PDA

View Full Version : Custom 3SGTE Intake Manifold - 5/16 Materials Arrived!



Pages : [1] 2

ChrisD
10-21-2009, 10:27 PM
I've been tossing around the idea of making a custom intake manifold for my ST165. Based on readings of the mr2 forums, it seems there is very little downside to this mod, especially so when you retain the stock runners, which is my plan. The lowest cost manifold on the market today is $500usd. I am confident a custom one would cost significantly less, all things included.

My initial thought is to make something similar to the ATS manifold: http://www.atsracing.net/ats%20intake%20manifold.htm

That is, add a large plenum atop of the factory runners.

Pre-requisites:

-Must keep my 3rd gen TB
-Must keep position for ST205 WTA IC
-Should support me from my current HP level (~265whp) to 350whp or slightly more on a bigger turbo

I would likely buy the plenum extrusion from RMR (http://www.rossmachineracing.com/plenumextrusions.html) - who sells the parts but no longer carries the manifolds. They have the end caps too. I would likely have to go to mcmaster carr for some sort of AL stock for the center feed TB plate.

Which leads me to my first question. Which plenum style is most beneficial and best suited to my requirements, and why?

Half-Round:
http://www.rossmachineracing.com/images/extrusions/Half%20Round/halfroundt.jpg http://www.rossmachineracing.com/images/large/halfroundwire.jpg

or

D-Shape:
http://www.rossmachineracing.com/images/intakeparts/dplenumt.jpg http://www.rossmachineracing.com/images/large/dplenumsize.jpg

Obviously, the D style is quite a bit larger. However, it is easier to mate a TB mounting plate to the half-round style. ATS seems to use a D-style that looks like it is quite large. I'd bet it's actually the same RMR extrusion that I'd be purchasing.

http://www.atsracing.net/atsintake%20manifold%20center%20feed.jpg

My next question would be in terms of mounting the plenum. Has anyone done this before?

Initial thought is to make a cut so that I am dealing with a straight line for welding. Pro - ease of welding, Con - may not be conducive to the best flow.

Anyways that should be enough to kick-start some discussion. Thoughts? What would you do if it were you? Lets see what comes of this...

beingblueeyes
10-21-2009, 10:59 PM
Well if it were me I would make a hole intake manifold, runners and plenum. Since I would be making a plenum I would use a INFINITI q45 (http://cgi.ebay.com/ebaymotors/INFINITI-Q45-90MM-THROTTLE-BODY-SR20DET-RB26DETT_W0QQcmdZViewItemQQhashZitem4837e25b62QQit emZ310175226722QQptZMotorsQ5fCarQ5fTruckQ5fPartsQ5 fAccessories) 90mm throttle body

and go to town on it

ChrisD
10-21-2009, 11:21 PM
I think doing runners and the whole manifold introduces way more effort and difficulty than I am willing to take on. :D

Currently there isn't much need for a larger TB since my gen 3 TB lines up perfectly with the 205 IC.

:)

Luni
10-21-2009, 11:29 PM
Chris, I want the damn adapter you got that allows you to use the gen3 TB.

Whered you get that from?

Shadowlife25
10-22-2009, 12:00 AM
Didn't Conrad make it?

Nice discussion Chris. :)

Klue
10-22-2009, 12:08 AM
I thought I would chime in having done a few manifolds myself.
first off, like 99% of things people do to their cars in the pursuit of performance, your likely wasting your time. Why you ask? Because sure an intake manifold is wonderful on certain vechicles that are producing enough power to warrant it, but really at your power level your betting of spending money where it counts, fuel and turbo.

500whp with stock manifold? YES, how? fuel and turbo. If you get a good turbo you are FORCING the air in, and to be honest the turbo doesnt give a shit weather you got WTA, 2 inch piping, stock or 1200 dollar manifold, 2.0 litres of displacement or 10L. as long as you can supply the fuel in the correct amount your going to make that power.

So it boils down to how much you want to spend, for how much your going to gain. with a 400 hp engine ATS gains just shy of 15 ftlbs... for 600 bucks? not a ice cube chance in hell that your convincing ME to buy that. and thats their gain with in house testing cough cough.

Now if you bored and just want to make something sure, i say go with the round plenum and stick some velocity stacks on there, and if you where me, then you and I would make it side feed and move the WTA off the engine. its pretty simple you need 2 90 degrees.

ChrisD
10-22-2009, 01:01 AM
Chris, I want the damn adapter you got that allows you to use the gen3 TB.

Whered you get that from?

Yep Conrad made it. It's damn slick. Will give you dibs if this turns out.

ChrisD
10-22-2009, 01:20 AM
Klue - appreciate your comments. Hopefully you can share some of your experience on fabbing your previous manifolds?

There's a lot of theory floating around about the validity of an aftermarket intake manifold - some saying not worth it, some saying there are gains to be had. Theoretical aside, there is proven empirical data (dyno's - eg. Hyde went from 258 to 302whp by adding an RMR manifold, although that car is a freak of nature) that +ive gains are available at my power level. My fuel system is fine for at least 330whp (injectors). Turbo, I agree, is a limiting factor (ct20b), however, I am quite confident that this will yield gains based upon the data I have seen.

I'd have to respectfully disagree with you saying it doesn't matter tho...removing restrictions in the intake path, imho, is almost always a positive. At the 500whp level I would place money down that you'd see a 40whp gain on an aftermarket intake manifold - and you'd have to *add* fuel just by making that change.

As for cost, I'm looking at probably $150 for parts total, plus any labour costs for welding. If I net 10whp, thats not a bad deal. Considering the alternatives of a turbo ($2300), cams ($600), or the ATS manifold ($500). With any luck I'll do a bit better. I'm not expecting crazy numbers at all.

Now - on with trying to make this thing.

ChrisD
10-22-2009, 01:22 AM
Also, IC is staying put. :D

Klue
10-22-2009, 03:21 AM
Good stuff, glad to share my opinions with you. Id like to keep this organized into our two discussions .I understand your apprehension to believe my remarks. First off the RMR manifold is completely different from what your making, that is a complete manifold with improved runners and KEY IS NO TVIS!!
stock system runners give a advantage to the 3rd cylinder because of the way the air is channeled into it. next up is the fact that they use velocity stacks, another huge gain in flow dynamic there. The manifold im running uses these points and is sidefeed, no tvis so i will give you some data once i finish tuning it. also using the same turbo for the winter months ct20b ceramic. But like I was saying the turbo is forcing in the air, so the gains from having a smooth path, restrictions etc are dulled to a great effect. Now if you had a 9k 4age then thats a different story when dealing with air restrictions and stuff on the intake path. with a engine relying on horsepower to get it moving then flow becomes a big issue but the air is "forcing" itself into a "smooth" pattern by the time it hits the valve bowl which in reality is all that matters.

Yes, i wont deny those guys get gains, and yes im sure you will have some. As far as your budget is concerned if you can weld that thing then it is Definately worth a few horsepower to increase the plenum volume HOWEVER, you will sacrifice what some call "spool" in the name of filling that plenum volume before positive pressure is achieved. Although you sound conviced you will net a better gain from moving the IC than making your intake :Ban1:
anyway back to topic:

If you keeping the stock runners then you dont really have that hard of a time, if you can encorporate some velocity stacks you will see gains, if not try to knife edge the runners as you take off the plenum. try and get as much volume as you can work with, and also watch the fitment against the firewall. Ummm before i go on lets clarify what motor you using and if your keeping the TVIS etc.

Grot
10-22-2009, 03:28 AM
he has gen 3 WRC 3sgte

joe's gt
10-22-2009, 03:33 AM
Hey Chris, if you tell me the runner diameters, how far apart the runners are from each other, and the length of the plenum I could probably rig up a quick flow analysis for you using COSMOSWorks for the two plenums.

I also agree with Klue that with a larger plenum without velocity stacks, you aren't reaching the full potential of modifying your manifold.

rizin
10-22-2009, 04:49 AM
I just started to get my toes wet reading about custom intakes. In my reading it says to use velocity stacks to reduce the loss of air entering the runners. So I would think about the possibility of using them.

ChrisD
10-22-2009, 05:09 AM
Cool. I don't mind good debate at all. :)

This is for my ST165 (the ST205 will stay mostly stock, so I can play with things like this on the 165). Gen II 3sgte, ct20b, power fc standalone, all the usual supporting mods and then some.

Got any pics of your manifold?

Not budging on the 205 IC. :) There is *no* room for it anywhere else. Plus with the massive heat exchanger I have, it stays plenty cool. Also have water injection for good measure. My intake temps are very good, so not too worried about that.

Adding velocity stacks to the stock runners might be a tough job as the ports are all in weird positions. Best job I could do would be smoothing them as best as is possible.

Keeping TVIS for the time being. *unless, maybe, I go with option 2 below*

Options are open right now. For fun, lets talk about 2 designs, pros and cons. 1st with stock runners, larger plenum, 2nd completely new design built ground up.

*If* I were to build my own, my first question is how do I curve/bend the runners? Most tubing is straight as supplied.

Also, I can not do all the welding. I can probably tack it together, but I will have to outsource the primary welding. (this may still be free, but I wont be doing it)

ChrisD
10-22-2009, 05:10 AM
Joe - I am going to see if I can pick up a manifold this weekend to take measurements. I will let you know what I can figure out.

Shadowlife25
10-22-2009, 05:14 AM
I also am in agreeance about the Velocity Stacks.

I do however have to respectfully disagree Klue.
TVIS is in no way a restriction at these power levels. It is not detrimental at all in my opinion.

Sang
10-22-2009, 06:02 AM
with a 400 hp engine ATS gains just shy of 15 ftlbs... for 600 bucks? not a ice cube chance in hell that your convincing ME to buy that. and thats their gain with in house testing cough cough.

Nice of you to leave out that it also made 24whp, and made more hp and tq at EVERY rpm point starting at about 3600rpms to redline. That kind of area under the curve has significant benefits even if it was "just" 15 ft/lbs.

http://www.atsracing.net/atsintakemanifolddyno.jpg

Notice there is nothing but gains here...same spool up until 3600-3700 and then it's the business end of the ats manifold to 8K.


KEY IS NO TVIS!!
TVIS is not a restriction. Here is some tvis testing: One setup is stock manifold and gutted TVIS, the other is a slightly larger plenum (similar to chrisD's proposal) with active control over tvis.

http://mysite.verizon.net/ressmw94/sitebuildercontent/sitebuilderpictures/Dyno%20gutted%20TVIS%20stock%20vs%20LPIM%20w_TVIS% 20SAE.JPG

No real gain in top end from a larger plenum, but the area under curve in the mid-range was pretty significant..


I would likely have to go to mcmaster carr for some sort of AL stock for the center feed TB plate.

I contacted Scott and Tom over at RMR a couple months back, they don't sell 3sgte TB plates on their own but they do carry universal TB plates. Be warned that they are 0.50" thick, so welding might be a problem if you aren't used to it. I also contacted Hux racing and he can sell you 3sgte TB plates, he quoted me $45 (?) or so, can't remember exactly. If you do call, talk to eric if you can, he's much more amiable than emily has been...

I'm going the same route as you. Except I plan to use a '88 3sge sidefeed manifold instead. Cut the plenum off and slapping on a half-round plenum from RMR. Use one of the threaded end caps from RMR and TB flange from HUX.

If you want my opinion:

-Stick with stock runners, chamfer the runners and knife-edge the divider.
-KEEP tvis and actively control it.

As far as mounting the plenum onto the runners, you have two options. You can have the base plate CNC machined to cut the ports for the runners, or you can mark it yourself on the baseplate and go to town with a die grinder.

Obviously if you want to make this worthwhile, going with option 1 would probably put you close to the cost of an OTS manifold from a vendor. Option 2 is what i'm doing. The plenum thickness from RMR is only .25" thick so you shouldn't have much problems with welding it on.

Option 2 shouldn't cost you more than $150 after shipping if you go with the halfround plenum, 2 solid end caps, and an AL tb plate from HUX.

Assuming you do all the labor youself and get the welding done for cheap...it should be a worth while project. Just time consuming which is why I can't even started collecting parts other than the 3sge manifold :(

ChrisD
10-22-2009, 01:33 PM
Is there a thread/link that I can see the manifold that was dyno'd above? hmm...

Sang
10-22-2009, 01:43 PM
http://mysite.verizon.net/ressmw94/sitebuildercontent/sitebuilderpictures/LPIM%20side.JPG
http://mysite.verizon.net/ressmw94/sitebuildercontent/sitebuilderpictures/gutted%20TVIS%20Stock%20vs%20LPIM%20w_TVIS.JPG
http://mr2oc.com/showthread.php?t=187886

ChrisD
10-22-2009, 06:23 PM
Cool. Not a super huge plenum, but an increase for sure.

Here's some mockup pics I found of the ATS manifold. Nice work smoothing & knife edging.

http://www.celicatech.com/gallery/files/4/atsmanifold1.jpg

http://www.celicatech.com/gallery/files/4/atsmanifold2.jpg

http://www.celicatech.com/gallery/files/4/atsmanifold3.jpg

http://www.celicatech.com/gallery/files/4/atsmanifold4.jpg

Klue
10-22-2009, 06:37 PM
HEY HEY HEY! those are some nice pics and EXACTLY what i was talking about.
Just to deflame my attackers, It could be my grammar and terrible mastery of the english language but I NEVER SAID TO DITCH THE TVIS, weather it be at Your power level Chris, or for any case. Tvis is definitely usefull in developing high intake velocity at lower engine rpm. This reasoning would be completely contrary to my "forcing " air in point.

Sang, Yes i did fail to mention the 24whp but i was talking about the 20 ftlbs which TO ME is more important. when you have a 400 hp engine and you get 25 hp and 20 ftlbs, then likely downsizing that to Chris's engine with a turbo that flows less then you would end up with about 1/2 the gain(not worth 600 bucks IMO) I failed to mention it because you can see that just by looking at the graph.

Like i Was suggesting with the "in house testing cough cough" the validity of their test is negligable along with your gained low end torque, they are comparing their TVIS equpped manifold to a NON TVIS manifold... go figure they gained mid range. You ALWAYS have to be scepticle about people who are selling a product. You can even go as far to change the brake loading on the dyno to get 1000whp more if you want. easy as a few keystrokes

Also like I was saying about the VELOCITY STACKS, im 99 percent sure this is where their gains are had, That is probably the best design you can do using the stock runners, Sang yours looks pretty good to, but the ATS one looks spectacular, probably done in CNC.

Anther point, having donemanifolds both exhaust and intake for 3sgtes in mr2s and celicas the firewalls are VERY differnt, I will get some pics later to show you my maniofold

good thread guys lets keep it going.

Berg
10-22-2009, 06:50 PM
Hey Chris,

Not sure if you are interested or if it will fit, but if you want a spare manifold to cut up I have a first gen 3S-GE manifold you can have. You can have the T-VIS plate as well if you want it. Like I said though, I'm not sure if it will fit on a 3S-GTE gen II. Let me know and I can pull it off my spare engine.

Cheers,
Berg

ChrisD
10-22-2009, 07:01 PM
Hey Berg! No, the 3sge won't work for me. the runners are all slanted to work best with a side feed setup. It will bolt on just fine tho.

ChrisD
10-22-2009, 07:12 PM
...

Yes - ATS plate is quite nice. From what I've found, it is CNC'd, then fitted and then smoothed by hand to ensure a perfect transition. They have definitely done a nice job.

For the firewall. I have several inches of space to play with on my car. The big clearance thing I will have to watch for is the strut brace by my first look at things. Would love to see your manifold!

-------------------------

On the flip side of things, lets also explore the full custom job.

Using RMR's oval tubing and velocity stack (this increases costs significantly), mated to the head flange from a stock manifold.

http://www.rossmachineracing.com/images/large/ovalrunnert1.jpg
http://www.rossmachineracing.com/images/large/ovalvstackt2.jpg

Again I ask - how do you properly bend this stuff? Mandrel benders aren't exactly common, would an exhaust shops pipe bender work?

Runner length would have to be close to the stock length due to the IC setup.

-------------------------

Pro's / Con's of each option? At my current power level? In the future on a larger turbo?

Any links to intake manifold build threads, with pics?

Berg
10-22-2009, 07:33 PM
Hey Berg! No, the 3sge won't work for me. the runners are all slanted to work best with a side feed setup. It will bolt on just fine tho.

Okay no problem.

(Rats, there goes my excuse to check out your 205)

Cheers,
Berg

Sang
10-22-2009, 07:36 PM
they are comparing their TVIS equpped manifold to a NON TVIS manifold...

Quite wrong.

I've yet to see an ATS built car that used tvis. ATS' school of modification always recommends removing tvis for a phenolic tvis spacer whenever possible. The plot was a back to back on the same car between an extrude honed stock manifold versus a modified manifold...tvis is not a variable.

...any other points you'd like to make to try and discredit the test?

Klue
10-22-2009, 08:37 PM
Quite wrong.

I've yet to see an ATS built car that used tvis. ATS' school of modification always recommends removing tvis for a phenolic tvis spacer whenever possible. The plot was a back to back on the same car between an extrude honed stock manifold versus a modified manifold...tvis is not a variable.

...any other points you'd like to make to try and discredit the test?


i think im picking up a slightly hostile undertone, for what I ask are you affiliated with ATS or something? Sorry i wasnt afforded the ignorance to believe what a company who sells a product is posting VS what my actual experiences has told me. In the dyno you show it clearly says GUTTED TVIS vs TVIS. Im just going on the documents your providing. none the less the points we are arguing is insiginificant to the thread, you win.

anyway. those velocity stacks and extrusions are really nice from rmr, problem is you cant really "bend" aluminum like that Im pretty sure it will just snap under the force.

I bought this plenum from someone else and slapped on my own runners, it was basically exactly the same as the one i had on my last engine.
here are some pics
http://i936.photobucket.com/albums/ad206/xiiicarparts/PICT2676.jpghttp://i936.photobucket.com/albums/ad206/xiiicarparts/PICT2675.jpg

ChrisD
10-22-2009, 09:03 PM
Ya I do think that the ATS dyno was both no-TVIS, there was a seperate one posted that compared TVIS vs non-TVIS....but whatever, lets just continue with the productive chat.

Bah you're right, AL will definitely just snap when you'd bend it. Doh.

So how does one properly fabricate runners?

I think what this will come down to is an unknown figure for performance boost, vs. price difference between going completely custom vs. modding the oem mani.

Conrad_Turbo
10-22-2009, 09:19 PM
Chris, I'd recommend velocity stacks that are not on the floor of the intake manifold (ie: not like the ATS manifold). For custom bending Aluminum (I've done the odd job here and there) I'd plug one end, fill it with dry sand, plug the other end with paper to pack down the sand. Clamp it, then heat the inside and outside radius of how you will be bending the tube and use some muscle. Or use a press and a wooden jig (cheap and doesn't damage the aluminum). Sand bending works for various applications, and is only really suited to one off stuff where tooling isn't justified. However it is all dependant on tubing thickness and bend radius (which you'd have to consider with hard tooling as well).

Sang
10-22-2009, 09:20 PM
i think im picking up a slightly hostile undertone, for what I ask are you affiliated with ATS or something?
I'm only going off what you say. You were the one inferring that basically their dyno's are crap.

And no, I am not affiliated with ATS, or any company for that matter.


In the dyno you show it clearly says GUTTED TVIS vs TVIS. Im just going on the documents your providing. none the less the points we are arguing is insiginificant to the thread, you win.

Perhaps stating your argument better would give you a leg to stand on?


Like i Was suggesting with the "in house testing cough cough" the validity of their test is negligable along with your gained low end torque, they are comparing their TVIS equpped manifold to a NON TVIS manifold... go figure they gained mid range. You ALWAYS have to be scepticle about people who are selling a product. You can even go as far to change the brake loading on the dyno to get 1000whp more if you want. easy as a few keystrokes

Notice I posted 2 plots. One from ATS, and one from an independent owner named Eric Baker that merely wanted to revert back to TVIS based on unsubstantiated claims of "tvis being crap" theories that made him gut it in the first place.

Berg
10-22-2009, 09:29 PM
Hey Chris,

If you are planning on doing your own runners, are there huge advantages to different runner lengthes with a forced induction application? I know that there are with naturally aspirated but wouldnt a pressurized intake negate that somewhat? And have you done any math to determine what length runners you would be doing? If so I'd love to see the equations.

Cheers,
Berg

ChrisD
10-22-2009, 09:29 PM
ok...lets move on...i like my threads drama free :P

ChrisD
10-22-2009, 09:43 PM
Conrad - thanks for the tip on the bending! That would probably suit my needs should I go that route.

I've seen velocity stacks placed above the floor before. Never understood why, although I'm sure there is good fluid dynamics behind it...can you explain?

Berg - No calculations yet. Still deciding which route I go. It still matters, sure. Generally, short and fat for better high range power, long and skinny better for mid-range. (huge over-generalization!) BTW - no need to make excuses to see the 205, heck, if you're free tomorrow night a few of us might be grabbing some beers somewhere. Interested?

Klue
10-22-2009, 10:38 PM
what do you think of the manifold chris?

Conrad_Turbo
10-22-2009, 10:42 PM
To make a long story short, a velocity stack on the floor can only pull air above the stack. While as a protruding velocity stack can draw air above and also pull air from below the velocity stack due to the air being drawn in. Effectively makes the velocity stack more efficient. I don't have any fancy CFD models to show you what I mean, so you're just going to have to believe me. :D Either way it is more work to build the manifold with protruding velocity stacks, but there are gains to be had with the extra effort.

ChrisD
10-22-2009, 10:53 PM
Makes perfect sense. Thanks. :)

klue - I like! Looks like you went with round runners - easy/tough to mate to the flange? Using velocity stacks or no? Able to share any thoughts on how it performs (what boost are you running)? Any idea what the volume of that plenum is? Length of runners? Obviously, trying to sort some specifics to give me a bit of a reference point. But the design looks solid, I like the rounded end caps. :)

Did some quick calcs, looks like the "half-round" plenum would be ~2.8L, while the "D-Shaped" would be 3.6L. (based on a 12inch length) Anyone know what stock is?

Sang
10-23-2009, 12:30 AM
Anyone know what stock is?

Roughly 1L

joe's gt
10-23-2009, 12:30 AM
It would be so awesome to do a CFD analysis in Floworks for various manifold designs. Never done it and it would be an interesting project. I'd probably need a lot of measurements and help with parameters along the way tho. If you want to do it Chris, I could probably model any manifold you come up with, and do CFD analysis on it, including heat transfer through the manifold. Idk, just a suggestion. I am by no means extremely knowledgeable, but I think it would be an awesome learning experience and help you make various decisions. I can't guarantee how fast my learning curve would be tho, since I've never attempted anything like this before. I have the software, just have never needed to design a manifold.

Klue
10-23-2009, 02:25 AM
Yes velocity stacks for sure, mounted on the floor of the plenum. Conrad has a good point. When i do N/A manifolds 100% i agree you need to lift the VS and also make sure its fully round, BUT once again your forcing feeding this bitch so i dont apply the same princepals. More concerned with plenum volume. I drove it today for the first time. Still tuning it so i cant comment on boost situation.

I can for sure say that throttle tip in response is excellent! and the induction sound makes me piddle LOL anyway, i choose the round runners to maximize flow and also because im using a fully circular VS. no problem mating to the flange just took some heat and light taps with the hammer to oval it. also needed to make a dimple for the fuel injectors. otherwise perfect.

here is the deepthroat
http://i936.photobucket.com/albums/ad206/xiiicarparts/PICT2656.jpg

and here is another sidefeed i have used using stock runners
http://i936.photobucket.com/albums/ad206/xiiicarparts/For%20Sale/PICT2650.jpg

and a 3sge manifold engine for good measuer
http://i936.photobucket.com/albums/ad206/xiiicarparts/Projects/PICT2578.jpg

rizin
10-23-2009, 04:47 AM
Ok I said I was getting my toes wet with custom intakes but. Love the T-Vis now how to make it better on a higher flowing engine is the next step. For me that is.

The only time a plenum volume matters is when you tune the runners then the plenum to them. So Chris if you plan on making a total intake the plenum can help. The runners can be tuned for where you want the best tourqe. A lot of math in the length calculation. A intake under boost and with out works the same on air flow. Put more mass in the intake it takes more force to move it. The only thing that I might not think is the same beacuse I am reading about n/a situations. Is the preesure wave that is what you tune the runners for. It helps the air flow into the cylinder. It travels at a certian rate. That is why length of the runners matter. But on a turboed engine I have not read if the wave travels faster at higher pressure I am betting it does. Since that is the same for sound.

So any thought on this guys I am new to everthing so correct me if I am wrong.

joe's gt
10-23-2009, 07:00 AM
Ok I said I was getting my toes wet with custom intakes but. Love the T-Vis now how to make it better on a higher flowing engine is the next step. For me that is.

The only time a plenum volume matters is when you tune the runners then the plenum to them. So Chris if you plan on making a total intake the plenum can help. The runners can be tuned for where you want the best tourqe. A lot of math in the length calculation. A intake under boost and with out works the same on air flow. Put more mass in the intake it takes more force to move it. The only thing that I might not think is the same beacuse I am reading about n/a situations. Is the preesure wave that is what you tune the runners for. It helps the air flow into the cylinder. It travels at a certian rate. That is why length of the runners matter. But on a turboed engine I have not read if the wave travels faster at higher pressure I am betting it does. Since that is the same for sound.

So any thought on this guys I am new to everthing so correct me if I am wrong.

Sorry dude, but that is not totally right. I applaud your effort to learn and understand the crazy theory and science behind intake manifolds. Hell, I am still learning as well, but this is what I think I understand so far.

The plenum serves 2 purposes. It must be designed in such a way to best allow air to flow equally thru all runners, and the large volume of air acts as a spring to bounce the pressure waves back and forth between the valves and the plenum. There is also harmonics in the intake, but that is a lot less important on an FI motor compared to N/A.

There for it is important to size the plenum so it is big enough so there is enough air for the engine at high rpms, but not so big that throttle response suffers due to the pressure having to equalize when the throttle plate is opened and closed. So the size is more based on the engine airflow demands than the runners.

As for the pressure waves, they actually travel about the same speed between N/A and turbo. Yes there is greater pressure in a turbo, so the speed should increase, right? Well the reason there is greater pressure is because you are forcing more air in (higher density = more mass). So when you try to move something with more mass it moves slower. I believe the way the calculations work out is that the excess mass caused by the turbo forcing more air in and increasing density, in effect cancels out the excess pressure, and the waves end up traveling at around the same speed (speed of sound).

With Chris' current situation, the stock 3s manifold plenum is too small for his engine flow demands based on all of the modifications he has made. So in essence, he is trying build a larger custom plenum to meet his upper rpm demands, while at the same time equalizing airflow to all his runners and retaining throttle response.

So in short, there is a little bit more leeway in plenum design then there is in tuning runners.

So now someone can read over what I said and correct me ;)

Berg
10-23-2009, 03:41 PM
I have another theoretical question, would there be any advantage in building a plenum for individual throttle bodies in this situation? I know that ITBs help with throttle response on NA applications I was just wondering if the same held true for turbocharged engines. I realize that this might be overcomplicating things, it's just a curiosity.

Chris - Beers tonight? Were you planning on going to a bar to watch the football game? I might be available, still got my number?

Cheers,
Berg

ChrisD
10-23-2009, 06:39 PM
Good pics, loving the deep throat action. :hehe:

Do you happen to know of a good source for inexpensive aluminum VS's?

Joe - good info. Maybe I should clarify my goals a little more now. Looking to maximize power above 4000-4500, at WOT to redline. Throttle response is somewhat less important to me. Also, would like the manifold to support me today and in the future. If I were in the turbo market today, I would select something that would get me in the 350-400whp range at under 20psi, with cams if required. No money today tho, so here we are trying to maximize what I have.

Berg - ITB's are not really a possibility for me personally due to the IC setup. I can't comment on effectiveness for an FI application, but I don't see a tonne of benefit being there under boost situations. Plus, with a standalone, I can tune around that a bit if I need to.

Klue
10-23-2009, 08:27 PM
I got these ones CNC onto the baseplate. Thats why the cost is usually so high. The manifolds usually get made first in some type of plastic, i forget the name the guy uses.

ITB's just add complexity when its really not necessary or without a real quantifiable advantage, IMO

Berg
10-23-2009, 08:34 PM
ITB's just add complexity when its really not necessary or without a real quantifiable advantage, IMO

Thats why I was asking if there was an advantage :wiggle:
According to the replies I'm guessing that with a turbo engine there isn't.

Back to the plenum designs!

Cheers,
Berg

joe's gt
10-24-2009, 02:11 AM
ahh, I see Chris. sorry for misleading anyone. ya, I don't have much info or knowledge on ITBs either.

ChrisD
10-25-2009, 04:13 PM
Hey no worries joe - a lot of people probably have different goals than me.

I'll toss in my dyno graph too. Remeber, this is a Dyno Dynamics, which has been tested against local Dynojets to be 14% lower. So that places me at 265ish whp on a dynojet. I'd still like to tune out that dip I have there - we had to be really conservative during the tuning as we saw some high knock readings...anyways the graph:

http://www.celicatech.com/gallery/files/4/dynograph.jpg

ChrisD
10-26-2009, 05:03 AM
Got the donor manifold. (stock 2nd gen)

So here's the dimensions of the runners at the head flange.

Small ports: 2.5 x 3.7 cm
Larger ports: 2.7 x 3.7 cm

With the plenum @ ~ 1L as claimed earlier in this thread.

The ports have rounded corners, so I'm not sure how tough it is to make calculations on that. If not, rectangular runners may do fine. Joe, able to work with that or need more info for the stock mani?

ChrisD
10-26-2009, 05:42 AM
http://www.not2fast.com/gasflow/Lecture08/Lecture08_files/v3_document.htm

I haven't read this yet, but it seems that a lot of the calculations we can use are in there...

Trance4c
10-26-2009, 01:12 PM
Chris, you will need to go forward with the cam tuning we spoke of in the past if you do this. At this level, your not maxing out the stock intake manifold. You will make it easier for the engine to breath, for sure. So might as well take advantage of it with a cam setup. It would maximize what you have for less than going to a bigger turbo.

ChrisD
10-26-2009, 04:08 PM
For sure. The plan is to install the manifold and cam gear, then go to the dyno. Check results with just the manifold change, then retard exhaust cam from to optimize results. (and of course adjust fuel maps as required)

I've already got the cam gear, just needs to be installed. :)

Bigger turbo...ahh ya, in the long term plan, but for now there's no budget for it. *...dreams...*

On a side note I had my ST165 out for a quick spin today immediately after driving the ST205. In comparison, my 165 seriously hauls. These cars fricken love the fall/winter air. :D

Trance4c
10-26-2009, 04:33 PM
No joke, its finally cooling down here in north FL, 60's in the early mornings and its just a crazy the difference from cool air. The air box helped quite noticably here in FL during the hot days too.

Sounds good though, I would be interested in you getting a 256 cam or a 3rd gen intake cam on the car once you have the manifold in. You'll be able to adjust the torque curve to your liking with the exhaust cam, but you should really get something with a longer duration intake to take advantage of all that air coming in from the larger manifold and TB.

An bigger turbo? who? why care? Nothing better on the street then a responsive turbo! With the intake and cam setup, you could net another 30hp, that should make for more than enough fun. Hell, on some race gas an a pound or two more of boost at the same timing, I wouldn't spend the money or time on something bigger. Driving it to the limit like that isn't an every day affair anyway.

ChrisD
10-26-2009, 05:02 PM
Remember I have the 3rd gen intake cam already. A small improvement over stock, but it is something.

haha, turbo is down the road anyways. Right now it is extremely fun to drive and I am enjoying it - but I always need some project to keep me busy...

Trance4c
10-26-2009, 05:09 PM
opps.. didn't realize you got a 3rd gen intake cam.... doh. Good deal, will be interesting to see how you like the intake manifold then! Good luck!

alistair
10-27-2009, 04:31 AM
Hey Chris,

If you want a project my 165 is always in need of some love ;)

ChrisD
10-29-2009, 08:59 PM
OK, so here's some measurements of the stock ST185 intake manifold:

Runner Length (along center line of runner): 25cm
Vertical distance from bottom of plenum to center of head flange: 17cm
Vertical distance from center of TB to center of head flange: 20.5cm
Horizontal distance from TB flange to head flange (middle, flange is angled): 2-3cm
Angle of head flange from TB flange plane: ~20 degrees
Arc radius of runner curve along runner center line: ~6cm

Dimensions of the runners at the head flange:
Small ports: 2.5 x 3.7 cm
Larger ports: 2.7 x 3.7 cm

Distance between runners at the head flange (# is the cylinder)
1-2: 2.8 cm
2-3: 4.6 cm
3-4: 2.8 cm
Between smaller/larger runners: 0.5 cm

Plenum width: 26.5cm
Plenum height: 7.5cm tapering to 5.5cm
Taper begins 18cm from RHS of plenum edge, viewing TB inlet straight on.
Center of TB: 11.5cm from LHS, 15cm from RHS
Depth of plenum: 7.5cm tapering to 6cm

TB diameter: 55mm (stock)
My TB : 60mm (3rd gen)

joe's gt
10-30-2009, 01:29 AM
Thanks Chris. I'll get started modeling it right away. The taper should be fun. I'll post a screen shot when I'm done. Modeling it is actually the easy part. Flow analysis is gonna take a little longer to figure out.

ChrisD
11-02-2009, 11:10 PM
Stock manifold pics:

http://www.celicatech.com/gallery/files/4/IMG_2202_small.jpg

http://www.celicatech.com/gallery/files/4/IMG_2205_small.jpg

joe's gt
11-03-2009, 06:39 AM
oh my, this is going to be fun. lol. This manifold is in no way easy to model as I expected. I'll see what I can do. This definitely helps a lot tho. Thanks Chris. I'll have to combine your measurements with the apparent angles of the the runners and head flange to come up with something as close as possible.

MrTurrari
11-06-2009, 03:25 AM
I'm just going to jump in here and make some comments about some of the assumptions that have been thrown around in this thread. :wedge:

Turbo manifolds work the same as NA manifolds. You use the same modeling and math to figure out both of them. Air is air. It doesn't suddenly start defying the laws of physics when you increase the pressure from 14.7psi (NA) to 29.4psi (1bar). The speed of sound even stays the same except when you increase the temperature then it gets a little faster. Air flows by volume and all the other forces, except the effects of heat, cancel out.

A plenum has only one purpose. To even out the flow between runners to keep them from starving each other. It's a buffer. The reflections happen whether you have a plenum or not even if you just join two pipes together as is done with a log style manifold. In this case though we don't want a low pressure forming like we get in a more restricted space like two joined pipes so we have a plenum. As a matter of fact the cross-talk pulses between runners get weaker when you have a plenum and the bigger it is, the less effect they have.

The size of the plenum is determined by the size of the engine, how efficient it is and where in the RPMs it makes power. Nothing more. Turbo or NA it doesn't matter. If you have a stock powerband, ie. stock cams and stock runners, you will NOT gain much from a larger plenum. Toyota already sized it right for you. A larger plenum will give you a couple more HP on top at the expense of power down low and spool time. You probably won't even notice the difference because it is so small.

Now if you add cams and larger intake runners then the cylinders will start to starve each other at high RPMs because the engine will flow more air and the intake events are closer together. In that case you need a larger plenum to combat the starvation effect. You will notice a much larger gain in power going from the stock size to a larger plenum in that case.

Now I'm sure some people will try to argue that NAs and Turbos are apples to oranges but realize this, turbos mask the losses you would see if you were running your engine NA because the boost controller varies the flow. It makes it seem as though what you did had no effect when in fact all you did was make the engine work harder to achieve the same effect. When you really want the most from your setup you need to consider all the parts though and how they work together. You stop thinking in peak HP and start thinking powerband.

joe's gt
11-06-2009, 04:18 AM
When you really want the most from your setup you need to consider all the parts though and how they work together. You stop thinking in peak HP and start thinking powerband.

Its funny you say that, cuz when I first got into cars I had that noob mentality of more horsepower = faster and funner. But now I can look at a dyno of some high horsepower car with no area under the curve and tell someone that car is not fun to drive on the street and then get looked at like I have no idea what I'm talking about.

Conrad_Turbo
11-06-2009, 03:46 PM
I'm just going to jump in here and make some comments about some of the assumptions that have been thrown around in this thread. :wedge:

Turbo manifolds work the same as NA manifolds. You use the same modeling and math to figure out both of them. Air is air. It doesn't suddenly start defying the laws of physics when you increase the pressure from 14.7psi (NA) to 29.4psi (1bar). The speed of sound even stays the same except when you increase the temperature then it gets a little faster. Air flows by volume and all the other forces, except the effects of heat, cancel out.

A plenum has only one purpose. To even out the flow between runners to keep them from starving each other. It's a buffer. The reflections happen whether you have a plenum or not even if you just join two pipes together as is done with a log style manifold. In this case though we don't want a low pressure forming like we get in a more restricted space like two joined pipes so we have a plenum. As a matter of fact the cross-talk pulses between runners get weaker when you have a plenum and the bigger it is, the less effect they have.

The size of the plenum is determined by the size of the engine, how efficient it is and where in the RPMs it makes power. Nothing more. Turbo or NA it doesn't matter. If you have a stock powerband, ie. stock cams and stock runners, you will NOT gain much from a larger plenum. Toyota already sized it right for you. A larger plenum will give you a couple more HP on top at the expense of power down low and spool time. You probably won't even notice the difference because it is so small.

Now if you add cams and larger intake runners then the cylinders will start to starve each other at high RPMs because the engine will flow more air and the intake events are closer together. In that case you need a larger plenum to combat the starvation effect. You will notice a much larger gain in power going from the stock size to a larger plenum in that case.

Now I'm sure some people will try to argue that NAs and Turbos are apples to oranges but realize this, turbos mask the losses you would see if you were running your engine NA because the boost controller varies the flow. It makes it seem as though what you did had no effect when in fact all you did was make the engine work harder to achieve the same effect. When you really want the most from your setup you need to consider all the parts though and how they work together. You stop thinking in peak HP and start thinking powerband.

You know exactly what you're talking about.

ChrisD
11-06-2009, 04:34 PM
The size of the plenum is determined by the size of the engine, how efficient it is and where in the RPMs it makes power. Nothing more. Turbo or NA it doesn't matter. If you have a stock powerband, ie. stock cams and stock runners, you will NOT gain much from a larger plenum. Toyota already sized it right for you. A larger plenum will give you a couple more HP on top at the expense of power down low and spool time. You probably won't even notice the difference because it is so small.

Now if you add cams and larger intake runners then the cylinders will start to starve each other at high RPMs because the engine will flow more air and the intake events are closer together. In that case you need a larger plenum to combat the starvation effect. You will notice a much larger gain in power going from the stock size to a larger plenum in that case.

Good stuff. Like I said, I'm no expert and I've never designed a manifold before. It may turn out that there is just too little to gain from this mod at present - we'll see.

For my specific set up, I have a gen 3 intake cam and also plan to retard the exhaust cam slightly as well.

The point about *where* in the RPM's I want to make power. We do not know where the stock manifold was designed to be most efficient, however I suspect that with a stock CT26 they were looking at the 2500-5000 RPM range. For me, I'd like to make the bulk of my power at 4500-7000 RPM. I am not looking to build a high revving engine, which basically rules out the huge manifolds like the RMR which seems to only show gains in the extremely high RPM's. Given this, does that make a case for a custom manifold? I suppose we'll have to turn to the math and flow modeling.

On the note of turbo selections, my initial thought is that the turbo selection would have a big impact on IM design. Especially in plenum size. Consider the three options. Stock CT26 which the manifold was designed with; CT20b which is my current turbo @~17psi; or the GT3071R which is the 'future' turbo that I'd like to go to at some point. With the change of only the turbo, would you not also want the specs of the IM to be altered? Like I said, I'm no expert so this is an honest question.

The last point is that the stock IM isn't designed that well...or at best, could be improved even using the same specs. So possibly there is room for improvement without a loss in the midrange. Remember that the ATS dyno comparison shows zero losses, all gains.

Good info tho. :)

MrTurrari
11-06-2009, 07:24 PM
Yeah I'm really no expert either but I did my homework when making my own intake manifold and came to the conclusion that most of the info out there is just generalizations and guesses. I had to come at it from a different angle and figure out just how and why air moves to figure it out.

Yeah you are right it does have an effect on the design. I should have added too that an engine with a larger turbo will also benefit some from a bigger plenum. They are less restrictive in the top end so your VE gets better up there. That combined with the slightly longer duration intake cam you have it may be worth it for you. I imagine that most of the problems people have with the 3sgte manifold comes from operating it far outside of it's intended purpose. Because of it's center feed design, it would tend to unevenly starve the runners when you try to move a lot more air through it then it was designed for.

Are you also planning to shorten the runners? To figure out where the stock manifold helps you, you need the cross-sectional area of each runner and how long they are from the plenum to the valve. Also you need to know the average temperature of the air going into it when under full boost. And lastly you need engine displacement and the specs of the cams, specifically the intake valve closing angle.

When you are thinking about this and trying to figure out what benefit a certain mod will give you, understand that engines have their biggest torque peaks determined by the cams. As long as boost holds steady you will always see two peaks when each of the cams comes in phase. All other mods you do will just add or subtract from these peaks or move torque up and down above or below each of them. The intake cam comes on first so by having slightly more duration by using a gen3 cam the peaks will get closer together. You will end up with a little bit narrower powerband but by shortening the runners some and increasing the plenum size you should be able to keep it from falling off so quickly. A big turbo helps with that too.

ToyMaster
11-06-2009, 10:22 PM
Thought I'd put my .02 in on this topic....

Firstly about TVIS I dont care what anyone's dyno says if you have never personally experienced a stock or even slightly modified (still uses a factory turbo) 3SGTE equipped car without the TVIS then you would know that the TVIS is NOT a restriction it IS an AID.
Toyota added them specifically for that reason because long divided runners were used in the design of the IM. That particular IM design was used because the TMIC necessitated the need for it - remember the 3SGTE was first introduced in the Celica NOT the MR2. Also the 3SGTE was based on the 3SGE which again was first introduced in the Celica NOT the MR2.
In theory if you want to eliminate the TVIS on a stock or slightly modified 3SGTE then the proper way to do it would be to shorten the runners by at least half, eliminate the dividers making the runners one and place the TB at the side of the plenum. We actually tried this once with the 3SGE IM almost 10 years ago. We cut the plenum off at the start of the runner and then shortened them by cutting them down a little more than halfway and grinding the divider out of what was remaining. But we never completed it because we could not figure out how to weld the plenum back on - didnt match up to the shortened runners. Now that I think about it I have an idea on how to fix it good thing I convinced my bro not to throw it out.

My brother and I have done almost every bolt on mod in almost every configuration you can think of and where ever a stock or upgraded turbo was used the TVIS was always retained in the end. To give some examples of the various setups we have tried:

-stock CT26 - 3SGTE & 3SGE IM w/ & w/o TVIS
-46 trim CT26 - 3SGTE & 3SGE IM w/ & w/o TVIS
-50 trim CT26A - 3SGTE & 3SGE IM w/ & w/o TVIS
-ATS CT27 - 3SGTE & 3SGE IM w/ & w/o TVIS
-CT20B - 3SGTE & 3SGE IM w/ & w/o TVIS

All those configurations were tested on a gen2 3SGTE with gen2 3SGE cams, 3angle, port matching, FMIC, 3" DP and exhaust, 550cc, TFFR, RRFPR, TTFP, ATS ROM tune. I know what your thinking that is a lot of swapping in and out an IM that is already a PITA to remove in the first place. We anticipated this so in order for us to make the job easier we fabricated two IC cold pipes - one for the center feed and one for side feed IM. We eliminated all the IM and TB braces by never re-installing them, eliminating the cold start injector and most importantly we secured the engine wiring harness to the bulkhead instead of on the engine itself. Makes removal of the engine or just the cylinder head a lot easier without having to remove the harness along with it plus it looks tidier.

Chris after looking at your current mods my suggestion to you would be:

A)Keeping with the TMIC go with the gen3 IM and have an adapter plate made up so it will bolt to the gen2 head - there was actually someone selling one on AUS ebay a couple weeks back with the IM.

B)Ditch the TMIC go FMIC and use a 3SGE IM retaining the TVIS

Taking into account your future mods might include bigger cams and turbo then I suggest you go full custom side feed IM not retaining the stock runners, loose the TVIS, loose the TMIC and go FMIC.

Personally I think you should loose the TMIC and center feed IM all together, no matter how nice of a plenum you use or port the stock runners the center feed IM will still lean out cylinders #3 and 4. Invest your money into a better IC and eliminate the weaker links in your setup, put the WI on the ST205 instead, it needs it more then the ST165 does especially if you DD it and expect the the ring lands to last in the gen3 motor.

ChrisD
11-10-2009, 11:05 PM
Well, keeping the TMIC. The 205 unit has shown *zero* signs of reaching its limits. Even after a few hours on the dyno, the IC was cool to the touch. Flow-wise I still do not see it being a restriction.

Anyways, for the purposes of this thread, lets assume that the intercooler is not changing.

Gen 3 IM eh...hmm. Well, I suppose we'd have to model that - but it requires a bit more work to fab an adaptor plate. Plus I lose TVIS. Granted, I don't miss it on my ST205. It has a great power band.

What if I throw into the mix larger cams? Lets say I could be convinced to adding some 264's at the same time. Change theories? Does it make a larger plenum more viable? Full custom job a better choice?

I would think we would be looking at an increase of VE and as such have less potential for torque drop off in high RPM's - similar in that respect to a larger turbo. So, more air going in might justify trying to make an IM that will flow somewhat better?

ChrisD
11-10-2009, 11:07 PM
Runners have physical space to be shortened by ~ 5cm, similar to the ATS design.

MrTurrari
11-20-2009, 12:40 AM
What if I throw into the mix larger cams? Lets say I could be convinced to adding some 264's at the same time. Change theories? Does it make a larger plenum more viable? Full custom job a better choice?

I would think we would be looking at an increase of VE and as such have less potential for torque drop off in high RPM's - similar in that respect to a larger turbo. So, more air going in might justify trying to make an IM that will flow somewhat better?
Yes cams would make a larger plenum more desirable. When it comes to intake manifolds, the diameter of the runners and the size of the plenum are based on the volume of air that must move through it. Making your engine more efficient at a higher RPM by adding cams increases the volume of air the engine uses at that higher RPM. A bigger turbo does the same thing by decreasing restriction in the exhaust and allowing for better cylinder filling at a higher RPM. Also when people talk about these mods increasing VE, me included, we are actually talking about moving the peak VE of the engine up in the RPMS and not actually increasing the peak VE value.

A custom manifold will be better then one based off the stock one simply because you will be able to increase the diameter of the runners too and match it to the ports. If you are doing this all out you would also install bigger valves and port the head to match.

ChrisD
12-09-2009, 09:46 PM
OK, lets for a moment assume I'm open to all options. Even going side feed and FMIC.

How do I go about calculating optimal runner size and length, and plenum size?

joe's gt
12-09-2009, 10:52 PM
OK, lets for a moment assume I'm open to all options. Even going side feed and FMIC.

How do I go about calculating optimal runner size and length, and plenum size?

There are entire textbooks devoted to this one question. lol

ChrisD
12-10-2009, 05:51 PM
cliffs? :laugh:

joe's gt
12-11-2009, 05:15 AM
eh, I just have regurgitated info that I'd have to sort through to determine what legitimate. Calling Turrari, WOT, Conrad, or Nuke...? They'd be the best ones to give u cliffs. Here's a link MrTurrari posted in a thread awhile ago.

http://www.grapeaperacing.com/tech/inductionsystems.pdf

MrWOT
12-12-2009, 07:41 PM
Cliffs?

I lol'd. :hehe:

Keep the TVIS, more separation between runner inlets per cyl, 12-14" runners, 2.25-2.5L plenum. Plenum reshape for larger TB and tapered feed from TB to plenum.

FourVeeSix
12-17-2009, 12:08 PM
Try this site for a wealth of calculators etc, may come in handy. :)

http://www.wallaceracing.com/Calculators.htm

ChrisD
12-17-2009, 05:05 PM
Interesting.

joe's gt
12-18-2009, 02:16 AM
Alexander Bell is awesome. I was gonna make a spreadsheet for myself, but now I don't have to. lol. Thanks for the link.

ChrisD
12-21-2009, 03:51 PM
Just updating with some pics of some MR2 custom manifolds.

These are the RMR velocity stacks. Is that enough protrusion out of the plenum floor?

http://i213.photobucket.com/albums/cc148/racerx1715/IMG_0783.jpg

http://i213.photobucket.com/albums/cc148/racerx1715/IMG_0798.jpg

http://i213.photobucket.com/albums/cc148/racerx1715/IMG_0811.jpg

http://i213.photobucket.com/albums/cc148/racerx1715/IMG_0814.jpg

Here is a really interesting way to utilize the stock head flange to save on costs. I have found one custom flange available for ~$120usd however.

http://i465.photobucket.com/albums/rr11/mr2ninja/DSC00758.jpg

http://i465.photobucket.com/albums/rr11/mr2ninja/DSC00751.jpg


Yeah, I used a stock manifold to start and did a BUNCH of welding and grinding.

At first I was just trying one to see if it would work, it turned out better than I thought it would, but I will just have a flange made next time.

Reference Link: http://mr2oc.com/showthread.php?t=373624

joe's gt
12-22-2009, 08:33 AM
Hey Chris, so sorry I have kind of dropped the ball on this little project of yours. I got extremely busy with school, finals, my senior project, and building a website and I just didn't have the time. I finally found some time to work on it. So here is the progress.
http://i23.photobucket.com/albums/b396/92ct-27gt/allrunnersfrontview.jpg

Now on the next picture, you can see the runners aren't perfect. Doing this 3D space curvature was SOOO difficult. The reason they aren't perfect is because based on the ATS pic of the cutoff runners I saw, the runners that meet the plenum are a slightly different size than the runners that meet the head port. I do not know if this was caused by ATS cutting horizontally across the angled runners, thus making them look bigger, or whether they actually are bigger and taper to smaller at the port. Therefore, I made the runners different sized at the top (based on visually estimated measurements) and the bottom which results in the thinning around the curvature that you can see in the picture below. I tried my best to eliminate the thinning by drawing a bunch of profile sketches of the runners along the curve, but it still is slightly thinning. To solve this problem, I could just sweep the profile at the head port all the way to the top, which would prevent the thinning and cause the runners be the same size throughout the whole curvature. Its up to you. I'll post a pic soon of the runners swept without the change in shape and therefore no thinning and you can tell me which one you want to go with.
http://i23.photobucket.com/albums/b396/92ct-27gt/allrunnersthinningview.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/sideview.jpg

Now, correct me if I am wrong, but we are mainly trying to determine how a plenum distributes flow to the runners. The thinning runners, while inaccurate for analyzing flow through the runners, should be adequate for just analyzing how a plenum distributes the flow to them. This, however, will all be irrelevant if you decide to go with the constant head port runner shape.

Also, if its ok, I'd rather start modeling your plenum designs than trying to model the stock plenum. The stock plenum has a bunch of angles and meets the runners in a V-shape and it will be a real pain to model. We already no that its too small to meet your needs and we already know that the flow distributed to the runners is horrible, so I really don't see the need. But if you want me too, I can model the stock plenum too, it will just take a little longer.

It sux not having a manifold here. lol

Shadowlife25
12-22-2009, 11:14 AM
Damn nice job there Joe. I am impressed. :)

Conrad_Turbo
12-22-2009, 04:52 PM
These are the RMR velocity stacks. Is that enough protrusion out of the plenum floor?

Looks as useful as no protrusion. Without doing CFD on it, you should be looking at ≈0.25" protrusion to actually be effective. This is all dependant on the manifold design as well so don't take that # as a defined goal. You need some space under the trumpet to allow the air to be pulled from underneath and into the intake runner.

Joe, nice job!

ChrisD
12-22-2009, 06:46 PM
Joe, awesome stuff. No worries on the time...I'm in no rush on this project.

Yeah, hard to say if the runners on the stock manifold are tapered or not. It is definitely possible although I cannot say without cutting up a manifold. Go with the design you've got for now. Analyzing flow through the runners may be required after if I consider building custom runners...then hopefully maybe we can try to compare some options.

Let's try two things to start. I want to compare the potential benefits of going from a center feed to a side feed plenum. Yes, this would require that I change my intercooler...let's say I'm considering going FMIC. Maybe.

Using the D shaped plenum and/or the half-round one for comparison's sake.

One other advantage of a side feed is that I may potentially be able to shorten the runners a bit more.

:) :) :) :)

Conrad - I figured as much. It's too bad, b/c those are really easy to fit up and weld in. All that is required is drilling a hole in the plenum floor. I suppose, really, the same can be said for a traditional VS. Just have to find a good, cost effective source for some.

joe's gt
12-22-2009, 09:18 PM
So the single profile sweep was worse than the previous one so it looks like the above pictures are as good as it is gonna get. If I had time, i would draw a bunch more sketch profiles but that would take FOREVER!

So Chris, do you need me to model the stock plenum, or can we start right out for your centerfeed and side feed plenum designs. IMO, the goal here is to just analyze flow distribution to the runners, so we can start right out with your designs.

Also, if you decide to make an entire custom manifold, modeling that will be much easier because you won't be using any ridiculous 3d space curves like the stock manifold and an accurate simulation of flow through the runners and the plenum can be obtained. I don't think the runners as they are now would yield absolutely horrible, unusable results either.

Modeling something as shown in the pictures of the manifold you posted above would be pretty easy.

ChrisD
12-23-2009, 07:55 PM
Lets just go ahead with some custom plenums.

joe's gt
12-23-2009, 08:40 PM
Oh I was hoping you'd say that. Looking at that stock plenum made me cringe after doing the runners. Let me know what you come up with, or are you still using one's from the first page? And if so, what volume of the one's on the first page were you shooting for? Taper?

Oh hell, just draw me a diagram when you come up with one. lol

ChrisD
01-27-2010, 04:22 PM
Want to get back on this a little bit.

Lets say I want to go side feed and use the stock runners. No velocity stacks, but at least do my best to replicate what ATS has done with theirs. Primary goal is to move to a side feed to facilitate optimal FMIC routing. IM performance, lets say, is secondary to that goal as I suspect real world gains may not be huge. But possibly, there are some on the table so I might as well grab them.

Power at present is based on the ct20b turbo with a view to a larger say GT3071/76/82 size. Current power 265-280whp @6200rpm, future goal 350whp. Peak torque currently measures at 5100 RPM but I suspect with some tuning I might get it to 4400 RPM. Want to look for best power at 4500-7500 RPM.

The question

Which plenum is best for me: D-shape (3.30L), or Half-Round (2.57L)? (RMR's from 1st page, adjusted for 11" length instead of 12" like first est.)

ChrisD
01-27-2010, 10:08 PM
Also want to do some flow analysis if we can on a tapered design, if is is worthwhile.

Joe or others - able to do something like this? (or teach me how?)

http://i123.photobucket.com/albums/o309/magnum126/manifold2.jpg

http://i123.photobucket.com/albums/o309/magnum126/manifold1.jpg

http://i123.photobucket.com/albums/o309/magnum126/manifold4.jpg[/QUOTE]

joe's gt
01-28-2010, 06:24 AM
Yup we can do that. I already have the profile of the plenums from the first page, so if you could just tell me how much of a taper you want? Ending height, or an angle for the taper would work.

Also, will you be cutting the runners at the point just above the wiring harness mounts as ATS did or if you could give me a vertical measurement of how far down you plan on cutting the runners? Either way, if you can give me a vertical measurement of how far down you will be cutting off the runners that would be great.

The above pics are just a flow analysis with things called "cut" planes down the middle to show the flow velocities in those planes. The program I am using (SolidWorks w/ CosmosWorks) also has this feature.

This is my first attempt at doing this, so it might take a few attempts to get meaningful results.

ChrisD
01-28-2010, 03:45 PM
Sweet. Let's start by making the cut where ATS has. I like the idea around shortening the runners somewhat for higher RPM power, but it also gives more space to clear the strut brace. And potentially on JDM cars, the brake booster.

So there's basically 3 different variations I'd like to check out. x2 for both plenum styles...

To help with the visualization for everyone, I've MS Painted them. :D

1. No taper. Basic design with just the plenum, side fed.

http://www.celicatech.com/gallery/files/4/manifold1.jpg

2. Taper with varying angles/lengths to optimize flow & distribution. Really, this is a shot in the dark. If you can, pick 3 different combination's, and we'll compare all 3 to try to see which is the best, and then start adjusting from that point. We'll have no idea which is best until we do the flow modeling.

http://www.celicatech.com/gallery/files/4/manifold2.jpg

3. Full length taper. eg:

http://i465.photobucket.com/albums/rr11/mr2ninja/DSC00758.jpg

joe's gt
01-31-2010, 02:14 AM
Plenums...


Half-round
http://i23.photobucket.com/albums/b396/92ct-27gt/Halfround.jpg
Half-round w/ custom taper
http://i23.photobucket.com/albums/b396/92ct-27gt/customtaperhalfround.jpg
Half-round full taper
http://i23.photobucket.com/albums/b396/92ct-27gt/Fulltaperhalfround.jpg
D-plenum
http://i23.photobucket.com/albums/b396/92ct-27gt/Dplenum.jpg

Me and Chris are still deciding on how to taper the d plenum because as you cut the d plenum, the base width will very. Any suggestions are welcome. There is also the "just don't do it" option. lol.

Chris, all of these parameters are adjustable. All I really need to modify the taper are the following things:

1. Desired height at the end of the plenum

AND EITHER

2. How far left of the right end of the plenum want the taper to start

OR

3. Angle of taper

MrWOT
01-31-2010, 02:29 AM
I think all the ones you have up there are too limited by the throttle body placement. So now you are forced to have a plenum shape that will turn the air, rather than a manifold that has inherently good distribution. If you did model those above, you would see the cylinder closest to the throttle body running quite rich and the furthest cylinder quite lean under hard boost. Dense air has greater inertia, it will be more expensive to turn it. You want the inlet at an angle. :)

joe's gt
01-31-2010, 02:34 AM
I think all the ones you have up there are too limited by the throttle body placement. So now you are forced to have a plenum shape that will turn the air, rather than a manifold that has inherently good distribution. If you did model those above, you would see the cylinder closest to the throttle body running quite rich and the furthest cylinder quite lean under hard boost. Dense air has greater inertia, it will be more expensive to turn it. You want the inlet at an angle. :)

We can do this too. All I would need would be a desired angle to have the throttle body. When I did a very, very elementary flow analysis on a random side-feed manifold I built for fun a while back I noticed this exact thing you were talking about. Runners 1 and 2 got much more air than cylinders 3 and 4 with a horizontal entry.

Not only that, but Chris the stock runners are also designed for a center feed design. I don't know of any side-feed plenum design that would work with the stock runners. Hell the air would have to come in and literally turn back around to reach cylinders 3 and 4.

DudeMan
01-31-2010, 04:49 PM
I don't know of any side-feed plenum design that would work with the stock runners.

Not sure if this helps but here's two pictures to compare the stock intake manifold and a 3SGE side feed manifold. Maybe the runners should curve towards the throttle body?

http://i283.photobucket.com/albums/kk284/dudemanxl420/PC310431-1.jpghttp://i283.photobucket.com/albums/kk284/dudemanxl420/P7280047-1.jpg

ChrisD
01-31-2010, 06:19 PM
Well, we can play around and see.

re: angled TB placement

Can you clarify this for me? Do you mean having the TB angled up? (ie. pointing toward the hood?) I can see that this would soften the angle with which the air would have to turn. But it is difficult to build proper IC piping. Also remember the stock center feed plenum imho has to make some crazy turns, and has very poor cylinder distribution.

But, what we're looking to do is actually model it and adjust to come up with the best design in the end. But at least with this kind of proper modelling, we can see actual results instead of just speculation.

The 3sge plenum is good if you don`t modify the plenum, but I`ll be cutting down the runners a bit so I`m not sure there will be as much value in using it.

Joe - I`ll get that measurement for you today sometime.

ChrisD
01-31-2010, 06:20 PM
Joe - also the TB size that I am using is 60mm in case you need that. (gen 3 TB)

joe's gt
01-31-2010, 07:35 PM
Joe - also the TB size that I am using is 60mm in case you need that. (gen 3 TB)

Told me that in a previous post so we're good.


Not sure if this helps but here's two pictures to compare the stock intake manifold and a 3SGE side feed manifold. Maybe the runners should curve towards the throttle body?

Pretty much exactly what I was thinking. But it would require either 3sge runners or custom runners. As it is now, air would actually have to turn around back towards the throttle body to reach cylinders 3 and 4.

MrWOT
02-01-2010, 12:48 AM
TB up ~40* towards the roof, pointed at the inlet to runner #2 (assuming center feed runners)

Conrad_Turbo
02-01-2010, 02:56 PM
Chris based on previous CFD intake manifold work I've done, #1 is not ideal for even flow (easiest to build), #2 is better and #3 is best for even flow. It's almost common sense when looking at the results.

ChrisD
02-01-2010, 04:40 PM
OK, so add one additional - the full length taper with the TB tilted 40 degrees up. We'll see what the results show. I'll have to angle cut the plenum to do it.

I will cut the runners 3 cm from the bottom of the plenum.

I still would like to do one analysis of the D-shaped plenum. What if, instead of having a varying width base, I cut the top of the plenum and weld on a flat peice of AL. This way the base is equal in width and I just have to trace out and cut the top piece to fit.

joe's gt
02-01-2010, 09:25 PM
OK, so add one additional - the full length taper with the TB tilted 40 degrees up. We'll see what the results show. I'll have to angle cut the plenum to do it.

I will cut the runners 3 cm from the bottom of the plenum.

I still would like to do one analysis of the D-shaped plenum. What if, instead of having a varying width base, I cut the top of the plenum and weld on a flat peice of AL. This way the base is equal in width and I just have to trace out and cut the top piece to fit.

I could do both of those.


Chris based on previous CFD intake manifold work I've done, #1 is not ideal for even flow (easiest to build), #2 is better and #3 is best for even flow. It's almost common sense when looking at the results.

Feel free to tell me anything I'm doing wrong Conrad, cuz you've got way more experience than I do and I can guarantee I'm going to screw up something the first time around cuz this is the first time I'm doing this.


Chris based on previous CFD intake manifold work I've done, #1 is not ideal for even flow (easiest to build), #2 is better and #3 is best for even flow. It's almost common sense when looking at the results.

Hey Conrad, how would he even fit the piping in the engine bay if it was angled at runner #2 or #3?

ChrisD
02-05-2010, 09:05 PM
Need anything from me? I'd like to finalize the design within the month if we can, then move to purchasing materials, fabricating, welding, testing and installing. If things go really well maybe I'll be able to do before/after dynos for results.

Obviously, I also have to purchase FMIC supplies.

joe's gt
02-06-2010, 06:42 AM
Sorry man, I've been very bogged down with school, senior project, and job interviews the past two weeks. I'll do the best I can at getting those modeled.

ChrisD
02-06-2010, 06:46 PM
No prob, school first! I am also busy with this parts car, still loads of work to do on that before I get my garage back. Not to mention trying to keep up with the parts requests...I can barely keep up!

joe's gt
02-08-2010, 12:08 AM
So I tried modeling it with Floworks. It went anything but smoothly as I expected. For some reason the program does not like the holes I cut for the runners in the plenum. It won't create a fluid volume. It works for some random circular runners, but the square runner holes with the filleted corners it doesn't seem to like. I have spent 4 hours troubleshooting it, so I am taking a break for the rest of the day and trying it again tomorrow.

joe's gt
02-08-2010, 07:51 PM
So I've run into a huge roadblock that I don't really no any way of getting around. When I cut off the runners, I have to project the resulting outline of the runners onto the plenum so I can use that to cut the runner holes in the plenum. For some reason, Floworks doesn't like this and refuses to recognize the holes as air outlets and therefore won't do a flow analysis. However, I think I may be able to redo the runners (which I spent countless hours trying to perfect) by redoing the lofted feature and physically moving the end of the sketch used to create it 3 cm below the top. We will see how this works I guess.

EDIT: re-did the runners. Took about 4 hours. Hopefully it will work now, I am probably done working on this till Friday. Got a tone of stuff coming up in the next 3 days with classes, machining stuff for my senior project, 2 midterms and a job interview on the same day.

joe's gt
02-21-2010, 08:03 PM
UpDATE: After I re-did the runners, I made the cuts in the plenum and the flow analysis WORKED with the new cuts in the plenum. However, the flow analysis only seems to work with the plenum right now. FloWorks doesn't like what I did with the runners and won't do a flow analysis through them so I have to figure out what the problem is there. But the first half of the analysis works. Once I figure out the runners, we'll be good to go.

EdIT: Major problem. FloWorks refuses to recognize the runners as a fluid volume. I am assuming the program just can't do analysis on lofted features and cuts. For those that aren't familiar with 3D Modeling, a lofted feature involves a bunch of sketches of the outline of what u want to model and then a guide curve(s). The program then runs the outline of the drawing along the guide curve(s) to create a solid 3D object. Same with any cuts you want to make. Outlined of cut swept along guide curve(s).

So Chris, I honestly don't know what to do from here. There is no other way I can model the runners short of sharp angles that won't be smooth radii of curvature, but could possibly be useful just for determining flow distribution in the plenum. The only possible way to model the 3s runners is sweeping them along a curve, and if FloWorks won't do an analysis on that, there is nothing I can do. I got the plenum to work, but the runners are giving me fits.

Calling Conrad...I need your help. so frustrated right now.

ChrisD
02-24-2010, 06:04 AM
Well, I think I'm set on using the 3S runners. I like the idea of keeping TVIS, the car is still responding extremely well with it. So lets not worry too much about the runners.

At this point the big thing is trying to figure out the best plenum design. So lets do that. I'd be extremely happy if we had a solid design proven to be as optimal as is possible.

I am getting ready to order my FMIC and manifold supplies, so I'll need to figure what I need for plenum stuff. Also, I've got another treat for my 165 - a set of HKS 256 cams. :D Going to be a fun driving season!

Joe - can't thank you enough for the time you've spent on this. :)

joe's gt
02-24-2010, 06:48 AM
I'm trying man. I sent a message to Conrad. I might even try posting the question on a 3D CAD forum. I'm just so disappointed. Good news is that the plenum analysis works.

ChrisD
02-24-2010, 04:52 PM
No prob. At least with this, we are light years ahead of any of the other manifolds that have been built for the 3S. Can't wait to get going on it.

:)

ChrisD
03-10-2010, 10:44 PM
Do you have the plenum analysis? Or still have to set it up? I'm pretty close to ready to order parts.

Shadowlife25
03-10-2010, 10:50 PM
I am still following this. :)

Really looking forward to seeing the results.

ChrisD
03-11-2010, 12:54 AM
Me too! Right now I'm thinking it will be the half round full taper design, but we'll see. :)

joe's gt
03-11-2010, 02:05 AM
I have it set up, but this quarter has been an absolute nightmare as far as school work goes. Finals end after next week so if you can hold off until then, I have plenty of time to work on it. Not to mention, next quarter I only have 2 classes, so time won't be a problem either. I can get it done by the end of next weekend if that's ok.

What boost pressure do you want me to do the analysis at? Also, does anyone have an idea of approximately what the pressure at the outlet (where it connects to the head) should be?

joe's gt
03-11-2010, 05:17 AM
Alright, here's a taste of it. This is the D plenum with horizontal entry. Cylinders 3 and 4 don't get enough air as expected. It just sux that the 3d space runners I modeled don't work, but I guess it is pointless considering we are just seeing how the plenum distributes air to the runners. I used a boost pressure of 15 psi and an outlet pressure of atmospheric. Idk what pressure I should use for the runner outlets, so if someone could enlighten me on what's the best value for that, it would be great. I could also specify a mass flow rate if that is preferred.

http://i23.photobucket.com/albums/b396/92ct-27gt/D-horizontal-velocity.jpg

http://i23.photobucket.com/albums/b396/92ct-27gt/d-horizontal-trajectories.jpg

Shadowlife25
03-11-2010, 08:01 AM
What about a 10deg downward angle on the tb inlet?

joe's gt
03-11-2010, 08:50 AM
What about a 10deg downward angle on the tb inlet?

yup, we'll do whatever angle anybody wants. I kinda did this up quickly, cuz I know Chris has been kind of waitin on it and I feel bad for offering and not being able to keep up with it. But this is kind of what all of them are gonna look like.

I could also do a time dependent animation type analysis Chris if you give me your intake and exhaust valve durations and at what crank angle they open and close.

ChrisD
03-11-2010, 05:33 PM
Cool! Yeah, I can wait till after next week no prob. Then we can really play around.

The above already shows you what an ATS manifold with side feed setup would look like.

For my setup, I'm around 17psi right now. For CFM, there's not a lot of very good data on this however, it makes more sense to me to specify the pressure and see the resulting CFM on the model? (I have heard of total estimates pegging CFM at ~125 from intake all the way through the head...but who knows how close to my setup that is) You can try both to see if there's much of a difference if you like.

Here's the specs on my new cams:

HKS 256 Intake
Gross lift = 9.0mm
Net lift = 8.8mm
Lash = 0.2mm
Open = 0
Close = 40
Centre = 110

HKS 256 Exhaust
Gross lift = 9.0mm
Net lift = 8.75mm
Lash 0.25mm
Open = 27
Close = 11
Centre = 98

Conrad_Turbo
03-19-2010, 06:32 PM
Whoa I've been gone for a long while. Seems things have been worked out quite well. CFD hiccups happen often with Cosmos, it's a PITA to figure out sometimes and it's usually different issues that cause the CFD not to work, all dependant on how the model was created.

I haven't done a lot of CFD on intake manifolds in recent months but from what I have done a tapered manifold is more efficient in terms of flow to the last runner than a constant diameter/size runner. Like this:

http://www.aceengineering.ca/Temp/cfd.jpg

Much better having it taper down near the last port like this:

http://www.aceengineering.ca/Temp/tapermanifold-01.jpg

The taper gets rid of the stagnant slow moving air along the top of the plenum and aids the airflow to the further ports. Also having a flat floor makes things easier for fabrication in terms of welding in trumpets and keeping them further from any walls/floor surfaces.

joe's gt, if there is any issue with your model just let me know and I can take a look. It seems to be producing for you though. :D

joe's gt
03-20-2010, 05:55 AM
Thanks Conrad. You can just call me Joe btw. There is really no other way for me to do the 3d space runners. I have to loft them along a curve to get that smooth flowing shape and Cosmos refuses to analyze it. Cosmos is elementary software anyways. So in this regard, I have just decided to do flow analysis on the plenum and run the runners straight off (same inlet shape tho) to analyze the distribution. If Chris wants to do custom runners, the limitations in fabricating those will result in an easier runner analysis.

As for the taper, I have so many configurations yet to do. Full taper, custom taper, and each of those with a different throttle body angle configuration.

Oh and Chris, FINALS ARE DONE! So I can get crackin on those this weekend.

ChrisD
03-21-2010, 05:00 AM
Awesome! I'm excited. I also now have a spare 3SGE manifold just in case we need it (or if this one goes bad haha). So I've got both to use.

:)

Conrad_Turbo
03-22-2010, 04:20 PM
Cosmos is elementary software anyways

Yep I agree with you on that. It does a good job on a lot of things, but it does have some major limitations that are EXTREMELY frustrating when you've done everything right.

One thing that really helps when doing manifold design is actually considering how the manifold will be produced. The 2nd pic I posted is an old design of something I produced in the past. It's formed out of one sheet of aluminum for the main plenum (only 4 major bend lines) and then one thick plate welded on to mount the TB. It's one to have the best flowing manifold, it's another to have a good flowing manifold that is easy to build. Just keep that in mind. :D

joe's gt
03-22-2010, 11:02 PM
Yep I agree with you on that. It does a good job on a lot of things, but it does have some major limitations that are EXTREMELY frustrating when you've done everything right.

One thing that really helps when doing manifold design is actually considering how the manifold will be produced. The 2nd pic I posted is an old design of something I produced in the past. It's formed out of one sheet of aluminum for the main plenum (only 4 major bend lines) and then one thick plate welded on to mount the TB. It's one to have the best flowing manifold, it's another to have a good flowing manifold that is easy to build. Just keep that in mind. :D


Totally agree. All the designs I'm doing can be easily implemented by Chris by just cutting a straight line at whatever angle is needed (ie: tapers). It helps that he is going off pre-determined stock, because that forces our designs to be simple.

BTW: Chris, I am currently working on a time dependent analysis using your cam specs to show a flow animation instead of just having a static 2d simulation.

EDIT: So the time dependent analysis just takes to long and the Floworks software is too elementary to do an animation of the results. The time dependent analysis took like 3 hrs on my crappy CPU and when it finished, the animation options aren't even advanced enough to show whats happening. So unfortunately we are stuck with a static analysis for all of the options.

ChrisD
03-23-2010, 01:40 PM
lol, no prob. Static should do just fine. Based on the first one you posted above, I think it'll get us more than close enough. :)

joe's gt
03-25-2010, 03:45 AM
Hey guys, what is a good estimate for a pressure at the runner outlets? Chris is using a boost pressure of 17 psi at the runner inlet.

Also, should the flow entering and leaving the manifold be laminar or turbulent. If no one knows I guess i'll have to break out my textbooks.

EDIT: nvm, its better to assume turbulent. anyone have any ideas on the proper runner outlet pressure because I am running it at atmospheric right now and I know that is not accurate?

ChrisD
03-25-2010, 04:48 AM
I have no idea. If you model at atmosphere and then at 17psi, is there a significant difference in the result?

joe's gt
03-25-2010, 05:21 AM
I have no idea. If you model at atmosphere and then at 17psi, is there a significant difference in the result?

oh sorry chris, I think you misunderstood what I was saying. I have the throttle body inlet set at 17 psi. I just don't know what to set the runner outlets at. I know intake cylinder pressure varies with crank angle, but I figured I would use the lowest cylinder pressure that occurs during the intake stroke. I have no idea what a ball park figure for that value would be so I just assumed atmospheric, but I am sure it is below that, but I don't know by how much.

Basically, I need to know a ball park value for the lowest cylinder pressure that occurs during the intake stroke so I can set the runner outlets at that value for the analysis.

Conrad_Turbo
03-25-2010, 02:55 PM
Hey guys, what is a good estimate for a pressure at the runner outlets? Chris is using a boost pressure of 17 psi at the runner inlet.

Also, should the flow entering and leaving the manifold be laminar or turbulent. If no one knows I guess i'll have to break out my textbooks.

EDIT: nvm, its better to assume turbulent. anyone have any ideas on the proper runner outlet pressure because I am running it at atmospheric right now and I know that is not accurate?

Pressure would vary due to soundwaves. Imagine shaking bowl of jello, it's never still because of each valve opening and closing creating those pressure waves.

I'd assume turbulent, but it'd come down to runner inlet shape, air velocity/temperature and the cross-sectional area of the runner.

joe's gt
03-25-2010, 07:38 PM
Pressure would vary due to soundwaves. Imagine shaking bowl of jello, it's never still because of each valve opening and closing creating those pressure waves.

I'd assume turbulent, but it'd come down to runner inlet shape, air velocity/temperature and the cross-sectional area of the runner.

yeah, I know about the pressure, that's what makes it difficult without physical measurements. But I need to assign some ballpark value so I can do an analysis.

I would be assuming turbulent as well. Most of the stuff I have read on the net says turbulent and I am thinking the necessity of around 90 degree bend near the throttle body inlet will induce a turbulent flow anyway.

joe's gt
03-26-2010, 07:42 PM
For Reference:
HR = half round

HR Plenum; No Taper; 0* TB angle; Density Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/D-horizontal-velocity.jpg

HR Plenum; No Taper; 0* TB angle; Flow & Velocity Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/d-horizontal-trajectories.jpg

HR Plenum; No Taper; 10* TB angle; Density Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/D0density.jpg

HR Plenum; No Taper; 10* TB angle; Flow & Velocity Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/D10flow.jpg

HR Plenum; No Taper; 20* TB angle; Density Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/D20density.jpg

HR Plenum; No Taper; 20* TB angle; Flow & Velocity Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/D20flow.jpg

HR Plenum; No Taper; 30* TB angle; Density Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/D30density.jpg

HR Plenum; No Taper; 30* TB angle; Flow & Velocity Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/D30flow.jpg

HR Plenum; Full Taper; 0* TB angle; Density Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/DFT0density.jpg

HR Plenum; Full Taper; 0* TB angle; Flow & Velocity Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/DFT0flow.jpg

HR Plenum; Full Taper; 10* TB angle; Density Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/DFT10density.jpg

HR Plenum; Full Taper; 10* TB angle; Flow & Velocity Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/DFT10flow.jpg

HR Plenum; Full Taper; 20* TB angle; Density Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/DFT20density.jpg

HR Plenum; Full Taper; 20* TB angle; Flow & Velocity Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/DFT20flow.jpg

HR Plenum; Full Taper; 30* TB angle; Density Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/DFT30density.jpg

HR Plenum; Full Taper; 30* TB angle; Flow & Velocity Plot
http://i23.photobucket.com/albums/b396/92ct-27gt/DFT30flow.jpg

It is clear now why velocity stacks and/or angled runners are important in a side feed design. Air does not like to take sharp turns, so when its forced to do that it creates a low density area on the right side of the runner. So far Chris, if you don't want to use velocity stacks or 3sge runners, it looks like the option with the 20* TB angle and full taper gives you the best flow distribution and velocity.

NOTE: These velocity figures are most likely not correct because the runner outlets were modeled as atmospheric where in reality, the cylinder pressure is most likely a little lower. Either way, the fact that these numbers may not be accurate is irrelevant to the analysis because we are comparing these designs relative to one another, so as long as the parameters are the same for each configuration, the optimal solution can be found without the need for "real" velocities.

Some things to consider are that you could always use 3sge runners, velocity stacks, or we can also move the runners left and right. We also can do custom taper configurations. Just let me know how far left of the plenum opening you want the taper to start and the end height, or how far left and an angle. Keep in mind that the lowest you can go on the end is approximately 2" (the radius of curvature of the stock), otherwise if you cut it shorter than that, the radius will become smaller and you will have to cut the plenum floor at an angle to fit the smaller radius of curvature.

Also, unfortunately Chris, FloWorks does not give you the option of looking at a plot of CFM. You can specify it as an inlet or outlet flow for the initial conditions, but it doesn't give you the option of a plotting it for the results.

ChrisD
03-26-2010, 10:32 PM
D plenum is the one that is difficult to fab tho correct? Have you tried the half-round variations?

I now have a spare 3SGE manifold that I could use if need be too. Think this would be hard to model? I'm not opposed to using it at all.

ChrisD
03-26-2010, 10:33 PM
Also, we could play around with variations in the taper angle as well.

joe's gt
03-27-2010, 12:53 AM
D plenum is the one that is difficult to fab tho correct? Have you tried the half-round variations?

I now have a spare 3SGE manifold that I could use if need be too. Think this would be hard to model? I'm not opposed to using it at all.

Sorry Chris, I always mix them up. lol. I edited the above post. They are the half round plenums. I didn't even try the D Plenum because fabbing a taper from that wouldn't be straight forward because of the varying radius as you cut.

I can't model the entire 3sge runners, because once again they require a 3d space curvature that FloWorks doesn't like to recognize as a fluid volume. However, all I need is the dimensions of the runners and spacing between them as provided before and the angle at which each of them leave the bottom of the plenum and I should be able to rig something up pretty quick.

While we won't be able to see the flow through the entire runners, we can see how the plenum will distribute the air to all of them which is what we really want to know and optimize.

Anybody have an estimate for a cylinder pressure I could use during the intake stroke? Yes I know it varies, I just need a ball park number.

If I had a better program, I could probably do more with the runners and animations as well, but unfortunately FloWorks is quite elementary compared to other software such as Ansys Fluent.

I've thought about downloading fluent via torrent, but that would introduce a whole new learning curve that would probably not coincide with time constraints.

ChrisD
03-27-2010, 02:55 PM
Ah good. Ya the half-found would definitely be easiest to work with. I don't suppose there's an easy way to figure the plenum volume of the full taper with 20* TB? I want to make sure it stays around 2 Litres if I can.

Thats ok, we`re getting real good stuff with this program. This is really cool.

Can you pull the angles from this image below...should be able to. Now that I'm seeing the flow results I can definitely see the benefit of the angled runners. Would soften that turn significantly since the air really doesn't like to make sharp immediate turns if it doesn't have to. I'd pull the angle #'s but I'm out of town right now and typing on a netbook as it is. :P

http://i283.photobucket.com/albums/kk284/dudemanxl420/P7280047-1.jpg

BTW thanks again Joe, this is completely awesome. In all the IM build threads I've read for the 3SGTE, I've never seen this type of analysis done. Either the builder has done it in 'secret'...IE saying oh yeah i've optimized it, or no consideration was done at all. Even looking at the aftermarket ones available, this sort of thing clearly wasn't done, otherwise the designs would certainly have ended up different.

But that's why this is CTech!

joe's gt
03-27-2010, 06:15 PM
BTW thanks again Joe, this is completely awesome. In all the IM build threads I've read for the 3SGTE, I've never seen this type of analysis done. Either the builder has done it in 'secret'...IE saying oh yeah i've optimized it, or no consideration was done at all. Even looking at the aftermarket ones available, this sort of thing clearly wasn't done, otherwise the designs would certainly have ended up different.

But that's why this is CTech!

Yeah, the lack of sharing of information bugs me as much as it does you. I have no secrets, and I firmly believe that the more minds that can collaborate on ideas, the more productive and efficiently things can be accomplished. This is extremely prevalent with tuning information because people make a lot of money doing it and feel that they will lose massive profits if the information gets out there, when in reality you can only get close on the street and should still go to the dyno to extract that last bit of power and torque on the machine. Whatever, there's nothing I can do about it and I personally think that kind of attitude is counter productive.

As for the runners, the best I can do is estimate the angles from the pics. They look like they meet the plenum right next to each other from the pics, so I'll just model them that way. I guess I can also use the same dimensions from the 3sgte runners until you can get back and measure the actual dimensions.

Unless someone viewing this thread has a 3sge manifold where they can measure the runner dimensions?

EDIT: I assume since this is a bolt on upgrade for the 3sgte the runners are the same size?

joe's gt
03-27-2010, 10:59 PM
Hey Chris, I think I am just going to wait for you to get back and do some measurements. I will build the runners so they are adjustable, but won't actually do a flow analysis until you can get back and measure the actual angles and spacing where they meet the plenum. All you should need is a protractor to approximately measure the angles and a tape to approximately measure the spacing at the plenum. It looks like there is a little space between runners 1 & 2 and 3 & 4 where they meet the plenum.

EDIT: Sketches are done and runners are ready to be modeled. Chris, once you get back and can get me those measurements on the angles and the spacing where they meet the plenum, I'll model those up and redo the flow analysis.

ChrisD
03-29-2010, 05:45 AM
Cool, no prob. I'll should be able to get you the info by tomorrow evening at the latest. I already suspect that the angled runners are the way to go, at minimum the lower density spots could be reduced somewhat.

:)

ChrisD
03-29-2010, 04:59 PM
The runners are angled at 65 degrees. Or 115 if you go from the other side.

Here are a couple more pics just because...

http://www.celicatech.com/gallery/files/4/3sge_manifold1.jpg

http://www.celicatech.com/gallery/files/4/3sge_manifold2.jpg

Conrad_Turbo
03-29-2010, 05:22 PM
The pressure at the intake plenum where it meets the intake runner would be dependant on the rpm of the motor, the length and diameter of the runner. Since the resonance frequency would change throughout the rpm band the pressure would as well. For your CFD what rpm are you basing the flow rate at? If the rpm is at resonance for the intake the pressure will be very low at the intake plenum meeting the intake runner area.

joe's gt
03-29-2010, 07:04 PM
The pressure at the intake plenum where it meets the intake runner would be dependant on the rpm of the motor, the length and diameter of the runner. Since the resonance frequency would change throughout the rpm band the pressure would as well. For your CFD what rpm are you basing the flow rate at? If the rpm is at resonance for the intake the pressure will be very low at the intake plenum meeting the intake runner area.

Hey Conrad, thanks for your response. Um, I am just using 17 psi inlet pressure and atmospheric outlet. Originally I planned on doing a time dependent analysis with an animation at 5500 rpm, but the calculation took like 4 hrs, and when it was done, FloWorks didn't even have the capability to properly animate it. If I was using Ansys Fluent, I could probably do it, but FloWorks is just too elementary. As a result, I set all of the runners' outlet pressure at an environment pressure of atmospheric (which is inaccurate, I know) and the TB inlet pressure of 17 psi. To my knowledge, FloWorks does not have the capability of analyzing resonance effects.

Feel free to tell me if you think there is a better way. For now I am disregarding resonance effects because Chris is not designing his own runners. I don't know how I would even do a time dependent analysis without pressure values to go off of. I just stepped down to atmospheric when I tried to do it originally, essentially creating a square wave instead of the "real" parabola pressure curve. Should I somehow specify a flow rate for the runner outlets instead of a pressure?

I figure since the goal of this whole design is to find a plenum that will equalize flow to all the runners, an analysis with single inlet pressure and single outlet pressure for all the runners (providing the rpm is constant) would be adequate. Don't be afraid to tell me if I am way off base here tho. I am not as knowledgeable with this stuff as you, so I hope I am doing it right. I'm just trying to explain my thinking to you so you can tell me where I may be wrong and point me in the right direction.\

Thanks for the pics Chris. I guess they all come off at the same angle then and are right next to each other.

Conrad_Turbo
03-29-2010, 08:41 PM
For a "good enough" CFD result I would specify the mass flow rate of the intake air entering the manifold and split the input by four (for simplicity sake). To figure out the mass flow rate you'd need the intake air density (boost pressure and air temp), approximate CFM at the particular RPM you are designing the manifold for (and assuming the VE of the engine the manifold is designed for). A start would be to look at the turbo compressor map for the particular turbo being used. That way you get a more realistic result instead of stating 17psi inlet and assuming an outlet pressure (the ratio of the two would dictate your mass flow rate anyways).

Hope this helps.

joe's gt
03-29-2010, 09:35 PM
I can do that! Thanks so much Conrad. I plan to try and find the CFM from a map and as a last resort the equation in Alexander Bell's book:

CFM = (displacement x rpm x VE x pressure ratio)/5660

and then multiply the result by the density of air at the boost pressure and estimated intake temp to get a mass flow rate.

Will this work?

Conrad_Turbo
03-29-2010, 10:37 PM
You betcha buddy.

Mass flow rate (lb/min) = density(lb/ft^3) * volume flow rate (ft^3/min)

Hardest part is figuring out the VE...especially when people go changing their cams or modifying the intake/exhaust. Without dyno info it's a best guess, ie: fudge factor, but if you guess too high that will mean the manifold is being designed for a higher rpm and guess too low and it'll be designed for an rpm lower. Not a major concern though as there are larger variables assumed than that. :D

ChrisD
03-30-2010, 05:25 AM
My dyno is a few posts back...but of course I'll be changing cams again.

Can I also get the measurements to try to estimate the volume? Unless the software can do that?

joe's gt
03-30-2010, 05:53 AM
My dyno is a few posts back...but of course I'll be changing cams again.

Can I also get the measurements to try to estimate the volume? Unless the software can do that?

I am pretty sure the software can do that. I'll let you know the volume of the no taper and full taper when I do the next analysis. I think with Conrad's advice on setting mass flow rates instead of pressures we can really start nailing this thing down.

Conrad_Turbo
03-30-2010, 03:57 PM
Joe if you have any other Q's don't hesitate to ask, I'll do my best to answer them. :D

joe's gt
03-30-2010, 11:04 PM
Joe if you have any other Q's don't hesitate to ask, I'll do my best to answer them. :D

Will do. Thanks. It wasn't until you mentioned mass flow rates that I remembered back to my fluid dynamics course where we always set the mdot in to equal mdot out and then it made complete sense as to why it would be better to do it that way. Granted its not completely accurate for flow through a intake manifold but as you mentioned, it will work for a "good enough" analysis.

joe's gt
03-31-2010, 05:53 AM
Ok, so we are doing the analysis now with the 3sge runners and instead of using a pressure differential we are using a more accurate mass flow rate thanks to Conrad's suggestion. We are assuming steady state, which allows us to say that the mass flow rate in equals the mass flow rate out. This steady state assumption is not technically correct, but it will work as a "good enough" analysis for seeing how equally the plenum distributes air to the runners while at the same time maintaining equal flow rates. While FloWorks does not give a flow rate distribution option, it does give a velocity distribution which can easily be converted to a volume flow rate by multiplying this velocity by the runner exit area.

For those interested in how I calculated a mass flow rate to use, I'll explain below.

Mass Flow Rate = air density x volume flow rate

Volume Flow Rate

Volume flow rate is easy. This is just how much air your engine is ingesting at a given rpm and pressure ratio (ratio of boost pressure to atmospheric).

The equation used was from Alexander Bell's book Forced Induction Performance Tuning

CFM = [Displacement in liters x rpm x VE (estimate & doesn't effect final result that much) x pressure ratio] / 5660

CFM = [2 x 5500rpm x 95% VE x 2.15]/5660 = 397 cubic feet per minute(28.5 lb/min)

397 x .0004719 = .187 m^3/s

Density

Density is a function of temperature, pressure, and something called an ideal gas constant.

density = pressure / (gas constant * temperature)

Pressure is easy. This is just Chris's boost pressure of 17 psi (218535 Pascals absolute pressure)

Temperature is where it gets slightly complicated with fuzzy estimates.

Using an equation from Alexander Bell's book for temperature increase through the turbo:

Temp Increase = [F (factor from a table for a given pressure ratio) x (ambient temp in Kelvin) x 100] / (turbo efficiency at specific flow rate)

Temp Increase = [.242 x 288.5 K (15.5* C, 60* F) x 100] / 77 = 90* C increase

77 is from the 77% efficiency line for a 28.5 lb/min flow rate at a pressure ratio of 2.15 found on this (http://www.atpturbo.com/root/maps/gt28rs.htm) compressor map for a gt2860rs which I've been told is close enough to the ct20b

This is where it gets fuzzy. Assuming an FMIC efficiency of 75% (let me know if you guys think this is too much)

Intake temp = [90*C *(1-.75)] + 15* C (ambient) = 37.5 *C (310.5 K)

So finally...

density = pressure / [gas constant * temperature (K)]

density = 218535 Pa / [287.05 * 310.5 K] = 2.452 kg/m^3

And AT LAST....

Mass Flow Rate = volume flow rate x density

Mass Flow Rate = .187 m^3/s x 2.452 kg/m^3 = .459 kg/s

And there is just a tiny lesson in thermodynamics kids. lol.

p.s.: feel free to tell me if I screwed up anywhere, but the answer seems reasonable to me.

Conrad_Turbo
03-31-2010, 04:51 PM
I didn't double check the numbers, but the principles look good to me. :D

ChrisD
03-31-2010, 05:14 PM
Looks good. The only thing I see is that we are designing this for a FMIC instead of the ST205 IC. However, still seems like a good estimate. I'll be using a high density 10.5" X 24" X 3.5" core from Treadstone Performance. Sticking with that estimate seems fine tho.

joe's gt
03-31-2010, 06:17 PM
Looks good. The only thing I see is that we are designing this for a FMIC instead of the ST205 IC. However, still seems like a good estimate. I'll be using a high density 10.5" X 24" X 3.5" core from Treadstone Performance. Sticking with that estimate seems fine tho.


ah, you are right. my bad. edited the post, and I'll probably keep the same estimate as you suggested. The only real way to know anyway is to get the efficiency from the manufacturer or calculate it from before and after temps. I plan on doing the analysis today.

Berg
03-31-2010, 11:46 PM
The runners are angled at 65 degrees. Or 115 if you go from the other side.

Here are a couple more pics just because...


http://www.celicatech.com/gallery/files/4/3sge_manifold2.jpg

That manifold is filthy! I hope you didn't pay full price for it! :haha:

Are you planning of just cutting off the plenum and using the runners with your own plenum?

ChrisD
04-01-2010, 12:18 AM
lol! Yep, that's the plan. Will clean it all up tho, it will look brand new when I'm done. Also a few runs with my water injection and it will be 'steam cleaned'...haha.

joe's gt
04-01-2010, 07:06 PM
ran into some stuff for school yesterday so didn't quite get it done. I will finish it no later than the end of this weekend tho.

ChrisD
04-01-2010, 10:54 PM
No prob, although I am definitely excited to see the new and improved results. :D

joe's gt
04-04-2010, 10:14 PM
GrEAT NEWS! I found out how to look at volume and mass flow rates at the runner outlets. The analysis was done with a .459 kg/s mass flow rate at the inlet and atmospheric pressure at the outlet. So the actual number values aren't accurate, because the runner outlet pressure is obviously not atmospheric, but the analysis of the distribution will be accurate. This was done at 5500 rpm.

Here we go!

HR = half round
NT = no taper
FT = full taper
TB = Throttle body

Runner 1 is the left most runner and runner 8 is the right most runner.

HR, NT, 0* TB Density, Flow Velocity, & Flow rate data
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT0density.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT0flow.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT0data.jpg

HR,NT, 10* TB Density, Flow Velocity, & Flow rate data
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT10density.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT10flow.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT10data.jpg

HR, NT, 20* TB Density, Flow Velocity, & Flow rate data
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT20density.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT20flow.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT20data.jpg

HR, NT. 30* TB Density, Flow Velocity, & Flow rate data
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT30density.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT30flow.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeNT30data.jpg

HR, FT, 0* TB Density, Flow Velocity, & Flow rate data
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT0density.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT0flow.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT0data.jpg

HR, FT, 10* TB Density, Flow Velocity, & Flow rate data
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT10density.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT10flow.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT10data.jpg

HR, FT, 20* TB Density, Flow Velocity, & Flow rate data
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT20density.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT20flow.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT20data.jpg


HR, FT, 30* TB Density, Flow Velocity, & Flow rate data
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT30density.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT30flow.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/HRgeFT30data.jpg

I haven't really looked at them that much, cuz I spent hours doing this and just want to take a break, so you guys tell me which one is best. lol

The density plots this time around are so much better. From a quick observation is looks like the 20* TB angle on either the no taper or full taper plenum are the best best options for the most equal mass and volume flow rates. I still would like to get these more equal. If every runner flowed equally they would each have a mass flow rate of:

.459 / 8 = .0573 kg/s

Any suggestions? I don't feel like thinking right now. lol

ChrisD
04-04-2010, 10:35 PM
Wow, again awesome stuff. Initially, looks like 10 degree TB, full taper. But looks like might a tad to benefit from a 15* TB...something in between 10 and 20.

Off to easter dinner right now, but will take a closer look a bit later.

Seriously, thanks again for all of this Joe. This is really cool stuff.

edit: looks like the angled runners have almost fully eliminated the low density zones on the inside of each runner. Good improvement there.

joe's gt
04-05-2010, 08:09 AM
Wow, again awesome stuff. Initially, looks like 10 degree TB, full taper. But looks like might a tad to benefit from a 15* TB...something in between 10 and 20.

edit: looks like the angled runners have almost fully eliminated the low density zones on the inside of each runner. Good improvement there.

Yeah, with the angled runners the improvement in density is amazing. All runners are within 20-30 g/m^3 of each other.

As for the angles, I need to get my head out of the gutter. lol. I was locked into 10* increments and didn't even think about that. Will have the 15*, full taper up for you tomorrow.

oh and btw,

Volume for the No Taper HR plenum is 1.96 liters.

Volume for the Full Taper HR plenum is 1.48 liters

This is assuming a plenum of 11 inches long w/out end caps.

So the full taper is approximately 50% increase in volume and 75% of engine capacity. I've seen varying theories on plenum volume, so let me know what you want to do with that. We could increase the length or decrease the taper.

Also let me know if you want to do velocity stacks. At least we've narrowed down the optimum angles to 10-20*

Shadowlife25
04-05-2010, 12:17 PM
Great work there Joe. Wow!

Personally, I would futz with the TB angle just a bit to find the optimum distribution.
I am also partial to the HR NT design.
Though as per your tests the FT does smooth things out considerably just due to the nature of its design.

But, I am no engineer sir, if my commentary is completely off the mark, kindly disregard it. :)

I will continue to watch with interest.

ChrisD
04-05-2010, 05:51 PM
Wierd, the calcs I had on the HR no taper were a volume of 2.57L for 11 inches, and 2.8L for 12" long.

I think I have space to go up to 12" safely and have space for mounting, especially with the taper. With that said, I'm not sure it will help with anything really? Help me out with plenum volume theory here...

For velocity stacks...I have one thought, I'll post up in a sec.

Mario - the NT designs seem to have a lot of "dead zone" in the plenums...looks like it is very hard to get optimum flow to #1 and #8 runners, no matter how the angle is set. Looks like the density is lower also on all of them. Any reason you are partial to them other than ease to build?

ChrisD
04-05-2010, 05:59 PM
OK here's the only thing I can think of to create some sort of velocity stack for the factory runners. This sketch is somewhat exaggerated, but gets the point across. Have the stock runners protrude through the floor and round them off as best as possible to create the velocity stack "effect".

Experts please tell me if this would be helpful/useful in any way...

http://www.celicatech.com/gallery/files/4/img00021-20100405-10502.jpg

joe's gt
04-05-2010, 07:08 PM
If I read a textbook on intake manifold design, I could help you guys out, but honestly, I don't feel confident with giving you solid, "this is how to do it" design recommendations. Yeah, I am an engineer, but I have had literally zero experience in this area, so I am learning right along with you guys.

IMO, with velocity stacks, the optimum TB angle changes entirely and the no taper design could be more beneficial, because now there is a low pressure protrusion drawing the previous "dead zone" air in. The only advantage I see in using the tapered design is if we are not using velocity stacks.

Plenum volume is also another concern. I can't seem to find a general consensus on the ideal volume. People talk about plenum size and how it affects throttle response, which I don't quite understand, because the engine is just acting as N/A when your cruising. So I think throttle response is more so a matter of tuning that transition from cruise to boost, rather than plenum volume. Sure it can have an effect, but I would assume the plenum volume would have to be very large to have a noticeable effect.

Hopefully some theory experts like Conrad or Turrari will chime in here to confirm some of our assumptions. I'll hold off on any more analysis until we determine a general course of action for configurations.

ChrisD
04-05-2010, 07:24 PM
OK. At this point my preference is no velocity stacks, tapered plenum, angled runners, 15* TB once you run that one we can verify.

One interesting observation is that air density is measurably higher in the plenum when tapered. Seems to me that would be good for making HP.

ChrisD
04-06-2010, 06:55 PM
Joe - can you run the 15 degree analysis?

Conrad - do you have any thoughts on the results?

joe's gt
04-06-2010, 10:52 PM
Joe - can you run the 15 degree analysis?

Conrad - do you have any thoughts on the results?

Chris = will do! :)

Conrad = ditto to what Chris said

Conrad_Turbo
04-07-2010, 05:08 PM
I can't say I've ever done any analysis with just the runners poking in like that. It's either been a flat floor or having true velocity stacks inside the plenum, the advantage almost always going to the velocity stack configuration.

I'd assume having the intake manifold poke in like that would have somewhat similar effect to a velocity stack, where there would be a low pressure zone below the end of the runner and the floor of the manifold (similar to a velocity stack), but the lack of bellmouth might kill any benefit. Hard to say without actual CFD testing, but the fabrication aspect would be easier having the runner meet up with the floor of the manifold.

Plenum volume... Can vary from 1.4-1.6 x engine displacement for a FI engine, but then some cases it can be much higher. The smaller plenum lowers the RPM peak torque but if it is too small it will starve the top end. The large plenum will raise the RPM peak torque but will also cause bogging if too large. If I am committed to one manifold design and do not have the oppertunity to make multiple plenums to dyno, then go with a manifold that has very good CFD results and then compare the plenum volume to see if falls within the 1.4-1.6 x engine displacement. So for the 3S it'd fall in around 2.8-3.2L internal volume as a good ballpark size.

I would suggest going with the taper design, which would mean the manifold will have to be taller to compensate for the taper eating up volume. The extra height should aid with a more even airflow distribution as well.

Joe, don't leave too much of a flat spot on the end of the plenum opposite to the throttle body. Having 1-1.5" (somewhat dependant on how tall the plenum is as well) above the runner is more than enough to reduce stagnant and turbulant areas, which is very noticeable on the non-tapered manifold design, but much improved on your tapered design.

That's why I like this manifold design:
http://www.aceengineering.ca/Temp/tapermanifold-01.jpg

Good work so far Joe. :D

joe's gt
04-07-2010, 08:24 PM
So for the 3S it'd fall in around 2.8-3.2L internal volume as a good ballpark size.

I would suggest going with the taper design, which would mean the manifold will have to be taller to compensate for the taper eating up volume. The extra height should aid with a more even airflow distribution as well.

Joe, don't leave too much of a flat spot on the end of the plenum opposite to the throttle body. Having 1-1.5" (somewhat dependant on how tall the plenum is as well) above the runner is more than enough to reduce stagnant and turbulant areas, which is very noticeable on the non-tapered manifold design, but much improved on your tapered design.


Good work so far Joe. :D

Thanks Conrad, that was extremely helpful.

Chris, the Solidworks volume calculations are correct. I re-did the No Taper one by hand and it matched exactly. I am guessing you forgot to divide the circular area by 2 or maybe forgot to subtract .25" from each side for the thickness of the stock.

Here's how I calculated the No Taper:

[(1.75^2 * pi *11) / 2] + 3.5*1.75*11 = 120.291 in^3 (1.97 L)

Circle Radius: 2 - .25 = 1.75
Plenum Width: 4 - .25 - .25 = 3.5
Plenum Length: 11

As for increasing the volume. It looks like you are going to need to cut the halfround and use flat plates to lengthen the sides to make it taller. As it is now, the plenum end is 1.75" above the floor (equal to the radius of curvature of the inner wall) so any more cutting of that would result in decreasing the radius and cutting off the excess on the plenum floor. With a taller plenum, there should be more of a reduction of stagnant air, so hopefully this 1.75" will turn out to be ok.

Would you be willing to weld some quarter inch flat plates on to increase the plenum height and volume? How much space do you have to work with? Increasing length by an inch or so is also an option.

ChrisD
04-07-2010, 10:23 PM
Bah, you are right. I did not subtract the wall thickness. ;)

Although the adding the flat plates is possible, this is a bit extra work and might be best using a different base for the plenum at that point. I could extend to probably 12", and may have some space above depending on how low I cut the runners.

ChrisD
04-07-2010, 10:33 PM
Lets try to optimize a tapered design with 12" length, with 15* TB (if that is still the optimal angle), and see where that puts it. That should give it another .25L I'd think...

Failing that, we'd have to design something custom to get the volume up. Which isn't totally impossible, could be something like what Conrad has designed...

ChrisD
04-07-2010, 10:58 PM
A u-channel from mcmaster carr could do the trick to boost the height...hmm.

joe's gt
04-08-2010, 12:39 AM
Conrad says 1.4-1.6 x displacement. Alexander Bell in Forced Induction Performance Tuning says .8-1.5 x displacement. Corky Bell in Maximum Boost says .5-.7 x engine displacement. I'm surprised there is this much variation in the theory. That's a plenum volume variation from 1 L to 3 L. I'm gonna google more to see what I can find.

So I have no idea what plenum volume should be. After 5 google pages of reading search results I have come to the conclusion that there is no conclusion. It seems to be dependent on where you want to make your power and even then, there is no clear cut size.

Idk Chris, if you want to make your power in the upper rpms then i'd say bigger is better and go with Conrad's suggestion.

ChrisD
04-08-2010, 05:29 AM
lol, trouble is the definition of "upper RPM's" varies so much. For example, the RMR manifold makes its best power from 7000-10000 RPM. And you need a big turbo to get it going. I figure the RMR manifold is ~3.1L, if they are 4"x4"x12"....a guess, but seems reasonable.

I want 5000-7000 optimized, if I can. But...that still doesn't tell us how big of a plenum we need!

Yah, very confusing.

Well, you've done the analysis at 5500 RPM with roughly my flow numbers, and there's no restriction, right? What if we do one with the 15* angle, at 7000RPM? What will we see? Also, extend to 12"...I have room for that.

In the end, if the stock is 1 L and we go to 1.5, that's an improvement, plus I have the benefit of proper flow to each cyl.

I have no idea, haha.

joe's gt
04-08-2010, 07:08 AM
Alright, so extending it to 12 inches should give you some more volume. I figure if we can get to 2L that would be ideal. That would be twice the stock plenum volume and equal to the engine displacement. I doubt there would be the restriction you were seeing if we double the capacity.

We'll see!

I'll keep doing the analysis at 5500 rpm, and when we find what works, I'll test it out at 7000 rpm

ChrisD
04-08-2010, 02:01 PM
Sounds like a plan.

joe's gt
04-09-2010, 04:12 AM
Alright man, so to get the taper to 2L its gotta be about 5.5" from top to bottom on the right end if we taper it down to the minimum radius, which is a height of about 1.75" on the left end. My options for increasing volume are pretty much limitless. So, is increasing the height an extra 1.5" on the right end gonna be a problem? If so, we can decrease the taper and have a higher left end height, but that will also introduce more stagnant air.

Also, increasing to 12" barely gave you any more volume.

ChrisD
04-09-2010, 04:56 PM
Hm.

The two boundaries I have to deal with are the hood and strut bar at the top, and the wiring harness and fuel rail at the bottom. I measured the spare manifold and I'd have to cut the runners probably an inch or so lower to maintain the plenum height. I might have to take a look at my car to make sure.

Looking at the ATS manifold, it looks like there is enough room to move lower...

http://www.atsracing.net/atsintake%20manifold%20side%20feed.jpg

If I get a U-channel from Mcmaster Carr:

http://www.mcmaster.com/param/images/aluminum/unpolishedmillrounduchannel.gif

It's $16.61 plus shipping per ft. Base thickness is only .19", so that's what I'd have to play with to curve out the runner entry points. Gives me 2.25" 'legs' and a 4" wide base. Assembly / welding wise, I don't think it adds any complexity.

I dunno, go for it then? I'll take some measurements, but I think it should be okay to do it that way. Lets do it.

ChrisD
04-09-2010, 05:10 PM
I also recalculated the ATS design volume, looks like 2.89 L. (based on 12" length)

ChrisD
04-09-2010, 05:21 PM
One other option, using the D plenum since it is larger to start with. Basically make an angled cut for the taper, then cut to fit a flat piece of aluminum for the taper part. If ATS has 2.8L with no taper, I wonder if we can "taper out" only .8L and keep our 2L target?

A crappy sketch by me:

http://www.celicatech.com/gallery/files/4/img00022-20100407-1516.jpg

joe's gt
04-09-2010, 09:57 PM
^^ this is what I was trying to tell you before about cutting the D plenum and the varying base. But if you are fine with slapping some metal on the bottom and cutting to shape, we can model the D plenum as well.

I'll do the 5.5" tapered half round at a 12" length and a tapered D plenum. We don't really have any options of extending the sides of the D plenum tho because of the varying radius as we cut at an angle. I think the only purpose of the D plenum anyway would be to increase the volume as is. If we decided to add height to the plenum it would really only make sense to do it with the half round.

I think I have enough info to start doing some simulations now.

Also I saw the way you cut the D plenum for the angled throttle body. Would you prefer to do something like the following picture so you don't cut down on the volume?
http://i23.photobucket.com/albums/b396/92ct-27gt/fabdiagram.jpg

ChrisD
04-09-2010, 10:25 PM
What I am thinking is that the "base" will be a basic rectangle, but I'll cut the D plenum thru the top and slap on a flat peice of metal on the top for the taper.

Like this:

http://www.celicatech.com/gallery/files/4/manifold.jpg

For the throttle body, I can order the plenum extrusion to any length, so it's just as well to make an angle cut on it (also saves a weld requirement). If we make the bottom width 12" then the top could be marginally less, say 11" depending on the angle...or whatever is needed if we need the volume.

ChrisD
04-09-2010, 10:28 PM
The top piece would be cut something like this out of flat aluminum:

http://www.celicatech.com/gallery/files/4/manifold2_436335.jpg

joe's gt
04-10-2010, 03:13 AM
If you are planning on doing the HR plenum this same way the 5.5" high volume might actually be much less then 2L. I can re-model both the plenums in this manner if you prefer?

ChrisD
04-10-2010, 05:30 AM
No, only the D plenum that way b/c I want the base to remain constant. The half-round is good the way it was.

Will definitely be going one of these two ways...

joe's gt
04-10-2010, 05:46 AM
gotcha. I think we are on the same page now.

karl
04-11-2010, 03:24 AM
what happens to the flow if you angle the TB upwards into the plenum rather than at a specific pair of runners? or angle it toward the side walls of the plenum. use the TB to fill the plenum and let the runners draw from that.

thinking in terms of the IM used on the toyota eagle mk3.
http://www.allamericanracers.com/images/eagle_mk3_f411x326.jpg

joe's gt
04-11-2010, 08:25 AM
what happens to the flow if you angle the TB upwards into the plenum rather than at a specific pair of runners? or angle it toward the side walls of the plenum. use the TB to fill the plenum and let the runners draw from that.

This is a great suggestion and I love the thinking out of the box. I am no expert on this, but I believe the problem with this configuration is that the runners will temporarily be drawing in low density air until the plenum has time to "fill up". That is with the TB angled upward, high density air will be angled toward the top of the plenum creating a low density area near the runners until the plenum fills up.

As it is now, with the throttle body angled toward the runners, any changes in airflow and density are effective immediately because air is flowing directly into the runners and eventually the plenum is going to fill up anyway. High density air is directed right into the runners in addition to the plenum filling up. When you have a high volume plenum with the TB angled upward, I assume that this waiting for the plenum to fill up can negatively affect response. I am just going off intuition here though, and have no actual data or analysis to support these statements.

ChrisD
04-11-2010, 02:32 PM
Interesting, could be useful in a setup with antilag also, where boost doesn't necessarily drop off....ever. I would guess you would see a pretty even distribution especially in higher boost applications...could be speculation tho.

Also, they could have just done it b/c they didn't have the space to run it any other way.

:)

karl
04-11-2010, 06:18 PM
i'm just thinking that, considering how dominant the car was (in 1993, the only race they didn't win was the one they didn't enter), it couldn't have hurt them much. you'll notice too on OEM plenums like the ST215 that they don't angle the TB into the runners, but rather at the back of the plenum wall.

another possibility is a dual plenum design. lower plenum connects to the runners, upper plenum connects to the TB, and a slot connects the two chambers. some 4cyl LMP cars use this design, theory being that the slot between the two chambers allows for even distribution since the TB can fill the upper plenum, but the lower plenum draws from an opening the same size all the way across.

i had also looked at the cosworth manifolds for the mazda MZR-R engine, but they're almost a JUN-style bottom center feed setup.

joe's gt
04-11-2010, 07:44 PM
i'm just thinking that, considering how dominant the car was (in 1993, the only race they didn't win was the one they didn't enter), it couldn't have hurt them much. you'll notice too on OEM plenums like the ST215 that they don't angle the TB into the runners, but rather at the back of the plenum wall.

another possibility is a dual plenum design. lower plenum connects to the runners, upper plenum connects to the TB, and a slot connects the two chambers. some 4cyl LMP cars use this design, theory being that the slot between the two chambers allows for even distribution since the TB can fill the upper plenum, but the lower plenum draws from an opening the same size all the way across.

i had also looked at the cosworth manifolds for the mazda MZR-R engine, but they're almost a JUN-style bottom center feed setup.

I know what you are talking about with the dual plenum design and that is definitely superior. I don't know if you read the rest of the thread karl, but ChrisD is also looking for ease of fabrication. A dual plenum design would significantly increase the complexity of fabrication and space is also an issue. Those dual feed plenums also use velocity stacks which helps those designs tremendously. With Chris using the stock runners, trying to add velocity stacks to 8 irregular shapes right next to each other would be quite difficult.

As for the upward angled TB, I'll try a flow analysis on it and we'll see actual evidence of what really theoretically happens.

__________________________________________________

Great News Chris! The D plenum gave you A LOT more volume. With a 20* TB angle (worst case volume scenario) and tapering down to a 1.75" end height we have about 2.24 L of volume. Changing the taper greatly increases and decreases the volume. So if you want, we could go bigger or smaller on the end height depending on how much volume you want.

Flow analysis to come! Oh, and fabrication of the D plenum honestly doesn't look bad at all. Just cut at an angle. Place flat plates on all three sides and trace the outline. Cut out the outline. Then weld on. Hell you could just weld the flat plates on. The cutting is only for aesthetics anyway.

ChrisD
04-11-2010, 08:18 PM
Nice! 2.24 is really great, significantly more than the HR plenum. It's in the right zone (bigger than 2L), so whatever shows best flow and distribution is the taper to go with. LOL, 'only aesthetics'...I think it will end up looking really good. :)

Hehe, yeah the complex designs would be difficult and expensive to fabricate, plus space is a very important consideration. There's not a whole lot of room for creative designs in there. :)

joe's gt
04-11-2010, 08:27 PM
Nice! 2.24 is really great, significantly more than the HR plenum. It's in the right zone (bigger than 2L), so whatever shows best flow and distribution is the taper to go with. LOL, 'only aesthetics'...I think it will end up looking really good. :)

Hehe, yeah the complex designs would be difficult and expensive to fabricate, plus space is a very important consideration. There's not a whole lot of room for creative designs in there. :)

I also forgot to mention it prevents you from needing that enormous height of the HR plenum to achieve the same volume. The wider radius D plenum utilizes width much better than the HR plenum.

karl
04-11-2010, 08:32 PM
I know what you are talking about with the dual plenum design and that is definitely superior. I don't know if you read the rest of the thread karl, but ChrisD is also looking for ease of fabrication. A dual plenum design would significantly increase the complexity of fabrication and space is also an issue.

also true. i'm trying to think of a way to make it less complex, but my brain is taking the weekend off. i bet finding an intercooler end tank of sufficient volume would make manifold fabrication a lot easier.

honestly though, i think it's worth trying to find pics of road race 4cylinder manifolds like the ones professional race teams like LMP, GT etc use (not drag 4cylinders). they've likely done much of the CFD research on them. i'm trying to remember what we used on our FSAE cars back in the day, though those manifolds had a restrictor to deal with as well. might be worth a browse through some school's SAE projects too, since many of them publish their data for class.

also, if you can find any SAE papers on IM design, let me know. i have a contact who can get any paper i ask for.

joe's gt
04-11-2010, 08:49 PM
also true. i'm trying to think of a way to make it less complex, but my brain is taking the weekend off.

honestly though, i think it's worth trying to find pics of road race 4cylinder manifolds like the ones professional race teams like LMP, GT etc use (not drag 4cylinders). they've likely done much of the CFD research on them. i'm trying to remember what we used on our FSAE cars back in the day, though those manifolds had a restrictor to deal with as well. might be worth a browse through some school's SAE projects too, since many of them publish their data for class.

also, if you can find any SAE papers on IM design, let me know. i have a contact who can get any paper i ask for.

Most of the FSAE team manifolds I have seen utilize a triangular design with the intake feeding the plenum on top directly in the center and a symmetrical taper to the left and right that usually feeds a total of 4 equally spaced runners. Yeah, I don't know how I would get SAE papers. Don't you normally have to pay for those things?

karl
04-11-2010, 11:07 PM
Yeah, I don't know how I would get SAE papers. Don't you normally have to pay for those things?

sorry, i meant if you come across any abstracts on IM design that you think would be interesting, i can probably get them for you. a have a friend at honeywell who has access to most all of the papers published in the past 10 years.

joe's gt
04-12-2010, 12:37 AM
sorry, i meant if you come across any abstracts on IM design that you think would be interesting, i can probably get them for you. a have a friend at honeywell who has access to most all of the papers published in the past 10 years.

Bastard...J/K. lol. That is so cool that you have access to that kind of material. I can only imagine how reading those things could have helped me understand concepts better instead sorting through pages and pages of BS on the interwebz trying to find accurate information. Unfortunately along with all the great information on the net comes an almost equal amount of useless information. Whatever you do, stay on good terms with that guy. Access to those papers is a huge advantage and could save you a lot of money and headaches.

Pic of dual plenum for those that want to know that karl is suggesting. He is totally right that it is the better option, but as always, fabrication is the issue.

http://i23.photobucket.com/albums/b396/92ct-27gt/dualplenum-1.jpg
http://i23.photobucket.com/albums/b396/92ct-27gt/dualplenum1.jpg

joe's gt
04-12-2010, 05:13 AM
Hey Chris, just wanted to let you know that I am not posting anything up tonight because I am at the point where I am just trying to optimize the designs with different tapers and throttle body angles. I just wanted to let you know that I've been running tons of simulations today and will post up pictures when I have optimized designs.

What is the absolute minimum and maximum plenum volume you want? That way I can design the taper on the D plenum to match those requirements. Getting the D plenum above 2L is no problem and there is much more room to grow.

ChrisD
04-12-2010, 04:24 PM
No prob at all.

Absolute minimum, lets say 2.0L, which I think is pretty easy for you to get. Basically anything in the range of 2.2-2.5 would be awesome. Definitely want a "decent" taper there to optimize the distribution, so I'd say if you can reach a point where distribution is not declining significantly but you can increase volume somewhat, then go for it. Conversely, if distribution is suffering then feel free to increase the taper and reduce the volume accordingly.

If the optimal distribution will come with the 2.24L example, then lets go with that and then optimize the TB angle.

Having 3 variables makes optimizing rather tricky!

joe's gt
04-19-2010, 07:25 AM
Alright Chris. This is the best I could do. I varied the TB angles and taper end heights. I also experimented with where the taper started which effected volume much more than distribution.

The result with the best distribution was a 17* TB angle and a taper that started right at the middle of the plenum and tapered down to .75" (measured from the very bottom of the plenum to the cut height). I located the center of the TB 2 inches below (along the 17 degree axis) the top of the plenum. We will have to do more simulations if you choose to move the TB to a different position.

The total resultant volume is 2.25 L

http://i23.photobucket.com/albums/b396/92ct-27gt/DgeFT17density.jpg

http://i23.photobucket.com/albums/b396/92ct-27gt/DgeFT17flow.jpg

http://i23.photobucket.com/albums/b396/92ct-27gt/DgeFT17data.jpg

I personally think this analysis is for more of a transient type response anyways. Once the plenum is fully pressurized, I'd imagine that as long as there is enough air to feed all cylinders for the given rpm, the cylinders will pull whatever they need. I'd love to read a textbook or something on manifold design, but those tend to be expensive.

ChrisD
04-19-2010, 05:41 PM
Un-freaking-real. Love it. ^repped.

Distribution looks like the best of any we've seen so far. Volume is great at 2.25L, a 125% improvement from stock. Also, .75L above the 3SGE. Construction looks very do-able.

I will need the dimensions of the components so I know what to order. What ended up being the length of the base? Also, I've calculated that my TB flange should be 4.6" in height, is this the # that you have? Max width is unchanged at 5".

I'll let Conrad take a peek and offer his opinion, but I think we're at the stage where I can probably place the order for materials! :woot:

One special request, can you get a screen capture of the 3D view of the velocity analysis? Only b/c it looks cool. :P

joe's gt
04-21-2010, 07:12 AM
Check it out buddy. So I just was not satisfied with how little air runner 8 was getting so I experimented with lengthening the plenum in quarter inch increments between 12 and 13".
And that experimentation led to this...

http://i23.photobucket.com/albums/b396/92ct-27gt/DgeFT17TB1225inDensity.jpg

http://i23.photobucket.com/albums/b396/92ct-27gt/DgeFT17TB1225inFlow.jpg

http://i23.photobucket.com/albums/b396/92ct-27gt/DgeFT17TB1225inData.jpg

Check out the HUGE improvement in the equality of mass flow rates and the amount of air runner 8 now gets. A full .01 kg/s more. I think this is the best design yet and you only have to extend the plenum an extra quarter of an inch. That's right, just a quarter of an inch extension yielded this improvement. The TB was just too close to runner 8. I also checked 12.5" and 12.75" and at that point it began to starve runners 1 and 2. So the best length was 12.25" on the bottom without including the end cap thickness. I'll have drawings for you up tomorrow.

ChrisD
04-21-2010, 04:27 PM
Wow. I am seriously impressed! Great stuff, love the improvement.

:bigthumbu

Conrad_Turbo
04-21-2010, 04:37 PM
Looks very good. Good job Joe!

joe's gt
04-21-2010, 08:11 PM
http://i23.photobucket.com/albums/b396/92ct-27gt/DPlenum17TB1225L.jpg

http://i23.photobucket.com/albums/b396/92ct-27gt/PlenumCuts.jpg

http://i23.photobucket.com/albums/b396/92ct-27gt/TBplacement.jpg

Total Plenum Volume = 2.29 L

I didn't put drawings of the end caps up because I think its better for you to draw an outline after you make the cuts, instead of going on specified dimensions because the caps are kind of weird shapes.

So Chris, is this where you were planning on the center of your TB? If not tell me how far from the top or bottom and then I can move it and do some more simulations to make sure the TB angle and length are still optimum.

Also here is a 3D pic of the flow. Let me know if you were looking for something different or want a specified view.

http://i23.photobucket.com/albums/b396/92ct-27gt/3dpic.jpg

ChrisD
04-21-2010, 08:23 PM
Thanks for those!!

TB position is perfect. The only other measurement I think I'll need is the distance from the edge of the base plate to the runner on both sides.

For the 3D view, maybe something looking a little more straight at it instead of looking up at it? Very cool regardless.

joe's gt
04-24-2010, 03:40 AM
http://i23.photobucket.com/albums/b396/92ct-27gt/Runnerplacement.jpg

http://i23.photobucket.com/albums/b396/92ct-27gt/3dflow.jpg

ChrisD
04-24-2010, 05:23 AM
You rock. Thanks. :)

I guess I need to move on to getting the supplies!

joe's gt
04-24-2010, 06:17 AM
yeah dude, I am hella anxious to see if this yields you any gains. That would be so cool if you make more power in the upper rpms. Also, if it works out really well, maybe other members could use this design on their 3sge manifolds. We'll see. Keep me updated :)

And I'd also like to say what a pleasurable learning experience this has been for me. Thanks for giving me the opportunity to help you out and having enough faith in me to aid in the design of this plenum for you. Also, thanks to everyone who contributed helpful information to this thread. This has honestly been the coolest experience I have been a part of on Ctech.

ChrisD
04-24-2010, 03:39 PM
That begs an interesting question. When do I dyno? Right now, I have a dyno dynamics dyno...I'd prefer dynojet in the end.

There are 3 variables I have to install. 1. manifold, 2. intercooler, 3. cams. Trying to orchestrate it to isolate the manifolds gains only is a bit tricky. Sure, I could do one after every change, but that's expensive! Or I could street dyno and call it a day since that is free, haha.

Unless I can work a deal with a local dyno to give me a deal on 2 baselines a month apart for a discount....

joe's gt
04-25-2010, 04:24 AM
That begs an interesting question. When do I dyno? Right now, I have a dyno dynamics dyno...I'd prefer dynojet in the end.

There are 3 variables I have to install. 1. manifold, 2. intercooler, 3. cams. Trying to orchestrate it to isolate the manifolds gains only is a bit tricky. Sure, I could do one after every change, but that's expensive! Or I could street dyno and call it a day since that is free, haha.

Unless I can work a deal with a local dyno to give me a deal on 2 baselines a month apart for a discount....

lol. Chris don't even worry about trying to isolate the manifold. Go ahead and do your other mods unless you can work out a deal with the dyno guys. As long as you don't lose power I'll be happy. lol.

ChrisD
04-26-2010, 04:40 AM
Well, we'll see what I can do. I would really like to see the difference too if I can. At a minimum, I'll be able to do street dynos. They are pretty accurate from my experience. I definitely want to get a dynojet run in to see how my setup compares to others I have seen with similar mods.

Probably a month or so before the changes will happen on the car, but I may be able to begin to build the manifold depending on what sort of free time I can get. (given that I'm also in the middle of Ryan's swap)

joe's gt
04-26-2010, 07:43 AM
No rush man. Just keep us updated :)

ChrisD
05-16-2010, 06:35 PM
Got some goodies in the mail!

The plenum is really large. Bigger than I had visualized it to be. I tried to take some pics to get the feel for the size, but it's really when you're holding it in your hand that your like :eek4:

http://www.celicatech.com/gallery/files/4/img_0292.jpg

http://www.celicatech.com/gallery/files/4/img_0294.jpg

http://www.celicatech.com/gallery/files/4/img_0295.jpg

http://www.celicatech.com/gallery/files/4/img_0296.jpg

And another new toy, a portable metal cutting band saw! Tried it out last night, makes really quick work of aluminum. :D

http://www.celicatech.com/gallery/files/4/img_0297.jpg

ChrisD
05-16-2010, 06:39 PM
You can see in the first pic, I bought a big chunk of 1/2" stock. I was thinking of using this for the base of the plenum, so that I could radius the runner ports (like ATS does). Thoughts? It might be tricky, b/c I have to keep the angle of the runners which means I can't just cut it out straight up/down. Or I could do that with the plenum without hacking it up, but radius'ing the ports might be tricky.

Just trying to figure out the best way to start assembly.

Rio_Cyber
05-16-2010, 07:08 PM
anybody has any CAD drawings off bottom plate maybe :)

joe's gt
05-16-2010, 07:43 PM
AWESOME!!! Can't wait to see a dyno when its done. Hey Chris, I am just kind of worried about that 17* angle. Are you going to be able to fit the pipes under the hood with that angle?

Also are we going to need to angle the throttle body toward the front of the car like on the stock 3sge manifold? Or are there couplers or pipes available with a gentle 90* bend?

ChrisD
05-17-2010, 04:09 AM
AWESOME!!! Can't wait to see a dyno when its done. Hey Chris, I am just kind of worried about that 17* angle. Are you going to be able to fit the pipes under the hood with that angle?

Also are we going to need to angle the throttle body toward the front of the car like on the stock 3sge manifold? Or are there couplers or pipes available with a gentle 90* bend?

I am hoping to be able to get the runners short enough that I can clear the hood. I should be able to rough-mock up everything before commiting to the taper angles.

I have 90* mandrel bends, should be ok. I suppose an angle similar to the 3SGE could be done, however that does make the cut a little more tricky. Doable, but tricky. I'd probably rather just use the mandrel bent pipe.

What do you think about radius'ing the runner ports?

joe's gt
05-17-2010, 05:17 AM
Anything to smooth that transition would help. I wouldn't put it together without doing that. Air does not like sharp corners as we could see from the low density areas on previous analysis. The analysis I did didn't involve radiusing, so I'd imagine the numbers would be even better if you could do that.

Yeah, I don't want you to have to do that forward angle either, cuz then I have to re-analyze the design. lol. If we do end up having to do that, it would probably be better to move the runners back since the air would flow along the back wall.

ChrisD
05-17-2010, 05:11 PM
Hm. I wonder if I can do the radiusing (is that a word?) without chopping the bottom off. That would certainly make it a bit easier to build, also save welding time/costs. For a one off, I could probably figure a way to do it, even if it is by hand with little dremel attachments or possibly sandpaper. Again, as a one-off type thing. haha.

The alternative is chopping off the bottom, having the 1/2" plate welded on, drilling, cutting, sanding, dremeling, to get the radiusing, then back to the welders to have the top welded on.

joe's gt
05-17-2010, 07:16 PM
radiusing (is that a word?)

no its not. but it should be.

Conrad_Turbo
05-19-2010, 07:08 PM
I cheat and make trumpets with dimple die tooling...

http://www.aceengineering.ca/images/IMG_8000.jpg

Cut the excess material off and weld the flared end, this is only if I can't flare the manifold runners to begin with.

In your case Chris I'd just get an air die grinder and a good HSS rotary file and some candle wax to help release any chips from the bit. A good sanpaper roll will clean up the last bit of it.

joe's gt
05-19-2010, 09:20 PM
I cheat and make trumpets with dimple die tooling...

Cut the excess material off and weld the flared end, this is only if I can't flare the manifold runners to begin with.

In your case Chris I'd just get an air die grinder and a good HSS rotary file and some candle wax to help release any chips from the bit. A good sanpaper roll will clean up the last bit of it.

Dude, Conrad, where did you learn all this information about all these cool fabrication methods?

Conrad_Turbo
05-20-2010, 12:58 AM
All on my own. Just do things that forces you to question and learn all the time. A lot of it comes from personal investment, $$$ and time.

Just realized I can't spell sandpaper though. :laugh:

You did fantastic work with the CFD work in this thread. :D

ChrisD
06-15-2010, 04:26 PM
So a bit of a snag. Although the 3rd gen TB is 60mm, the rubber coupler that the ST205 is actually quite a bit bigger. I had expected that my 2.5" (63.5mm) intercooler piping would fit in snugly and be ok. However, it's looking more like 2.75-3" pipe would fit in that coupler. So I don't think I can use that unfortunately.

Now the 3rd gen MR2 TB inlet is a bit different than the gen 2 as well. As you can see, there is a sandwhich plate and then the inlet, which from what I have read is pretty close to 60-65mm and would work, technically.

http://admin.rpmgarage.com/images/attachments/14206_7fa6u9e6fzpn.jpg
http://admin.rpmgarage.com/images/attachments/14206_x2cib14zu2kw.jpg
http://admin.rpmgarage.com/images/attachments/14206_a7lk9hes9053.jpg

So to purchase this would cost a few extra bucks, but also more importantly in the design, would extend the throttle body use of space by several inches (versus a TB with integrated inlet, such as the Supra versions). And since we have the TB angled towards the hood, I want to keep it as compact as possible!

So I'm not sure using the 3rd gen throttle body is the way to go here. I might be able to switch to a different TB since I have to buy something anyway, and also save some space. On the plus side I could sell my whole adaptor/TB/coupler as a 205 WTA TB package and probably pay for the new TB.

So I'll likely try to look for something comparable in size, 60-65mm since my IC piping is 63.5mm already. But there are other options such as the Mustang (75mm) and Q45 (90mm). Or aftermarket TB's, which I am more skeptical about quality, and also I'd need to research the effects of the ISC valve and what exactly my Power FC does to control it.

If anyone can save me time by confirming which Toyota TB's might fit the bill that would be awesome, otherwise I'll start researching. (likely starting with 5MGE & 7MGTE)

Also, Joe, will it be very difficult to remodel this if I do end up changing the TB size? I'll shoot for 60-65mm, but would like to do a quick check to make sure none of the other measurements need to change if I do change the TB size slightly.

:)

racerx1715
06-15-2010, 06:31 PM
All of the flow modeling is very impressive! Sadly I haven't had access to that since I was at toyota.

ChrisD
06-15-2010, 10:59 PM
Thanks! Props to Joe and Conrad, they are the brains. I just ask for stuff. :P

For those who don't know, racerx has been developing some really cool intake and exhaust manifolds for the MR2 community as of late. All made to order, and feature a lot of very cool features such as tapered runners and some very interesting velocity stack ideas. He recently compared the RMR to one of his designs, and demonstrated some significant gains throughout the powerband. (not a shot against RMR, it's best suited to 600+whp cars imho...)

Anyway, thanks for posting up! :)

joe's gt
06-16-2010, 12:54 AM
Yeah, no problem. Just let me now what you want to try and we'll do some more analysis to make sure the current dimensions still yield the best flow.

Gandalf
06-16-2010, 01:58 AM
I am watching all of this with great interest... I am using a 3sge (lc) manifold, utilizing the runners, and my 3sgte TVIS... and making a custom plenum from a lot of aluminum tubing I have lying around.

Keep up the great work!

ChrisD
06-16-2010, 05:28 PM
OK, 7MGTE has a 60mm TB, looking at this one closely.

joe's gt
06-16-2010, 07:35 PM
Ok. cool. Yeah all I need to know is the diameter of the TB for now. The only thing that will affect the distribution is if the TB hole opening is moved up or down along the angle cut or if we need to shorten or extend the plenum. Otherwise, the length of the TB is irrelevant, because the air is still entering the plenum at the same place. The only thing the length of the TB affects on your end is fitment.

Berg
06-16-2010, 08:17 PM
I have a question about the TB and its placement. Nothing really to do with where it is placed on the plenum, but how it is oriented. Are you just looking at wide open throttle situations? I'm curious about partially open throttle, say half throttle. The throttle plate in TB at half throttle allows air in around the edges of the TB. Would it be better to have this plate with the openning at the top and bottom with reference to the intake runners (like a stock 3S-GE or 4A-GE) or would it be better to rotate it 90 degrees (or some other rotation)? Would it be better for the top to tilt towards the intake as it opens to angle the air down into the runners?

Or as I asked is this simply for WOT considerations and my questions are irrelavent?

Cheers,
Berg

joe's gt
06-16-2010, 08:27 PM
Hey Berg, the design was optimized for 5500 rpm. I could do a calculation at a lower rpm, but I would have to recalculate the mass flow rate. I could do it if Chris wants to. We were mainly concerned about flow distribution rather than resonance tuning or anything like that where different loads would have a much greater effect. This is because for ease of fabrication we are just going with 3sge runners, so we don't have any control over where we can make the power besides the size of the plenum. Chris needs more up top, so naturally we just designed for more volume. He was clear from the beginning that he was going for upper rpm power.

As for the TB thing. That's a good question. Haven't thought about it much. Maybe Conrad could chime in.

Something I thought about as well, but never modeled or looked into that much was having the TB angled toward the back wall of the plenum and having the runners near the back so the air could flow along the back wall and into the runners, instead of spreading out in the wide open plenum space.

Gandalf
06-27-2010, 01:00 AM
Any new updates/developments on this? :)

joe's gt
06-27-2010, 01:47 AM
I think Chris is close to building it. I understand he has been really busy working on his brother's car, so I think its been put on the back burner for now. He is gonna build it tho, he already got the materials.

takka
06-27-2010, 12:42 PM
hey joe iv been following the thread, an you guys have been doing some awesome work. but i cant help but think the half round would perform better with a constant taper, i did a quick sketch in paint of a good way to bump volume just use the piece removed from the base for the taper in reverse. it would give the 2 liters of the none taper, or more by adding side wall material and adjust throttle angle accordingly. i would love if you modeled this design?? http://i964.photobucket.com/albums/ae130/takkaz/plenum.jpg

takka
06-27-2010, 02:07 PM
also 1UZ-FE thottlebody has a 70mm diameter it would nice to see how that effects flow

ChrisD
06-27-2010, 04:58 PM
Yes still working on things. My car is already taken out of commission and work has begun. The IM is one of several items I will complete before starting it up again. Cams, FMIC, fixing oil leaks, TRD T-stat, and a lot of cleaning up will happen.

That design would cause a lot of installation issues, and also cost more for welding. I have to be very careful with hood clearance, and building it up like that makes an already tough job even harder. The D-shaped plenum does a much better job of creating volume with limited space.

Thanks for the tip on the 1uzfe TB! I'll check it out. :) Although I'm still more likely to scoop up a 7mgte due to cost/availability...also, I'm not likely to outflow it at this point. :)

joe's gt
06-27-2010, 06:53 PM
hey joe iv been following the thread, an you guys have been doing some awesome work. but i cant help but think the half round would perform better with a constant taper, i did a quick sketch in paint of a good way to bump volume just use the piece removed from the base for the taper in reverse. it would give the 2 liters of the none taper, or more by adding side wall material and adjust throttle angle accordingly. i would love if you modeled this design??

Thanks takka :)

I see what you are saying, and this has already been thought about and modeled. I just didn't post it here. The main problem with this idea is it significantly raises the throttle body height and hood clearance is a huge issue here. The plenum is gonna sit high as it is, not to mention the upward angle of the pipe that is going to have to come off of it as well. You would be surprised at how little a full taper vs. a partial taper effect there is based on the CFD. The main thing affecting distribution was the throttle body angle and placement of the runners. A full taper vs. partial taper yielded VERY similar results. The only thing it really affects is volume, which is important for when the plenum is full.

But you are right, except you could yield much more volume if you use the "D" plenum and cut it in half and then added the side walls and base. The halfround is of a lesser diameter. The whole problem here is added height.

This CFD is more pertinent for a transient type of analysis, because once the plenum is full, there is enough air anyways, and the cylinders just pull what they need. With this whole project, the idea was to basically increase stock size and equalize as much as possible transient distribution.

takka
06-28-2010, 12:51 AM
yeah ok, i keep forgetting about the hood im thinking about it from mr2's perspective an the added 2 inches shouldn't be a problem whereas width of the D plenum might. iv been planing this for awhile now an i have the 3SGE manifold ready to go, chris whats the plan with the runner openings, im thinking if you designed a nice raised stack base plate(similar to racerx's latest manifold) that matches the runner ports it wouldn't be to expensive to be machined?

an joe i know what im asking doesn't apply to what you guys are doing, but think the the removal of the turbulent zone created by the short taper which is disturbing flow to runner 1/2 and the added density of the full taper would make a big difference and also the larger 70mm throttle would have to increase the flow span across the floor and could yield allot better distribution between runners. i know the throttle angle has a massive role in this and thats why i wanted to see this design analysed before i slap it together and find im starving one cylinder, im studying engineering in australia maybe i should get some software an start playing around :) anyway cant thank you guys enough there is really good information here.

joe's gt
06-28-2010, 04:37 AM
yeah ok, i keep forgetting about the hood im thinking about it from mr2's perspective an the added 2 inches shouldn't be a problem whereas width of the D plenum might. iv been planing this for awhile now an i have the 3SGE manifold ready to go, chris whats the plan with the runner openings, im thinking if you designed a nice raised stack base plate(similar to racerx's latest manifold) that matches the runner ports it wouldn't be to expensive to be machined?

an joe i know what im asking doesn't apply to what you guys are doing, but think the the removal of the turbulent zone created by the short taper which is disturbing flow to runner 1/2 and the added density of the full taper would make a big difference and also the larger 70mm throttle would have to increase the flow span across the floor and could yield allot better distribution between runners. i know the throttle angle has a massive role in this and thats why i wanted to see this design analysed before i slap it together and find im starving one cylinder, im studying engineering in australia maybe i should get some software an start playing around :) anyway cant thank you guys enough there is really good information here.

I think I could try rig something up for you. Are you planning on using the 3sge runners? This is by far the best option second to making your own. You are right in that a 70mm throttle body will spread across the floor much better and the optimum angle might be different than for Chris' design.

And believe it or not, that turbulent zone is still there even with a full taper. It is caused by the height of the left end, and not necessarily the taper. In my simulations, the angle and length of the taper barely had any effect on the distribution. It was more so the left end height.

But if you want to start a new thread, we can find out more about what you want to do and I can help you out. We don't want to hijack Chris' thread ;)

ChrisD
06-28-2010, 04:36 PM
Checked out the 1UZFE TB - it's a little too long compared to the 7MGTE, so 7MGTE is still leading the pack. But I'll take a look at others as well, maybe some Mustang TB's...not sure. I don't have very much cash for the TB, so I've got to go with something stock/used.

Joe - from a design perspective, would you prefer a larger TB, or do you think the 60mm will be sufficient, as planned?

ChrisD
06-28-2010, 04:36 PM
Checked out the 1UZFE TB - it's a little too long compared to the 7MGTE, so 7MGTE is still leading the pack. But I'll take a look at others as well, maybe some Mustang TB's...not sure. I don't have very much cash for the TB, so I've got to go with something stock/used.

Joe - from a design perspective, would you prefer a larger TB, or do you think the 60mm will be sufficient, as planned?

joe's gt
06-28-2010, 10:09 PM
Checked out the 1UZFE TB - it's a little too long compared to the 7MGTE, so 7MGTE is still leading the pack. But I'll take a look at others as well, maybe some Mustang TB's...not sure. I don't have very much cash for the TB, so I've got to go with something stock/used.

Joe - from a design perspective, would you prefer a larger TB, or do you think the 60mm will be sufficient, as planned?

I would imagine, as takka already said, that a larger TB would spread across the floor better and provide more air. However, with a larger opening you also sacrifice flow velocity. But since you want to make your power in the upper rpms, I don't see how a larger TB could hurt you that much. I'll do more research on it and try to find out how big of an effect it would have.

So I looked up some stuff and it appears that for your desires of upper rpm power and throttle response being irrelevant, bigger is better. Also throttle body is less of a factor on a turbo car as compared to NA.

takka
06-29-2010, 02:15 AM
there are some guys on mr2oc using a modified 70mm bbk mustang throttle?? looks pretty neat an you can pick them up for under $200

Conrad_Turbo
06-29-2010, 05:03 PM
I have a question about the TB and its placement. Nothing really to do with where it is placed on the plenum, but how it is oriented. Are you just looking at wide open throttle situations? I'm curious about partially open throttle, say half throttle. The throttle plate in TB at half throttle allows air in around the edges of the TB. Would it be better to have this plate with the openning at the top and bottom with reference to the intake runners (like a stock 3S-GE or 4A-GE) or would it be better to rotate it 90 degrees (or some other rotation)? Would it be better for the top to tilt towards the intake as it opens to angle the air down into the runners?

Or as I asked is this simply for WOT considerations and my questions are irrelavent?

Cheers,
Berg

It's all about compromise. With a performance orientated manifold it will be designed at WOT, so the throttle plate will be disregarded. The orientation of the TB at partial throttle affects airflow regardless of it's position. I did some testing a long time ago and seemed to notice the longest distance from the axis of the TB to exterior manifold wall should have the axis of the throttle body shaft oriented in that direction. It had the least impact on airflow under partial throttle as opposed to the TB being rotated 90. However this is done with CFD and with an actual engine there are larger factors that come into play (intake layout pre TB and valve lift and duration being a major factor).

At the end of the day it's easier to design for WOT. :D

Sifu
06-29-2010, 08:40 PM
there are some guys on mr2oc using a modified 70mm bbk mustang throttle?? looks pretty neat an you can pick them up for under $200

Just make sure it's a BBK and NOT Accufab... Accufab's have been known to snap from pressure.

Sang
06-29-2010, 11:13 PM
there are some guys on mr2oc using a modified 70mm bbk mustang throttle?? looks pretty neat an you can pick them up for under $200

By modified you mean it uses the stock TPS and IACV? If so they go for quite a bit more than $200. If you find a BBK for dirt cheap you can send it to Eric at Hux and he'll put the IACV and TPS (after you provide the BBK and stock TB core) for $150 which isn't too bad at all.

ChrisD
07-08-2010, 05:50 PM
Remote IACV is another option, not too tough to do either. And I'm on standalone, so the specific TPS isn't too important although I'd rather Toyota so I can just use the stock one.

But yeah...my budget for the throttle body is <$50. So I'm going with a junkyard dog 7MGTE. :D

I might make some progress this weekend, we'll see!

donteatbugs
07-15-2010, 02:57 PM
i made a remote plate for my iacv and am running a q45 TB that i got at the junker for 30 bucks. it had a bad tps though and had to buy a new one of those for 30 more. it is overkill for the 300 wph im making but its reliable and ill never out flow it.

Funkycheeze
07-15-2010, 04:44 PM
Hey no worries joe - a lot of people probably have different goals than me.

I'll toss in my dyno graph too. Remeber, this is a Dyno Dynamics, which has been tested against local Dynojets to be 14% lower. So that places me at 265ish whp on a dynojet. I'd still like to tune out that dip I have there - we had to be really conservative during the tuning as we saw some high knock readings...anyways the graph:

http://www.celicatech.com/gallery/files/4/dynograph.jpg

14% lower? holy fuck - that means i'm making more than 360 WHP? no wonder it feels so fast!

http://www.telusplanet.net/public/rasp08/supradyno.jpg

Also, a 7M TB should be more than sufficient - you can make 400+ whp on them in a 7M, and 550+ on one bored out to 64mm like mine.

ChrisD
07-15-2010, 05:44 PM
^^For the record, Ryan's car kicks the crap out of mine! If only you had traction... :hehe:

I'm thinking about doing my next dyno runs at RCTS to get a dynojet reading. Maybe I'm wrong, but that's what I've heard. With that said, my performance at the track pointed towards my #'s being closer to the 235whp level. I was almost a full second slower than my best, and still a good half second slower than the best I've done in Calgary, at lower boost levels. So I dunno! Dynojet next time!

I've thought about going Q45, but it is really large and is probably overkill...if I can stay with the 7M one that just makes the install easier if nothing else. But if I have to, remote IACV is an option. Actually, I've been inadvertantly running without IAC for a few years now without issues, so not terribly worried about it!

Latest update: Got everything but one of the stays removed from my manifold. Need a longer wrench for more leverage. Should have the manifold out within the next couple days, then I'll start mocking up the 3SGE manifold and begin cutting.

Funkycheeze
07-15-2010, 10:00 PM
I get good traction in second and above, especially once the tires are warm

Damn AWD =P

Did you get ahold of gord from supraclub? he has a crapload of those TBs.

ChrisD
07-16-2010, 04:24 PM
I think I might have a lead on one from flateric. Fingers crossed! Although it looks like my car won't be ready for Ogopogo. Luckily I have the 205 for backup. :D