grgguy

I recently had a major tune up done to my 94 Z28 A4. I had a new opti with rotor ( MSD vented) put in along with new plug wires (msd) , coil ( also MSD), a new water pump ( a/c Delco) new plugs ( stock autolites ? I can find out if it helps someone answer my question) a new radiator, thermostat and all the hoses in the cooling system. Almost right off the bat the idle began to surge. It never once surged before the tune up. First it only happened when warming up then it progressivly got worse. I have changed the TPS, IAC, ICM, PCV valve, all of the vacuum lines,double checked everything. Ive even tried adjusting the timing by advancing or turning back the adjustment screw on the opti. Everything ive done has made for a better driving car. However the MPG is stiil terrible (around 12 -13 MPG) and the idle surging makes it very interesting at stoplights.... Tonight I used scan94/95 and made a short run to the store and back. The car didn't disappoint.At first she did good but then started surging. I think i should mention that i've seen the idle hold around 2000rpm while just sitting there warming up. Also when standing behind the car the exhuast smells rich or sweet. I dont know what that means. I'm not much of a mechanic but I can follow directions. So Id be open to almost any ideas. Also is there somebody on this site that can deciefer what the scan says.. I can/will put it up on my Google drive acct and make it public for anyone to access.
I may have forgot a part or two that I've changed in the last few months but i'll double check and add them if there are.
I'll put up the link momentarily.

https://drive.google.com/folderview?...mc&usp=sharing


Thanks in advance

GaryDoug

It's hard to tell from that scan because you did not spend much time at idle. Maybe do another scan idling from cold start to peak engine temp at least.

By the way, you don't need to create the Comms display log file. That's mostly for debugging the cable and connections.

Injuneer

Some things to consider:

-irrational RPM spike at 343.77-seconds.

-very slow scan rate, initially about 1 frame in 1.3 seconds. You want to record a max rate, closer to 9 frames per second.

-note record at scan time 50.23-seconds. It is full of bad data, and shows almost 6 seconds since the last data frame.

-speed to 0 MPH at 684.49-seconds. Results in 32 seconds of idling, with very little variation in idle speed - no surge. This may be masked by the low resolution scan, less than 1 record per second.

-speed to 0 MPH at 891.37-seconds. Idles for 77 seconds, and there is a moderate surge. When vehicle stops moving, IAC is way down at 5, and the PCM starts opening it up rapidly (again, masked by low resolution) to 112 counts, accompanied by a matching increase in RPM from 567 to 1047. PCM then starts to pull the IAC counts down to control the RPM, and it drops to 6, brings the idle to spec, and then starts climbing again. Idle/IAC count stabilizes until 950.61 and then burps again.

What we need to figure is what is triggering the sudden increase in the PCM idle counts, and we can't see that because of the low scan rate. Max the scan rate, and I want to look for sudden irrational spikes in the RPM log. I've seen those before, and they cause the IAC to over-react. In at least two cases, it has been the result of seeming erratic reading from an MSD Opti. In one case the Opti had just been rebuilt and returned to the owner. He had the surge problem (up to 2000 RPM). After we noticed a sudden twitch in the RPM reading, just before the IAC went open, he changed out to an AC Delco Opti and the problem went away.

Not saying its the MSD, but indicating you need to get the scan rate up so we can at least look for similarities with a previous case.

grgguy

I Thank you two very much and will redo the scan tonight. The surge has calmed down. But there are days when I have to turn the CSR off and walk away. I believe its the med opti myself. I never had this problem till I installed the MSD opti. I will post the new scan this evening .. Again thank you very much !

GaryDoug

Turn off everything except the engine section. Also don't turn on the Comms display this time. Don't have any charts active or displayed. Make sure the Wait (ms) setting is zero.

grgguy

So I did as you asked. Only engine data, no logs , no charts, and a good 20 min of just idleing and surging. However I couldn't figure out how to change the scan rate. As soon as I started the car she went into surge mode. There's a good 10-15 minutes of data on that. I don't know that the idle hit 2000 rpm. But she sure tried. I then took her on a nice little run down an old back road here in SoCal and let her run. Then when I got home I let her run parked for maybe another 10 min. She tried to start surging but calmed down almost right away. If this scan doesn't work. I'll do it again tomorrow. This time maybe you can tell me how to change the scan rate. I really tried and it's probably stupid easy but I coulnd't figure it out.

Side notes 1- I kept noticing the sync button going red for a second or 2 then going green again.
2- When I took her out I kept noticing the engine would just cut out for a half a second or so and the idle would drop 100-200 rpm while under throttle. Almost as if the fuel pump was going bad. This is also something I noticed when i first got her back. It usually only happens at freeway speeds 50-75mph.


Here's the link
https://drive.google.com/file/d/0B5B...ew?usp=sharing

Once again Thank you for your help !

Injuneer

You increased the scan rate by turning off the items Gary suggested. It's scanning at about 8 frames per second most of the time.

I'll try and go through the new log this morning. As a result of the increased scan rate, and the fact that it covers almost 42 minutes, it's a very large file.

GaryDoug

The "telltale" rpm spikes are present big time. For a short while they stop and the IAC counts return to normal. Then it starts again.

grgguy

That's the car doing that all by itself ? I looked at the file myself last night to look for anything odd..I barely caught it but I got a whole bunch of DTC's that will be on for 1 frame and then dissapear. Is this normal ?

Injuneer

Gary:

My concern is why is the idle speed/IAC so high to begin with? Target RPM is 888, but it's holding ~1700 RPM.

What are the spurious DTC's that appear for one frame? Some of them are not even valid codes. They seem to pop up when a bad piece of data appears in another column

There are both major surges, that seem to be tied to the very high RPM spike readings, and minor surges that result from a one or two frames that show slightly lower than expected RPM readings... e.g. 635 RPM indicated vs 650 target. That is the "twitch" I saw in another log with the MSD problem. Look at scan time 933.28. There is a one frame indication of 614 RPM, which seems out of place, but the PCM is already forcing the IAC upwards. Something similar at 927.27.

I just don't understand the hysteresis in the IAC command vs. the single frame of bad RPM data.

What are the blank lines in the record? Seems to be a 6-second gap in data in one instance.

grgguy

I have no idea why that happens

Injuneer

My questions were for GaryDoug.... he doesn't own an LT1, but he has assembled an LT1 PCM, an Opti, and the various inputs, so he can simulate the operation of the engine, and play "what if" with various inputs.

Electrical engineers are mad scientists.... not stable like mechanical engineers.....

GaryDoug

The idle counts start out high because there are already rpm positive spikes occurring from the start of the datalog file, and perhaps even before the datalogging starts (36 secs after scan starts) and the scan starts (11 secs after the engine starts). In my previous "study" of the effects of rpm spikes on the IAC operation, it was also shown that the target idle speed did not change even though the IAC counts did change. I don't think I reported that at the time. Just to make sure, I repeated the test today with the same results.

The rpm is derived by the pcm measuring the time between the leading (positive going) edges of the low res pulses, since those points are always in the same place regardless of which cylinder is being used. So it's very possible there are rpm spikes not captured here. With a scan time of about 130ms and a repetition rate for the low res pulse of about 20-25 ms at idle, we could easily be missing some that the pcm sees but does not report.

Blank lines are probably due to either loss of comms temporarily or bad checksums causing a restart. I may look at doing something different there to be a bit more lenient in allowing bad checksums without doing a reset immediately.

BTW, some of the other irregularities where the graphs and total scan time variable jumps around are probably due to the app skipping one or more data records when an error occurs. That leaves a record of all zeros or, more confusing yet, a record of data from the last scan. You can see that when the TOD is out of sequence. I will try to address that in the next revision.

GaryDoug

I forgot to add this. I noticed the timing variables of TOD and total scan time had reset during the scan and that would complicate things. I am pretty sure that happened because the operator clicked on the Connect/Reset button either during the scan or between scans. That is my fault for letting that button stay active during the scan, especially since they are both green. That has been corrected and will be reflected in the next release.

Injuneer

I wouldn't expect the target idle speed to change. It appears to be based solely on coolant temp.