strange and very occasional loss of throttle response

irish44j · 06-18-2019, 03:13 PM

After fixing my "repeatable" stalling issue that was due to a bad connection at the reference sensor, the car drove fine for a 300+ mile road trip. Drove perfectly, actually.

But the last three times I've taken it out driving, the following happens:
1. Car drives fine for at least 20-30 minutes. Idles fine, power good throughout the band, no hiccups, nothing. This is regardless of what type of driving I'm doing (traffic, highway, whatever).
2. At some point after 20-30 minutes I will be driving along normally and then nothing, throttle loses all input (though engine doesn't stall immediately) and the car will just slow down and may recover or may eventually stall out. Letting off throttle (to engage IACV) makes no difference. Mashing the throttle (to engage WOT switch) makes no difference.
3. When this happens, if I'm moving I can just quickly turn the key off and back on and the car starts right up, or sometimes it just recovers before it stalls.
4. Usually within a few minutes, the car will stall again at a stoplight. It fires back up immediately, with no heistation.
5. After the 2nd time doing this, it generally runs fine for as long as I drive (last time 20 minutes), totally back to normal. no hesitation, misfires, or anything else like that.

I'll also note, no warning lights at any of these times, unless/until the engine finally stops.

So, I'm having a lot of trouble figuring out these symptoms.
- Car has new fuel pump, fuel filter, good gas
- Has the solid-state DME relay
- I opened up the AFM and no significant grooves in the "track". Tried my other AFM and same symptoms.
- Throttle body cleaned
- No vacuum leaks
- New ignition switch
- Shoudln't be IACV-related since it will do it under cruising throttle.

So, any additional thoughts? I suppose it could be loose wiring or something, but can't figure out why it would follow this apparently-random sequence of events and run fine otherwise. Going over rough roads doesn't cause it (as I would expect if it was some kind of loose wiring). This always happens on really smooth roads (most roads around here are really smooth.

Jfrahm · 06-18-2019, 03:44 PM

If you keep it in gear does the tach wind down or drop dead?
Do you have a spare DME relay?

irish44j · 06-18-2019, 04:08 PM

Tach just winds down. It's like I just lifted off the throttle, but the pedal does nothing. As soon as the engine recovers (or I restart it), throttle works perfectly.

I have spare DME relays, but the solid-state relays don't have the same heat issue as the OEM ones, so can't see how that would affect things. As noted, I can restart the car the moment it stalls, and everything is fine for the remainder. It's not as if anything has time to cool down.

Plus, car starts right up perfectly and runs well otherwise even for extended periods right after this happens the 2nd time.

irish44j · 06-18-2019, 04:15 PM

I guess the next thing I can do is try my spare DME. The one in it was just recently rebuilt/tested and chipped by Russell from MaxHP before his untimely death.

thomasryan · 06-18-2019, 07:25 PM

I would try wiggling the speed and reference sensor wires while it's running. The little tab on the terminals that keep it in the plastic plug are pretty fragile as well as the plastic over time. Replacing the DME side of the plug is common.

Any chance the injector driver is overheating?

Jfrahm · 06-18-2019, 08:34 PM

The solid state DME relays may be perfect but the truth is we do not know if they stand the test of time or what their failure modes might be. Making assumptions is the top thing you do not want to do when you are asking for help to solve a problem.

At any rate it's probably not the DME relay although I suppose it could be the fuel pump side of it going into a bad state which is reset by the power-cycle. Maybe it's the fuel pump dying. We all gotta go sometime and fuel pumps will sometimes get stuck and unstick with a power cycle.
Injector driver is perhaps the most likely, that's a good suggestion.

If it were my car I'd swap the DME relay and if that does not address the problem maybe next time it stalls jump out and see if there is fuel pressure at the rail before trying to reset the problem. Maybe that and rig up a noid light I could see while driving.

thomasryan · 06-19-2019, 02:37 AM

the dash light not illuminating is the interesting part.

one would think the fuel pump electrics dropping out would cause a hiccup or lean backfire into the intake unless it was instantaneous. it would be nice to see the female spades on the fuse board side of the dme relay.

it kinda sounds like a fuel pump shut off that mr wilk developed the work around for. like the tps harness shorting and falsely triggering idle mode.

new parts aren't always good parts as far as electrons are concerned.

my 74 vega used to do the same thing but it was the ground strap at the rear of the motor. one of my bosses beetles the ground was the wrong transmission commutating with the body until it didn't.

irish44j · 06-19-2019, 02:41 PM

Quote:

Originally Posted by thomasryan

I would try wiggling the speed and reference sensor wires while it's running. The little tab on the terminals that keep it in the plastic plug are pretty fragile as well as the plastic over time. Replacing the DME side of the plug is common.

Any chance the injector driver is overheating?

This is what was causing the aforementioned stalling issue, which had different symptoms, and that's EXACTLY how I found it. The car has a new Paragon wiring kit all the way to the DME for the speed/ref sensors and I did wiggle them multiple times recently to double-check, with no effect.

irish44j · 02:48 PM

Quote:

Originally Posted by Jfrahm

The solid state DME relays may be perfect but the truth is we do not know if they stand the test of time or what their failure modes might be. Making assumptions is the top thing you do not want to do when you are asking for help to solve a problem.

At any rate it's probably not the DME relay although I suppose it could be the fuel pump side of it going into a bad state which is reset by the power-cycle. Maybe it's the fuel pump dying. We all gotta go sometime and fuel pumps will sometimes get stuck and unstick with a power cycle.
Injector driver is perhaps the most likely, that's a good suggestion.

If it were my car I'd swap the DME relay and if that does not address the problem maybe next time it stalls jump out and see if there is fuel pressure at the rail before trying to reset the problem. Maybe that and rig up a noid light I could see while driving.

The fuel pump has about 500 miles on it, and is a brand-new Bosch unit. I mean, anything can "be bad" out of the box, but if I'm buying new stuff like that and they're failing that fast, it's time to get rid of this car lol.

This isn't having the symptoms of losing fuel pressure though. When I had a bad DME relay last year and the car would stall, you'd have to crank it several times to get fuel pressure back up to running. With the current issue the car literally starts back up instantly without a "priming" turn of the key, so in my experience that would be an indicator that its holding fuel pressure at the rail.

But yeah, will try a different DME relay next time out (or just jump the pump/DME). I have 3 or 4 of them that seem to be good.

gtroth · 06-21-2019, 07:32 PM

I'd take a longer look into the AFM. My 'S would cut out intermittently when it was very cold out and car had warmed up. A little similar. My AFM looked good and measured good, but replacing it solved this problem.

irish44j · 06-23-2019, 02:39 PM

Quote:

Originally Posted by gtroth

I'd take a longer look into the AFM. My 'S would cut out intermittently when it was very cold out and car had warmed up. A little similar. My AFM looked good and measured good, but replacing it solved this problem.

Interestingly, I just swapped my stock DME back in the other day, and I've driven the car twice on 20-30 mile trips since then with no stalling. So assuming it doesn't come back, I'm guessing a loose connection or solder crack in my stock DME causing the issue.

Which is odd, since this DME was chipped and checked by Russell at MaxHP (and I can see evidence that he reflowed some solder joints on the board).

Guess I'll see if the stock DME continues to eliminate the issue, and then swap the chip into that one.

Bukowski · 06-23-2019, 05:37 PM

I admit these are recent problems of mine :

-computer problem
- if it stalls when the fans come on, wires on fan resistors shorting out on blower housing.

... does it always start from cold without problems?

Always handy to have a cigarette lighter voltage meter to check things (~$15.00?) and a Harbor Fright alternator/ battery tester (~$5.00).

irish44j · 06-25-2019, 04:01 PM

Quote:

Originally Posted by Bukowski

I admit these are recent problems of mine :

-computer problem
- if it stalls when the fans come on, wires on fan resistors shorting out on blower housing.

... does it always start from cold without problems?

Always handy to have a cigarette lighter voltage meter to check things (~$15.00?) and a Harbor Fright alternator/ battery tester (~$5.00).

yeah, cold start is generally fine. With the chipped DME, cold start is immediate on first crank. With the stock DME, takes a couple cranks, but still fine.

irish44j · 07-07-2019, 12:24 PM

Well, so much for whatever I thought I fixed.

last night I moved the car out of the garage to move a different car in. The 924 started up quick, ran fine.

This morning, decided to do a test drive. Car would crank, but no start. Tried this a half-dozen times. Even hit it with a quick spray of starting fluid. Cranking, not starting. Stopped for a minute or two, then it started right up. WTF....

Then went driving, and literally almost the exact same spot on my route as last time, the car stumbled, then stalled coming to the stopsign, This time took a few seconds to get restarted. Got going, all was fine again until I get back to my neighborhood and then stalled again going like 25mph, and took a minute or two to restart.

Something I noticed, thinking back, is that this almost always seems to happen in hilly areas. It will initially stumble coming up a steepish hill, and then usually stall within 10 seconds or so after coming over the hill and going downhill or level. This may just be conincidence since my "test route" has a number of hills, but seems odd.

I'm almost wondering (thinking back to something that happened to my e30 years ago) if on hills the catalytic converter substrate is 'sliding back' and blocking the exhaust, causing the stall. This is what happened in my 1980s e30, and it wouldn't stall the car but would ALMOST stall out. I may have to check the exhaust out, just to see.