Sgt.Stinkfist

iTrader: (6)

ive been trying to chase down spontanious misfire episodes and i think ive found it, now i would like to know what its cause could be. the misfire (dead mis) sometimes happens at idle and is always happening at light throttle (5-15% throttle) other than that 20-100% throttle accel is fine. i hooked up my scope to the injector wires at the ECU plug and at idle, when not misfiring, both primary injectors are operating at ~3.6ms pulsewidth. as i slowly press down on the gas pedal to induce the misfire, the rear primary injector pulsewidth climb smooth and steady, while the front primary cuts out completely. if is stab the throttle it comes back online and is operating on par with the rear until RPMs drop back down itll occasionally cut out again.


im pretty confident that the wiring for the injectors is ok. they all have 12v at the plugs and i personally installed, wired, tested and inspected the harness (S4 swap into a FB chassis)

what would.....or could cause the 1 injector driver to drop out under these circumstances asides from a bad ECU? (im already scouting for another ecu, but in the mean time i can still test other ECU sensory inputs for faults)

gxl90rx7

iTrader: (1)

i think under certain throttle inputs, it will cut fuel to rear rotor. like under 5% throttle or something?

Sgt.Stinkfist

iTrader: (6)

I've read a few things here and there about fuel cut under decel, which both injectors do when throttle is shut and I let it idle down or engine break down. But its my front injector that's acting screwy, the rear seems to be doin what its supposed to do

j9fd3s

iTrader: (3)

odd that its the FRONT injector that cuts out, there's no condition where its SUPPOSED to do that, could you maybe have front and rear swapped?

there are actually 2 decel modes, there is one with fuel cut on BOTH rotors, and then there is one that is fuel cut on just the rear rotor. the FC literature isn't explicit, but the GSL-SE makes the choice by the TPS. with the 1 rotor operation the throttle can be open a little.

i'm not sure but if the injector CAN fire, that would mean the driver is good, and its likely to be an INPUT to the ecu?

Sgt.Stinkfist

iTrader: (6)

im pretty sure the injectors are where the belong. the injector wire thats cutting out is the light green wire (3E) at the ecu, which is the front primary injector according to my wiring diagrams. im testing the circuit backprobing directly at the ecu.

before the problem started it would idle at 800rpm with no hesitation and cruise relatively smooth, then a week or 2 ago its developed the mis at light throttle or if i held an rpm in neutral. through out the week the throttle range where it happened started to grow a little more, and then it started to do it at idle too. ide rev it quickly and it would smooth out, but start mising again as soon as it came back down to idle.

the fact that it got worse on its own without me adjusting anything lead me to belive the problem lies in a failing electrical part. as far as im able to test, the TPS, boost sensor, ATP, MAF, and the coils are all withing spec for resistance values, all neccesary sensors have 5v from ecu and KOEO and KOER sensor output voltages are withing spec.

im stumped...

satch

iTrader: (1)

Have you tried starting the car and then disconnect the TPS to see if the problem persists.

j9fd3s

iTrader: (3)

yeah actually in this case, unplugging the tps is a good idea! plan B might be a spare ECU

HAILERS2

*********front primary cuts out completely*********

Your looking at pulse width with the meter doing this. Have you ever just flipped over to dcvolts and looked at it again? If the voltage is there but no ECU pulsing a gnd to the light green wire.........I guess it's like you suspect, a ECU driver problem.

And or swap the two primary wires AT the ECU and see if the problem travels or not. Can't hurt and takes but moments to do.

Sgt.Stinkfist

iTrader: (6)

Ill try unplugging the tps and confirming voltage at the ecu's injector wire during the drop tomorrow. I hope the tps isn't bad because its the only one I have that ohms out good and has a smooth sweep with no dropouts. The only other thing I was leaning towards, asides from a fubar'd ECU was maybe there is a frequency or duty cycle based failure of the injectors internal coil winding causing a break in the circuit....i know its a longshot, but u never know... They are all fresh remans and electronics, esp reman'd ones, can fail in odd ways

satch

iTrader: (1)

Not sure if you know this, but (in case you don't) the Light Green wire should have battery voltage w/key to on and then if there is a pulse present the voltage will drop to 8 volts or so w/the engine running.

Sgt.Stinkfist

iTrader: (6)

thx for the tip, i was just gonna compare the 2 primary injectors wires in how the behave interms of voltage. get a baseline idea of how it looks on a known good circuit and compare to the one in question. i will give an update tomorrow based on what i find. any other suggestions are welcome. i want to get my car up and running in a safer (for the engine) manner. i need to put on another 300 miles before i can consider break in complete

Sgt.Stinkfist

iTrader: (6)

ok, so i did what you guys suggested and here are the results.

i unplugged the TPS while the engine was running, and it made the car run craptacular. the idle would creep up and down and ran rough, so much so that i couldnt run it under the same conditions to induce the misfire. measuring injector pulsewidth it would cycle between 2.6-4.2ms regardless of rpm and throttle. the injector didnt cut out, but i also couldnt make it hold an rpm or maintain a constant pulsewidth.

after that i checked the voltage at the the ecu plug during operation. for a baseline the rear primary, while the engine was idling, would measure a peak of 13.5volts on my scope (actually showed the rapid cyle of 8-13.5 volt during injector operation). on the front primary injector, the wire would show the same operation and voltage, but when the injector stops working, shows a solid flatline of ~13.5-14v, indicating that the injector circuit is still getting full battery voltage through the injector and wiring up to the ecu and the ecu is failing to trigger the injector

i didnt swap the primary injectors wires at the ecu to see if the failure would follow, will the engine still run/idle normal with them switched? if so, i can try that tomorrow


im leaning heavily towards a bad ecu, so im goin to keep searching for known good N326 or N327 ecu



on a sidenote, what will cause my stock ecu to seem to have an incorrect ignition timing split? with leading on the yellow 5ATDC mark, the T1 flashes half way between the 2 marks

satch

iTrader: (1)

As far as disconnecting the TPS did you do this after the engine completely warmed up or not as it might make a difference with regards to how it runs and then again, it might not make a difference at all.

It sure would be nice if you could measure both primary injectors simultaneously to see if they both react the same or if it is just the one cutting out while the other behaves properly.

Have you made sure the problematic injector has its Light Green wire snuggly inserted fully into the small ECU plug such that it is not pulled back creating an intermittent reading? The battery voltage on the wire does not eminate from the ECU itself but the grounding affect does.

I don't know about the other issue regarding the timing gun other than it might be a function of using a timing gun on a rotary engine perhaps.

Sgt.Stinkfist

iTrader: (6)

i did unplug the tps when it was warmed up, but it shouldnt make a difference cause all the cold start and thermowax bits are gone as well as all the emissions.

when the mis starts itll continue until i rev it up, so i can/was going back and forth between the 2 primary injector wires and comparing dc voltages and pulsewidth while the injector cut out. i know that the injector driver grounds the injectors wire closing the circuit, the voltage readings i got indicates that the circuit is without breaks from battery positive, through the injector wiring and up to atleast the back side of the #3 plug on the ECU. one of the earlier things i did when the mis had developed 2 weeks ago were to go though and check wires for intermittant/bad connections by the simple push/pull/wiggle tests and found no change in misfire or loose connections at the ecu

Sgt.Stinkfist

iTrader: (6)

final update. i had a buddy grab some ecu's and the 1 un-f'ed w/ throttle body from a scrap yard around here. initially I swapped ECU and same thing was happening. following hours upon hours of searching this forum using specific keywords, along with searching 2 other forums, i think i found the issue.

a post i found from loooooong time ago...2002 :O had the only explaination ive found of the deceleration control on the factory ECU. apparently there are 2 stages of deceleration fuel cut. at higher rpm, sudden decrease of the throttle will cause BOTH rotors to cut fuel. BUT as the throttle is made less closed, or if you start to re-open the throttle, the FRONT ROTOR ONLY will cut fuel, atleast on NA's, before both injectors come back online. essentially my throttle stops, lack of the dash pot and TPS setting combined to make the ECU think that i was just slightly tipping back in throttle at low RPM causing it to go into the second stage of decel fuel cut killing the front injector signal.

the throttle body i got was 100% factory set, primary and secondary throttle set screws both still has the epoxy "lock" on them as well as all the cold start bits were still in place. i proceeded to Mod/port the the throttle body very carefully, and asides from the thermowax removal and smoothing the transitions around the secondary throttles blades left the rest of the throttle body as is. i installed the "new" TB, and low and behold, the front rotor fuel cut is gone now

only problem i have to deal with now is a new supposed vacuum leak (cheap *** rubber vac caps dry rot and crack within 1 month of use , gonna measure tonight and order some silicon ones from mcmaster car tonight), as my engine will hold a solid idle, but its at ~600rpm.... and the idle speed set screw on th TB is maxed out :O normally this wouldnt bother me, but my alternator doesnt charge till ~900RPM, so idling will in theory eventually kill my battery if left for long enough

arghx

iTrader: (3)

this is why we have electronic throttles now

Sgt.Stinkfist

iTrader: (6)

i still hate electronic throttles, the lag in resonse to the throttle input annoys the crap outta me, esp with manual transmissions...its my only complaint about my new mazda6. if the rx7's EFI was even more archaic and lacked a secondary deceleration mode then i would have had no problems with the factory ecu programing. but the enviromentalist and penny pinchers made it neccesary to force the programming into place. i think my next step, after i fix the vac leaks, in the following months will be an RTEK and i can do away with all my annoyances and smooth out the fuel curve. i want to jump passed the SAFC and get into a more useable control like the RTEK. i think that with my build it would be more benneficial to use a moderately programmable ECU vs. a completely stock ECU