strange issue with 78 280z running - I'm at the end of my rope!

[sscanf]

Hmmm... I have some observations to make here based on the work I have done on my 1976Z.

1. When I bought the car, most of the EFI parts had been removed (it was a jigsaw puzzle). While it was apart, I replaced the injectors with FJ707T (Standard Brand).
   
2. After putting it back together it ran rough and rich. Had periodic cut outs, and the tach was jumpy.
   
3. Replaced already upgraded ZX distributor with a rebuilt from Rock Auto which includes the ignition module.
   
4. This fixed the tach and the cut out problem. Ran pretty good but still had issues.
   
5. Replaced the AFM with a rebuilt from MSA and fixed vacuum leak BDCC (or is it BCDD). This made it run great except when cold.
   
6. Replaced the Aux Air thing with a ball valve from Home Depot. When its cold, I pop the hood and open it half way before I start and let it run for a few minutes that way, then I close it and go.
   
7. I have been running it daily for a while now (maybe a month). Getting 19MPG which I am disappointed in but I do lots of short trips. Work is only 8 miles from home. It does still seem a little rich to me based on exhaust smell.
   
8. I suspected a problem with the TPS so .... I dis-connected it (yes, I will put a meter on it soon). It runs just fine without it. I want to see what the gas mileage is with it disconnected. My theory is that if it was stuck on rich, then it should now be stuck on normal. Its main purpose, as I understand it, is to richen the mixture a bit when cold and when full throttle (and shut off fuel when coasting). So, if you suspect the TPS, you can disconnect it but, based on my experience, I kind of doubt it would cause some of these problems unless there was something else going on as well.
   
9. I have a home depot fuel pressure gauge after the fuel filter and its right on specs.
   

Dan

[Zed Head]

Your theory looks right. When disconnected, no idle enrichment and no full throttle enrichment. Mileage would only get better if you spent a lot of time at full throttle, I assume.

When the TPS gets wet, the idle circuit shorts, apparently, then when RPM go above 3200, the ECU cuts fuel because it thinks you've just shut throttle. Obviously, the algorithm doesn't consider that 3200 RPM could be reached with the throttle closed.

[FastWoman]

Here's a thought: If the ECU has so high an input impedance on the TPS circuit that a bit of water between the contacts can cause trip the logic, then it might also be sensitive to a buildup of electrolytes over the years, perhaps from exposure to road salt. (I note that you're in west Ohio.) Considering this, I would recommend you remove the TPS, scrub it down with gasoline. Let dry. Then wash with detergent and water. Finally, soak the switch overnight in distilled water (because electrolytes might be sandwiched between parts), do a final rinse in distilled water, and let dry. Re-lube. Reinstall.

Edited May 22, 2014 by FastWoman

[Zed Head]

Fastwoman - I did have the TSP apart and all looked well. My son and I check for continuity at idle and at full throttle at the 35 pin connector. I don't remember the exact pins, but the continuity checked OK at both positions.

Expanding on FastWoman's point - you're not reporting "no continuity" or open circuit at 4 degrees + off idle. You should see the idle circuit go open when the throttle blade is moved (looks like Pins 2 and 18). You might need a good meter to be sure that there is not some slight conductivity that would "trip the logic" as she suggests.

In the same vein, I assume that the ECU TPS idle circuit monitoring itself could be bad. But I've never seen anyone report that.

Still, the observations at 2800 and 3200 RPM should tell if any of this is even worth following up on.

[gheiser70]

Weekend update:

So new injectors from rock auto were installed with new O-rings. Slight miss at idle is now gone. Screw driver to ear test confirms all injectors "clicking" now.

TPS saga: Zed - I reconfirmed that pin 2 at 35 pin shows continuity at idle and open when accelerator is depressed all the way to full throttle. Also, pin 3 is open at idle and all the way until the last couple of degrees till full throttle is reached then it shows continuity. I cleaned the contacts with 1500 grit sandpaper and spray contact cleaner and repeated the above tests. They are the same at the TSP and 35 pin connector.

The fuel cut seems to be consistent at about 3K rpm, only when at full throttle - ie. it will rev past 3K and accelerate well unless the TPS sends the full throttle signal. As soon as I release the throttle, normal running resumes.

As a further test, I disconnected the TPS wiring from the TPS, and there is no cut out in this mode of operation - ie. I can run at WOT with no cut out of the engine. I think these tests isolate the issue to the ECU. It seems to be sending the fuel cut signal under WOT operation.

I was wondering if anyone has a schematic for the ECU. My son and I were trying to trace the circuitry for the TPS, but it seems nearly impossible without a diagram.

There is also a test in the FSM (section 8.3) for throttle valve insulation. The reading between all three pins on the TSP and body ground should be open. My TPS passes this test as well.

Has any one purchased a rebuilt ECU from MSA? And if so, how was it? If I can't find the schematic for the ECU that may be my next step.

[Zed Head]

TPS saga: Zed - I reconfirmed that pin 2 at 35 pin shows continuity at idle and open when accelerator is depressed all the way to full throttle.

The fuel cut seems to be consistent at about 3K rpm, only when at full throttle - ie. it will rev past 3K and accelerate well unless the TPS sends the full throttle signal. As soon as I release the throttle, normal running resumes.

As a further test, I disconnected the TPS wiring from the TPS, and there is no cut out in this mode of operation - ie. I can run at WOT with no cut out of the engine.

.

I guess I'll keep harping on the TPS. This is not right - open when accelerator is depressed all the way to full throttle

I don't know if you can see the picture I attached in my last post, or if you haven't seen that page in the FSM. Pin 2 should show continuity at idle then OPEN when the throttle is just off-idle - 4 degrees is barely cracked open. Basically, when you crack the throttle, Pin 2 shows NO MORE IDLE. What you're writing is that the ECU sees IDLE all the way up to full throttle. That's what I've been proposing. Plus the fact that you don't see the cut with the TPS unplugged supports that it's the IDLE circuit continuity (Pin 2) causing the fuel cut. There appears to be something wrong with the TPS signal to the ECU, not the ECU itself.

[240260280z]

Seems like the TVS wiring in the harness is reversed? The WOT pin on the TVS seems to be sending the Idle signal to the ECU by the behaviour?

[gheiser70]

Zed,

I guess I'm not saying it correctly. Pin 2 shows continuity at idle. Then, when the throttle is depressed, it is open (no contiguity) all the way through the rest of the stroke of the throttle, including wide open. At all positions other than WOT pin 3 shows No continuity. At WOT pin 3 shows continuity. This is correct per the FSM.

Blue, I also thought the wires might have been reversed, but they are correct at the 35 pin connector, so I don't think that is the case.

[Zed Head]

I guess I read it the way I thought showed that the TPS would be the culprit. Sorry about that, looking back I did read it wrong. Bummer then, problem not solved.

It's odd though, that disconnecting the TPS removes the problem. That doesn't fit with the ECU being bad. Going back to what FastWoman suggested, when you say open, do you mean infinite resistance? Do you have a high quality meter that shows infiinty, or are you using a test probe with a beep? It might be that there's enough conductivity to trip the ECU's idle circuit, bit not enough to show on your probe.

[gheiser70]

Zed,

I'm using a middle of the road multimeter to do my checks. It's no Fluke, but no harbor freight either. It shows infinite resistance. Dan's post above (# 25) shows a similar result with the TPS disconnected. I surmise that the disconnecting the TPS removes any signal from entering the ECU's fuel cut circuit. I'd love to get a schematic for the ECU, so I could see how the signal is processed.

[Zed Head]

Send a message to superlen, he's been deep in to the ECU functions. He's building a standalone engine management system to replace the Z EFI.

And the ECU is really easy to take apart. Maybe you'll find a shorted or corroded circuit. I've seen some crusty ECU internals, they live in a spot that gets moist if there's a windshield leak.

[240260280z]

If the RPMS are over ~2800 and the idle contact touches the center contact then the ECU cuts fuel (this is normal behavior).

From the description in post #29 paragraph 3 seems to indicate the fuel cuts off over 2800 rpm when the WOT contact connects to the center contact thus my supposition of reversed wiring. The fact that it does not happen with the TVS disconnected further points to the source.

Comrades... a reference with chocolate hugs:

http://www.atlanticz.ca/zclub/techtips/tps/index.html