Hey there guys,
Here's the deal. I have an 88' GTU I put an 87' engine in. The 87' was an auto and ran fine, took it out and swapped the flywheel, and dropped it in the 88'. I used the wiring harness from the 87 and swapped in the injector impedance box etc. It runs great until you start to drive it and put some load on the engine. The car will start to cut out, like you hit a fuel cut or something. Sometimes it will turn off, and if you sit for about 5 minutes then start it back up, you're fine for another few minutes. Even after it has idled for a while , then you rev it up it does it.
I've got fuel and spark the whole way through. It's got a new fuel filter and I checked the gas tank. Clean air filter, no loose vacuum lines. I've got the engine running on the 88' N327 ECU.

I'm stumped now and am down to 3 things I think it could be.
1. Clogged cat, builds up too much back pressure and causes the engine to cut out.
2. Dirty fuel injectors.
3. I need to put the N326 ECU back in.

Any help?
Thanks in advance.

 

Did you use the low impedence injectors from the '87 engine, or the high impedence from the '88?

Is there a particular RPM range where it just bogs down? Clogged cats will usually make you hit a wall around 4k.

And the ECU is a quick swap, but if that fixes it, it'll be because one ECU is bad. N326 and N327 are interchangeable.

 

I used the low impedance injectors from the 87' but I have the injector resistor pack since I used the 87' harness.
When you first start it, and warm it up it normally hits a wall at around 6k, then just gets worse and worse, then eventually stalls out.

I lean away from thinking it's the ECU because it runs completely fine sometimes.

 

No one else yet?

 

Hard to say exactly what it might be, but I supposed an intermittant spark problem could cause the kind of problems you're seeing. You could start by pulling any codes that there might be, then go around testing the coils, pressure sensor, AFM as per the FSM. Really anything electronic that's related to the fuel injection system.

 

This problem first surfaced last Monday. (Molotovman lives like half a mile down the road from me.)

We were driving the car. Through neighborhoods in lower gears it was good. I got on a main road and had it in third and floored it from like 20 MPH. Up around 6000 RPM the car bucked hard a few times as the engine sputtered. I eased up on the throttle.

We had decided that maybe the car had sat too long and the fuel was stale or there was water in the fuel... So we stopped at the fuel station and filled the nearly empty tank and added a bottle of fuel treatment. (Contains alcohol, plus our fuel around here has ethanol in it, so that should absorb any moisture.)

We went to get on the highway and again, floored it on the entrance ramp. It fell flat on its face at high RPM again. Eventually, it wouldn't even hold an idle or go down the exit ramp at walking speed. We pushed it to the side of the road in a safe place. I couldn't get it up on the tow truck under its own power. (I have a flatbed... It's a necessity of being an RX-7 owner. )

This was before the fuel filter... The old fuel filter came out and was clogged with rust... but that still didn't fix it.

Cats look clean, but there's an RB Cat replacement pipe on it now, so it's not cats...


Considering the ECU. Molotovman took apart the N327 ECU last night and found two burned resistors. I've gutted around twenty S4 ECUs from running cars in the past few years, both NA and turbos and have found those resistors in various stages of burning...


We examined the plugs last night... NGK BUR7EQP and BUR9EQP for leading and trailing respectively. Minor carbon fouling on only one of the leadings. Others looked good. We used a plug out of another engine to replace the carbon fouled one.


I have a Ignition voltage tester that measures Spark, Power, WasteSpark (kV) and Burntime (ms) It's a Snap-On Vantage (v 6.0 software). I hooked it up to leading ignition and we went for a drive... But I don't think my meter can keep up with the spark intervals. It drops out at high RPM. (The car does not drop out every time the meter does, but the meter does drop out every time the car does. I think.)


Also, the tach drops a little too fast when the engine dies. While we were road testing the car, even while coasting in gear, the needle was headed toward 0 RPM.
For instance... Say we had loose injector connectors or a cut wiring harness to the injectors and we were losing fuel. Even while coasting, the tach should still register the engine RPMs which wouldn't go down too much due to the momentum of the car.
This leads me to suspect that the ECU is indeed losing power and we're losing both spark and fuel injector pulse...


Does anyone know where the critical grounds for the ECU are? (Are they in the engine harness (Passenger/right side) or the power harness (Driver/left side)?

 

One of the grounds that i think you are looking for is on the Driver side shock tower. Under the coil pack. Just infront of clutch fill cap

 

^

I'll definitely check that out. We did detail the engine bay while the old engine was removed. Some moisture may have gotten there.

I also remember there being a page that had pics of the ECU connectors and the pinouts. I want to find some of the key grounds and stick an Ammeter between those and ground...

Ideally, I'd be looking for no current, but any readings would indicate a higher resistance ground path.

 

Just curious...why did you guys not just swap kegs and be done with it?

I read the engine was fine, but check compression anyways, just to be sure.

Pele, my sig has links to the FSM - the elec. section has what you need.

 

We did the engine swap... It's a lot easier to pull and drop in an engine with manifolds, etc attached, and wiring harness connected. Simply unbolt the bellhousing bolts, pull the harness from the ECU and out the body gromet, pull the engine mounts and all oil and coolant lines, then pull the engine...

When we pulled the blown one out... It was severely overheated and took out the OMP lines and possibly other items as well due to the heat. The ends near the OMP looked new. The ends to the oil injectors were all a pile of goo... Imagine what that heat would do to injector o-rings, 6 port actuators, the wiring harness, and anything else in the neighborhood.

Also, some jackass did a halfassed amateur looking throttle body mod with lots of JB-weld and a dremel. A pristine body does not deserve that kinda engine.

When we pulled the engine outta the running Automatic SE model, we inspected the harness, retaped it, replaced the coolant line under the intake, swapped the counterweight for the flywheel, put a new pilot bearing and clutch on, and made sure the rear main seal was clean... No halfassed job here.



I know of the FSM links. I have seen a larger pic, all color coded and everything. I just never saved it because I never forsaw myself working on FCs.

 

Like Pele said, We worked on it the other day.
I put a presilencer in place of the cat, and changed the ECU back to the N326. We also checked the resistance of the spark plug wires and the spark plugs. One plug was a little fouled out so we replaced it with a good used plug I had.

I'm still having the stumbling. It always initially occurs at 6,000rpm, then gets worse. I'm just about at my wits end with this thing and I don't know if Pele has given up or not yet.

I don't think it's the AFM because the AFM is the workign one from the 87' and I'm using the stock airbox and it's all mounted properly. The coil packs are the 88 ones, but they seemed unaffected and tested out okay as per the FSM.

It drives me crazy because it does not happen all the time. Could a loose or cracked vacuum line cause an intermittent problem like this?

 

You may be onto something with that vacuum leak theory....

I had a DSM that I had the damnedest problem with:

When it was cold/normal temps, everything was fine. Right to redline with no issue. After it started to warm up, if you slowly accelerated, it would be fine up to redline. If you got on it at all, it would fall on its face around halfway though the tach... It turned out that the rubber hose connecting from the air filter box (with the MAF) to the throttlebody had a tear in it, and when you revv'd the engine up, it would flex and open the hole...

Don't know if that'll help you, but it may be worth a look

 

I have an 88 na that does exactly the same thing. It has about 20k miles on a mazda factory rebuilt motor. At around 5 k rpm (+/-), it just cuts out, badly.

 

Not given up on it yet... I've just gone into "think about ****" mode...

Just going and replacing parts after parts is a good way to waste time and/or money...

Here's my theory...

When we're out driving around, the engine is hot. Somewhere there's a failed component that's only going bad when hot... Like a contact is broken and when it heats up, it expands and breaks contact...

Suggestion:
Drive it hard while the car is still cold. (Obviously not recommended on a regular basis.) If it runs well while cold. We may need to do this over a period of a few days to let everything cool down overnight.

If it does not fail when cold, the next step is to start driving it hot and cooling off individual components (Coils, CAS pickups, etc). We can turn a can of that sprayable air used to clean out computers upside down to freeze stuff.

 

Have you checked vacuum lines for cracking? Might be worth doing, I'm in the process of getting together all the silicone lines for replacing every one in my engine (3.5mm silicone vac hose is about as easy to find as a working FC sunroof in this country for some odd reason).

Certainly can't hurt to throw some water around on the vac lines and see if you get some bubbles?

 

Just a thought...you said the old fuel filter was clogged with rust? Changed it and still have the problem? I had a similar issue, rust built up in the tank, passed through the filter and at this point was very fine particles...then proceeded to clog the screens on top of my injectors (where they connect to the fuel rail). I replaced the screens, pintle caps, and o-rings, had them cleaned, removed my tank, drained all fuel out of it then blasted all the rust out with a pressure washer, then while the tank was drying, removed the fuel filter and shot brake cleaner through the line from the filter to the tank and blew compressed air to make sure all the crap was out, and same goes for line from filter to rail. Then I re-installed all of my clean parts and never had that issue again. Might be worth the time to pull the injectors and look very carefully at the screens, even try a small pick or screw driver and carefully run it along the inside of the screen to see if you get any paste like build up. Hope this helps

 

Yeah, thats definately on my mind. I popped open the gas tank though and it was clean. I'll try messing with some more stuff today, it's not raining anymore.

 

We just drove it around last night...

Swapped leading coil. Problem reduced but still existed...

A couple blind turns of the TPS screw and it quit doing it all together.
Bad TPS adjustment.

I thought we had it right before... I thought between the TPS signal to the ECU and ground, you're supposed to have 1v for closed and 4.5-5v when fully open...

 

glad you found out the problem, I was going to say did you check the tps lol
I remember when I adjusted my tps and it was buckling and back firing like crazy at 5500rpm, re adjusted it and everything was cool.

 

I wouldn't think the TPS would cause it to not idle right or not to start and work under low load properly. That threw me for a loop...

Plus, I thought the TPS was properly adjusted.

I don't have the two lightbulb guy that everyone else uses.

I was following the instructions provided in my Snap-On vantage power graphing meter. It said the procedure is the following:

1.) KOEO (Key on, Engine Off) Connect between known good ground and TPS sensor output. (Shows picture of connector, TPS output is the bottom section of the "T". The two sides of the "T" are +5v from the ECU and ECU ground.)

2.) Voltage at fully closed should be ~1 VDC.

3.) Cycle throttle slowly and smoothly to full open position. The voltage should increase as the throttle opens, to approximately +5 VDC at Full Open. There should be no dropouts or glitches, the voltage should follow the throttle position.



I guess Snap-On is wrong. I'll use the Mazda procedure from now on.

 

Fail.

It's happening again... Happens with the TPS disconnected.

New theory:

Faulty AFM.

It happens at high throttle while accelerating from low speeds (40ish MPH) and while cruising at 80-90 MPH.

Sometimes flooring it or backing off on the throttle will stop it.

 

Check ignition timing, quick and easy. Get that out the way. The car will drive perfectly fine without the TPS connected. I think you can do some resistance tests to rule out the AFM...it is very unlikely to have one go bad...kinda like the coils.

 

^

Timing is dead on... Both trailing and leading line up with a red or yellow stripe, although I forget which is which... I think it's physically impossible to get them mixed up though.

Theory:

Timing or TPS will not cause a fail to start.
Timing would be wrong throughout the entire RPM/Load range and shouldn't move about unless the CAS was loose.


AFM if the flapper door did not move smoothly could bind up and cause a fail to start.
The adjustable resistor in the AFM could be dirty and register as closed when the flapper door was indeed open in the air flow...