PDA

View Full Version : Rebuilt engine, ~3 mpg drop with other fun



Hiko
08-14-2014, 03:41 AM
FOR FUTURE SEARCHES: It turns out the that while the coolant temp sensor was broken (took many months to throw an 'undercover' code), and a stuck front caliper wasn't helping, the most likely culprit is my new driving style of short drives of about 4 miles, engine usually cold. Driving highway I've been getting ~30mpg, so for the most part, this post can be chalked up to a form of driver error.

Okay, so I have my 1990 GT 5spd, 5sfe rebuilt with ~1400 miles on it now. Since the rebuild, I've gotten notably worse gas mileage (24 mpg city compared to 27-28 before). She also has a somewhat rough idle, somewhat rough revving no load around 2-3k rpm, and somewhat rough running ~5k rpm WOT, general lack of power, and she smells rich.

What I can say that the problem is not....
-Engine compression (145 psi across the board....Colorado remember)
-Spark plugs, wires, cap, rotor (new)
-Ignition coil (tested per BGB, in spec)
-Dirty air filter (K&N, cleaned when rebuilt engine put in)

And I can say that it is not likely a/the....
-Vacuum leak (lots of carb cleaner and strained ears to attest)
-Throttle body being dirty (more carb cleaner)
-Injectors (Replaced w/ 2g green tops. Idle jumped to 1k rpm, bit more rich, but still rough)
-Cold start injector (could still be leaking, but tested in spec per BGB)
-Fuel leak (smell none)

Additionally, I've put 2 cans of BG 44K through her, once during break in due to sitting for ~1 year, and another afterward to rev it out and get some pressure in the system.

More details that might or might not mean anything to the problem....
-Engine will actually cool under WOT. It's not much, about half a tick mark on the gauge, but definitely noticeable.
-Experienced precisely one random engine hesitation. WOT, about 3k rpm.
-Had electrical short from battery to chassis a while back. (Here (http://www.celicatech.com/forums/showthread.php?57795-Melted-Trans-to-Fender-liner-ground-strap&highlight=) )
-Engine bogs for a second when electrical load is applied, to the point of almost dying once
-Have a Surewin (ebay) header that also acts as a cat delete. As the 5sfe is tuned to post cat o2 numbers, I'd be tempted to think this is my problem, but got 27-28 mpg before with this header.

So I'm narrowing down my list to a few things (o2 sensor, coolant sensor, and fuel filter among a couple others), but I'd really like any new ideas as to what this could be from you guys, and any idea as to my best place to start. I've been at this for months, and need some fresh ideas here. Any and all input is, as always, greatly appreciated.

CollapsedNut
08-14-2014, 04:28 AM
I feel like you have a timing issue or something to do with the distributor. That's just my redneck licked finger in the air though.

Hiko
08-14-2014, 04:40 AM
I feel like you have a timing issue or something to do with the distributor. That's just my redneck licked finger in the air though.

Ah yes, thank you for reminding me! Set the ignition timing to 12 degrees BTDC, broke it in with stock 10 BTDC. Also rebuilt the distributor due to seals ~300 miles ago. But checking all that hasn't changed anything before and after, symptom-wise.

I'll double check my cam timing too, but it doesn't *seem* bad enough to be off a tooth....

Doowstados
08-14-2014, 06:49 AM
I had similar symptoms with a bad TPS. Might be worth checking.

UtahSleeper
08-14-2014, 06:57 PM
Why does 145 psi seem low on a rebuilt engine? Thats what made me jump to timing cause I though PSI would be higher once broke in.

Also, maybe thermostat since it cools at WOT.

Hiko
08-14-2014, 08:14 PM
Why does 145 psi seem low on a rebuilt engine? Thats what made me jump to timing cause I though PSI would be higher once broke in.

Also, maybe thermostat since it cools at WOT.

I should clarify that the test was done at ~5600 feet elevation and with the throttle closed (oops), so the numbers are a touch low compared to BGB spec of 176 psi minimum. But I am quite encouraged that the compression is exactly the same for all 4 cylinders.

I'd had the same thought on the T-stat, think I'll replace it with a new one just to make sure.


I had similar symptoms with a bad TPS. Might be worth checking.

Thanks for the idea! I went out and checked it this morning per the BGB, came back that it was switching a little late from idle to mid-throttle, but past that was in spec. I adjusted it and drove it, can't really tell a difference with the adjustment throttle wise.

However, I think the car enjoyed the few extra miles to adjust to the larger injectors (220cc vs stock 210cc), and the idle trimmed itself down to 850-900 rpm, where she's no longer quite as rough. With the uncleaned yellow tops, it was so bad I couldn't use the outside rearview mirrors due to the vibration at idle, now I can at least focus on things in them. That and she no longer smells quite as rich. So I think I might've been premature on calling it not the injectors, and at this point my tasks'll be...
-clean/replace yellow top injectors
-Fuel filter at the same time
-Check intake->exhaust cam timing marks
-Check cam timing at the belt
-Replace T-stat just to be on the safe side

But I am still open to ideas on what else would be worth checking out. Thanks for the help so far!

Hiko
09-06-2014, 11:13 PM
So I've been a bit busy with the start of school, but the problem returned with the green tops being installed a bit longer.

In the interim, I've checked...

-Cam timing at the belt (MAYBE off by a degree or two, but well within one tooth difference)
-Intercam timing (Marks solidly aligned)
-Replaced fuel filter (Figured why not. Didn't do anything)
-Checked cold start injector for leaks (Pulled it out of the manifold, pressurized the system, saw no leak)
-Ghetto checked EGR valve (Sucked on the top [easy there partner] and heard it click up and down. Applied engine vacuum to it and killed the engine, so I'm guessing it works)
-Pulled vacuum hose to FPR and saw no fuel leak after ~2 minutes staring.
-And I realized I haven't mentioned my tire pressure is fine. They're at ~35psi all the way around.
-Unplugged the coolant sensor and she definitely responded poorly to it, so I'm guessing it at least marginally works.
-Remembered O2 sensor isn't all that old, pulled and looked at it and seems clean (highly scientific, I know)


I'm running out of things here guys. I must be missing something either in my previous tests or something else to check. And if anything, fuel consumption has gotten even worse.

alltracman78
09-07-2014, 12:31 AM
If you're running larger injectors, that's the problem there.

Check how often the O2 signal switches [changes]. Should be 8+ times in 10 sec? I forget off the top of my head, it's in the BGB. There's a test point in the DIAG box. Also you might want to use an analog meter to test this. Or use a cheap A:F gauge, it will give you the same thing.

Do a leakdown test, if all 4 cylinders aren't broken in it won't necessarily show on a compression test.

What do you mean by "the 5sfe is tuned to post cat o2 numbers"?

It might be as simple as you clearing out the ECU memory.

Hiko
09-13-2014, 11:57 PM
Been busy again. Sorry for the silence.


It might be as simple as you clearing out the ECU memory.

Has been cleared many times with plenty of miles in between for relearning :(



If you're running larger injectors, that's the problem there.

Yeah, I know I'm not doing the problem any favors by running the slightly larger injectors (<5% difference IIRC), but this was happening before switching the injectors (did so for diagnosis).



Check how often the O2 signal switches [changes]. Should be 8+ times in 10 sec? I forget off the top of my head, it's in the BGB. There's a test point in the DIAG box. Also you might want to use an analog meter to test this. Or use a cheap A:F gauge, it will give you the same thing.

Cool. Didn't know Toyota designed that kind of test. I performed it today, and it *seems* to be pointing to a rich condition. Reason I say that at one point the reading says to be either 0 or 5 V and I'm getting ~2.3V....so there's that. But with a vacuum leak, it does read higher than 0V, so I'm guessing the O2 sensor is good and properly reading rich.



Do a leakdown test, if all 4 cylinders aren't broken in it won't necessarily show on a compression test.

What would a leak down tell me? I was under the impression that a leakdown pinpoints sources of compression loss, but I seem to have good compression. What am I missing?



What do you mean by "the 5sfe is tuned to post cat o2 numbers"?

I mean the the O2 sensor for the 5sfe sits after the catalytic converter. I'm no chemical engineer, so I'm unaware how the perceived A/F ratio changes after the cat, but I'd guess it is a least different. I've also read of other people having similar problems with cat deletes. I've just run this engine with that header on it before and suffered no mpg drop, so I don't think that's what's doing it.

Thank you for all the suggestions! I feel like I'm chasing my tail on this one, and it's nice to have fresh input, even if it does take me a while to make use of it :hehe:

alltracman78
09-14-2014, 12:44 AM
If your O2 is reading 2.3v, either the sensor is bad [unlikely with that high a reading] or you're checking something wrong.
Or you have other voltage effecting the signal.
The O2 only sends a signal from 0 - 1v. And for practical purposes its more like .2 - .8v.

That would be like your battery in your car reading 20v with the engine off. The battery can't raise it's own voltage like that.


Leakdown test is just to make sure. If you're grasping at straws you should check all of them. :)


A 90 Celica shouldn't have an O2 sensor after the cat. Except for maybe Ca emissions?
Regardless, the ECU doesn't use the after cat O2 for fueling; it's used purely to check how well the cat is working.
The O2 before the cat is used to read/adjust fuel.

Hiko
09-14-2014, 01:24 AM
If your O2 is reading 2.3v, either the sensor is bad [unlikely with that high a reading] or you're checking something wrong.
Or you have other voltage effecting the signal.
The O2 only sends a signal from 0 - 1v. And for practical purposes its more like .2 - .8v.

That would be like your battery in your car reading 20v with the engine off. The battery can't raise it's own voltage like that.

I was doing the test per the bgb. The test point is in the same port as e1 te1, so it's not directly off the sensor and can be >1V. The BGB does explicitly say 5V, so I assume it's just another signal gets modified by the o2 sensor goes through there.


A 90 Celica shouldn't have an O2 sensor after the cat. Except for maybe Ca emissions?
Regardless, the ECU doesn't use the after cat O2 for fueling; it's used purely to check how well the cat is working.
The O2 before the cat is used to read/adjust fuel.

Ya know, I went down to look at my spare manifold and you are correct. The o2 sensor definitely sits above the cat, my brain was playing tricks on me :P So it definitely isn't my header that's doing it.


Leakdown test is just to make sure. If you're grasping at straws you should check all of them. :)

Well put :hehe:
I will try to get that done relatively soon, just to make sure.

MCcelica
09-23-2014, 10:52 AM
When's the last time you replaced your evap hoses?

Luni
09-23-2014, 07:23 PM
Re-run your comp test with throttle open.

I live at 4800 feet. Not much different than your elevation, and Ive seen 180+ on 5SFEs when theyre healthy. Your compression is low. Low compression is indicative of mechanical timing off, or something wrong with your rebuild, but since its across the board, I want to go with mech timing off. Ive seen the crank pulley outer side slip away from the inner side and actually screw up the timing mark. Have you zeroed it out by finding TDC with a screwdriver or a rod, and making sure it coincides with 0 on the crank pulley?

Hiko
10-03-2014, 09:34 PM
Your guys' advice hasn't fallen on deaf ears, just still busy.


Re-run your comp test with throttle open.

176-174-175-180

Throttle open, engine nice and warm, ~5600 feet altitude. They're certainly not race engine numbers, but I'm now pretty confident compression isn't my problem.


Have you zeroed it out by finding TDC with a screwdriver or a rod, and making sure it coincides with 0 on the crank pulley?

Actually yes, didn't mention it though. Took a while to be sure I'd found TDC properly, but the mark on the pulley and plastic lined up when TDC was hit. So I'm pretty darn sure my mechanical timing is set properly.


When's the last time you replaced your evap hoses?

...never? When I replaced the fuel filter, I had to pull out the charcoal canister and got a nice HUGE blast of gas smell outta it, so I kinda went on my way. I'll check the system better here shortly and report back.

I should emphasize that the car sat for a full year with gas in the lines/completely unprepped for sitting, and the more diagnosing I do, the more I think that it's the sitting that has caused this problem, but I'm tapped out for exactly how.

alltracman78
10-04-2014, 02:14 AM
I was doing the test per the bgb. The test point is in the same port as e1 te1, so it's not directly off the sensor and can be >1V. The BGB does explicitly say 5V, so I assume it's just another signal gets modified by the o2 sensor goes through there.

The BGB says 5V for the O2 signal?
If it does it's a typo. The signal in the DIAG box is the same as the one that goes to the ECU. Which is 0-1V.
You're probably checking the wrong wire.

Hiko
10-04-2014, 03:52 AM
The BGB says 5V for the O2 signal?
If it does it's a typo. The signal in the DIAG box is the same as the one that goes to the ECU. Which is 0-1V.
You're probably checking the wrong wire.

The thing to remember is it's not saying 5V for the O2 signal, it says 5V for the test port. The test specifically calls out for the VF1 and E1 ports, neither is directly connected to the O2 sensor. The one I believe you're thinking of is OX1, which is in fact connected to the O2 sensor. I'm getting this all from pages FI-237 and page 70 from the 5sfe wo ect wiring diagrams.

Finally got out and rented a fuel pressure gauge today, figuring if I'm gonna be pulling the charcoal canister I may as well test it while I'm there, and I got some interesting results. The pressure at idle is sitting at 40psi with spec supposed to be 31-37 psi. With the FPR vacuum hose removed, it jumps to 50psi, spec is 38-44psi. I'm not sure how much I trust my O'Reilly rental unit on this one, but even if it reads high by 10%, I'm still the high side of spec.

I think I'm gonna get another gauge and double check the readings, but it the meantime, anyone have any ideas what would cause high pressure? I know the main ones (clogged lines, FPR broken), but is there anything else I should be on the lookout for?

Thanks yet again!

Edit: Think I'll edit rather than bumping to say that the second gauge read the same thing. Checked for a clog after the FPR by literally undoing the return line on it and made no difference to pressure. A new FPR is in the mail and we'll see how it goes.