PDA

View Full Version : weird coil on plug ignition setup on a stock ECU



karl
09-26-2010, 08:26 PM
i want to know how in hell this works:
http://www.86garage.com/forums/viewtopic.php?f=24&t=1122&start=60

it's a "batch fire" COP setup. he's feeding all 4 coils the same ignition signal and firing all 4 at the same time. i have no idea how he's not lighting off the charge at the bottom of the intake stroke.

maybe it's because it's an NA engine and the charge is still too sparse/lean at that point? there's no way i would try that on a turbocharged engine, for fear of sending a rod out the side of the block.


i just think it's interesting because i finally finished the circuit that can distribute the spark from a single ignition channel. i tested it on the bench last night with no problems, and i'm going to try to test it on my car very soon. eventually i want to run 350Z coils, but i have a set of k-series coils that will work in the meantime. apparently these type of denso coil packs have seen over 900whp in various applications. i'm more interested in just getting some crap out of the engine bay, not having to deal with a distributor cap/rotor or plug wires anymore. the eventual plan is to get rid of the distributor completely and run crank/cam triggers for the honda ECU, but that will be a wintertime project. i may just go back to the stock distributor too, since it's small and i'm finally motivated enough to test the translator board.

anyway, i'm still kind of in disbelief that the above setup actually works!

pitcelica
09-28-2010, 03:48 PM
read some of this (scroll down until half the page) : http://www.msextra.com/doc/ms2extra/MS2-Extra_Ignition.htm

Sam

andy
09-28-2010, 08:09 PM
this has been going on for a while now.
check this thread for more info

http://forums.club4ag.com/zerothread?id=18900

karl
09-29-2010, 01:34 AM
jesus. that first method is ridiculously overcomplicated. mine almost fits in a tic tac box and doesn't need any extra sensors.

i'm just concerned that if someone were to try that method with a turbo car, they'd blow the motor apart. despite what they're saying in that thread about the fuel injector opening after the spark, they're wrong. at high load, the injector is open for ~80% of 2 crank revolutions, spraying fuel at the back of the closed valve. if the mixture were compressed enough on the way in, it would be ignited by that setup while still on the intake stroke.

there's a whole lot of not-so-educated guessing going on in that thread.

i did get a chance to test mine on the car yesterday. it works perfectly. i used a timing light and an RSX coil and connected the timing light to each cylinder while the RSX coil was connected to that cylinder on the COP board. sparks synced up perfectly for each cylinder. i need to get a wire harness for the CBR coils i have and wire up an igniter bank for them and then i can throw away my plug wires.

Mafix
09-30-2010, 03:01 PM
what the hell? that makes absoutely no sense what-so-ever

KoreanJoey
10-01-2010, 06:36 AM
if the mixture were compressed enough on the way in, it would be ignited by that setup while still on the intake stroke.


Um... no. Like you said, the mixture is still spraying on the back of a closed valve and the ignition is occurring in the cylinder. I don't really see your concern... EDIS = batch fire.

If you've got a leaky valve, you've got enough issues already.

joe's gt
10-01-2010, 07:06 AM
The way I understand it, karl is referring to one cylinder being on the compression stroke where the spark is suppose to fire, but with batch fire, another cylinder will be on an intake stroke with fuel inside that could possibly be pre-ignited when all four are firing at the same time.

Mafix
10-01-2010, 10:21 PM
no i mean i just can't fathom how this works. not how it's not supposed to work, i got that part. hell if it was this easy every tuner would have COP. instead of spending big bucks getting it.

karl
10-02-2010, 02:36 AM
Um... no. Like you said, the mixture is still spraying on the back of a closed valve and the ignition is occurring in the cylinder. I don't really see your concern... EDIS = batch fire.

If you've got a leaky valve, you've got enough issues already.

no, EDIS is wasted spark. compression and exhaust stroke cylinders fire together. intake and power stroke do not fire. this setup is firing every cylinder on every stroke at the same time.

i'm saying the injector is spraying at the back of the closed intake valve, continuing to spray as the valve opens and lets the mixture in at the top of the intake stroke. the intake charge has plenty of fuel in it, it's just not compressed. the problem is that there is a spark happening at the bottom (end) of the intake stroke near BDC. my only guess is that the momentum of the intake charge has most of the charge/fuel still at the bottom of the cylinder near the piston, so there isn't enough to light off up near the spark plug.

try doing that under boost and i have a feeling it will blow the engine apart.

---------- Post added at 01:36 AM ---------- Previous post was at 01:28 AM ----------


no i mean i just can't fathom how this works. not how it's not supposed to work, i got that part. hell if it was this easy every tuner would have COP. instead of spending big bucks getting it.

really all they're doing is sending an IGT signal to all 4 coils at the same time. the coils they're using have built-in igniters so they just need the logic level signal from the ECU to fire them. for 3 of the cylinders there is no problem- one spark happens in the cylinder on the compression stroke, one in the exhaust stroke, one on the power stroke.

i'm not sure if the coils themselves handle the coil dwell or not. i'm guessing not, and they're not running enough cylinder pressure to need huge spark so a weaker spark wouldn't manifest itself on their applications.

another thing that had me laughing and scared at the same time is that they were using the stock igniter to fire the stock coil to provide IGF to the ECU, but just leaving the coil secondary unconnected. a few of them melted their coils/connectors and nearly caused fires. they should just be grounding the coil secondary, or just connecting the stock igniter to the coil of a relay or something that doesn't have a secondary winding.

i definitely DO NOT recommend trying their method on a 3SGTE. like i said, i built a small logic circuit that electronically distributes the IGT signal to each coil in sequence for my application. i just need to find someone local with some xB/xA/yaris/echo/prius coils to see how well they fit the 3SGTE valve cover. the K series coils stick out way too far.

joe's gt
10-02-2010, 04:09 AM
like i said, i built a small logic circuit that electronically distributes the IGT signal to each coil in sequence for my application. i just need to find someone local with some xB/xA/yaris/echo/prius coils to see how well they fit the 3SGTE valve cover. the K series coils stick out way too far.

How do you adjust spark timing with this method, karl?

karl
10-03-2010, 07:55 PM
How do you adjust spark timing with this method, karl?

change it in the ECU's timing tables. also, the ECU still uses the distributor as a timing reference sensor, so moving the dizzy will alter base timing. all the circuit does it split the IGT into 4 channels. everything else is the same as if it were being sent to the dizzy igniter/coil setup.

joe's gt
10-04-2010, 05:59 AM
change it in the ECU's timing tables. also, the ECU still uses the distributor as a timing reference sensor, so moving the dizzy will alter base timing. all the circuit does it split the IGT into 4 channels. everything else is the same as if it were being sent to the dizzy igniter/coil setup.

Gotcha. Sounds cool.

andy
10-04-2010, 06:26 AM
i have a JDM 4age ecu and would love to run COP, whats the dizzy wheels look like in the 3s? i have 2 a 23t and a 4t wheel.

KoreanJoey
10-04-2010, 06:36 AM
Better option...

www.autosportlabs.com

karl
10-04-2010, 06:22 PM
The 3s uses 2 wheels, but 3 sensors. There's the Ne sensor which reads the 24t wheel and the G1 and G2 sensors which both read the same single tooth wheel, but 180 degrees apart from one another.

Sounds like the jdm 4age ecu is a batch fire injection ecu, since the ecu can't determine cylinder position from a 4 tooth cam sensor. You still could run cop ignition like my setup, but you'd need to add a hall sensor to sync it to engine position.



better option...

Not for me. The cop driver cost a grand total of $3 in components :) there are 3 logic ICs required. The board should fit inside my ecu case.

Luni
10-04-2010, 07:52 PM
Pics of the final result or it didnt happen!

karl
10-04-2010, 10:58 PM
getting there. probably going to etch the PCB for it tomorrow once i finalize the layout. i have some aluminum extrusions i got at a surplus store that i'd like it to fit into. tonight i'm working on the boost controller.