Intermittent Stalling

[Martin Wagner]

My car doesn't have a real TPS so the accel enrichments are MAP driven. I had problems with it nearly stalling during cold idle. I was getting a spike from the MAP that would activate the accel enrichments. There are several enrichments, one based on the rate of change in the TPS/MAP, and 2 during warmup. One of the warmup increases is a percent of the TPS/MAP enrichment. The second is a fixed pulse width. I had to remove the warmup fixed pulse and decrease the TPS/MAP enrichments to get the car to idle correctly when cold.

Marty

[Bruce Bowling]

Looking at the datalog I agree that the unit is losing sync or a pulse arrival is out of tolerance. The immediate RPM drop to zero indicates that the code detected an error in pulse arrival time and had to resync, which produces the zero value.

You indicate that the issue is the same b/w MSD input (and I assume opto input) and the VR input. So this tends to diminish the VR circuit as a root cause since it it is totally different from the opto input.

Also, when the uS dies and you are datalogging, does the second count also reset to zero, or does the seconds keep on clicking away during the event? If the seconds reset to zero then there is a processor reset. If the RPM resets to zero but the seconds keep on clicking up at the same numerical increasing rate then it is an ignition software reset.

I suspect that there is a noise source that is triggering back into the uS unit. I am sure you did this but make sure the ignition primary and secondary wires are separated far from the uS box. Same goes for the MSD, the oscillator flyback in the MSD generates a lot of EMI. A lot of this comes back into the power supply (battery) circuit. SOme have had good success (especially on dyno and race car setups) using an external capacitor across the battery, like the ones used for car stereos. These are real inexpensive and would be easy to verify, and can help with other electrical issues.

Finally, make sure the grounds are correct. You indicate noise the TPS enrichment causing triggers, again this is an indicator of external noise.

- Bruce

[DoD70]

When the motor dies, the seconds are still counting so it would seem the theory about the ignition reset is true.

In order to minimize the possibility of electronic noise, I mounted uS about 6' away from the MSD box. The MSD is mounted in such a way that the primary and secondary wiring is as short as possible and away from the rest of the electronics that are on the car. The IAT, CLT and TPS cannot avoid being near (< 12") from the spark plugs so if there is feedback along the sensor wiring from this there won't be much I can do about that. However, the unsed uS ignition input and output wires are running with the rest of th uS wiring. Could this be it?

FYI, all of the uS wiring is segregated from all other wiring and runs inside one of the frame rails of the dragster. All grounds run back to the engine block and chassis ground. And I have a 1.25 farad capacitor mounted right next to uS. Power wiring for uS comes straight off the capacitor which is wired parallel to the battery.

[Bruce Bowling]

OK - sounds like you have a pretty nice setup with the wiring. Here are a few things to try in order to help track this down:

1) The second VR input, tie this direct to Vref. This input has a pretty high voltage threshold so I do not think it is the issue, but this will eliminate this.

2) Check to see if there is a mechanical vibration issue with uS. Take the uS unit and tap it while the engine is running (not hard, just enough to introduce a little shock). See if the dropout follows any mechanically-induced noise.

3) You con sometimes narrow the issue to noise induced in the VR sensor. Get a capacitor, say 0.1uf, and hook direct across the VR input - between the VR+ and VR-. Run the engine and see if there is any difference in dropout.

4) I do not think this is the issue (since there is no processor reset) but you can temporarily run the uS off of another battery for a test. In fact you can even run it off of a 9-V battery, the injector battery correction may richen things up a little bit but this is just a short test. Unhook the +12V input going into the uS from the rest of the vehicle and attach this to the + terminal of the battery. The - battery terminal just hook to the chassis - leave all of the grounds, etc connected as-is. What this does is just power the uS internals, and since all of the drivers are low-side types they will still work (getting the current from the existing vehicle battery). Again run this way and see if there are any differences in the dropout rate.

Each of the above test isolate a particular potential issue, so try these out and see if any makes a difference. When it is narrowed down then we will know what we can do to track this down...

- Bruce

[DoD70]

Ok, here's the latest. I've tried your segguestions (thanks!) but nothing changed. I did notice that the reading from the MAP sensor will float widly even with the engine shut off. In fact, even with uS disconnected from the MAP sensor. While the engine is running, the "dancing MAP will trigger the AE flooding and stalling the engine. I have moved the slidebar on the AE all the way to the TPS side and the stalling problem vanished. Along with excellent throttle response.

Something else odd occured today. Coolant temp was around 200F when the WUE indicator on Megatune lit up. The idle dropped and the pipe started to belch black smoke. A quick look at the WUE dialog box confirmed that the WUE had decided to fatten the mixture to 120%.

Also, I'm still experiencing the problem with the rpm not spinning past 5000 rpm or so for the first minute or so after I start the engine. Warm, cold, it really doesn't matter.

On the upside, the car ran good enough at Norwalk this weekend to win the race. It was an ugly win, but who's counting?

[Bruce Bowling]

The MAP bouncing wildly - this should be pretty steady, maybe a couple of KPa while running, and locked at 100KPa or thereabouts with the engine off. This alone can cause all sorts of issues.

With the engine off but the uS powered up does it still bounce around? How much?

- Bruce

[DoD70]

The bounce that I'm seeing is in the AE area. I have seen a rate of change from -50 to 80 kpa with uS powered up and the engine shut off. Same results even with the MAP disconnected from uS. Setting the lag time down to 10 has masked the problem and at least allows the car to idle but throttle response is not as good. Actual MAP readings with the engine shut off are real close to 100 kpa and vary no more than 1 kpa. It's the rate of change that seems to be the culprit.

Any idea on the other problem? It's actually more trouble at this point than stalling.