Log data freeze

[Rusty]

Spent the day at the dyno. We tried to use autoVE with EFI 7.5.5 V1.2 to log and tune. The car would lose power during the runs after 4500 RPM. We tried to determine why but the log seemed to lock up during the most important times. We ran several runs and had the same problem every time.

After 7+ tries, we stopped logging and it appeared that the data could be monitored. When logging - the dashboard would freeze for a while (5+ sec.) versus when monitoring it appears smooth.

Attached is an example. There is almost 10 seconds between frame 157 and 158. Anyone have a theory or tell me what I am doing wrong. I have less than 24 PIDS.

[Rusty]

I spent another day at the dyno and having to tune via trial and error to get the VE tables close to right. I really need to resolve the logging issue. Does anyone else have a problem logging? It only stops logging when the car is starting to produce strong horse power.

Is it possible that the PCM stops send packets to log when the PCM is busy? I can just monitor and it appears to get the data in the scan tool. Its just when I log data that it freezes.

For testing purposes I created a PID list that just monitors RPM, AFR, and spark. I removed other gauges from the all three dashboards just in case it required the data for the dashboard. Still freezes on the way up the RPM (around 3300 to 4200) and then resumes until after throttled down.

Is there and other information that I can read?

[joecar]

Rusty,

I pm'd tech support.

You're saying after 4500 rpm engine loses power and log freezes...?

From log 0023:

Frame 157 at 22.804s
Frame 158 at 23.014s
Delta is 0.21s

Frame 159 at 23.243s
Frame 160 at 32.717s
Delta is 9.474s

Something did happen there... looks like the PCM got very busy or got reset... (are all your power/ground connections good...?)

You have 24 pids but are using 28 channels (some pids occupy 2 channels)...
get your channel count to 24 or less [no more than 4 packets] (see bottom of PIDs tab)
by removing some PIDs (DYNAIRTMP_DMA, CYLAIR_DMA) for now.

[Rusty]

I am attaching a couple more logs that show my problem. Log 11 only has 4 channels with a delta of 6.7 seconds.
The car does not loss power. We have got it producing medium results via 60+ trial and error dyno runs.

Keep in mind that when I select monitor instead of log, I can see the data streaming into EFI scan. I am currently trying to write down spark, AFR, RPM and basic info with a pencil during runs.

I have tried all four logging methods. Is it possible that Flashscan V1.2 has problems with EFI 7.5? I have several gigs of HD space left to write logs.

Thanks

[Blacky]

If your laptop is too slow and too much data arrives for it to process effectively, it can cause a comms fault. Once the underlying drivers see the comms fault, they will attempt to reset the connection with the vehicle. That includes sending initialization and reset commands to the PCM. I don't think that would cause the PCM to mismanage the engine control but I also can't rule it out 100%.

Can you please display the EFILive "profiler" window (press Ctrl+Y to show/hide) the window and see if it shows the PC lagging. You can tell if it is lagging by the value labeled "Display". It should be at 100% if the PC is not struggling to keep up. Anything less than 90% indicates the PC is skipping display updates to try and keep up with the incoming data. Anything less than 50% is bad.

If the PC is struggling to keep up, try changing: Properties->Playback->Update display every: 4 frames.

By the way what version of Windows are you using? What is the CPU speed? How much RAM is in the laptop?

Lastly, can you use V1 in black box logging mode without causing any PCM/engine problems?

Regards
Paul

[Rusty]

Blacky - Thanks for the help.
I will go back to the dyno to find out the information from the EFILive "profiler" window.

The laptop runs XP home SP2, 1.7Ghz, 1G RAM
I usually have EFI Tune running during EFI scan. I tried to close the tuning software but still get the problem.

The car's PCM does not seem to be effected by my logging problem.

If I monitor (red button) instead of logging (yellow button), it seems to get all the data.

Is there anything else I can try in addition to black box logging mode and changing update display to 4 frames?

[joecar]

You might try BBL logging and post some logs from this.

[Blacky]

If I monitor (red button) instead of logging (yellow button), it seems to get all the data.

I presume that;'s just a typo and you meant that the red button (logging) causes the problem, while the yellow button (monitoring) does not.

The difference between the two is that one is writing data to the disk (red button) and one is not (yellow button).

Is it possible your hard drive is severely fragmented and or failing intermittently, causing disk write retries?
Maybe its worth trying the de-fragmentation option, and/or running some sort of disk check on the hard drive.

Or, try using an USB memory stick as the temporary cache for log files.
To do that:
1. plug in the USB pen drive (assume it installs itself as drive E:).
2. Run the Scan Tool and open the Properties window.
3. Select the [Folders] tab page.
4. Change the "Temporary cache" folder name to the pen drive (i.e. E:\). To do that, select the Tempory cahce entry and click on [Browse].
5. Click [Ok].

Try logging and EFILive should log to the pen drive instead of the hard drive. Once you stop logging and select a filename, the temporary cache data will be copied to the filename you select.

After testing that option, don't forget to put the "Temporary cache" folder name back to the original setting or you won't be able to log any data if your pen drive is not inserted.

Regards
Paul

[Rusty]

Blacky,
Yes, just a typo on the colors. The hard drive was defragmented before the second day of dyno testing.

I have tested a different car and EFI logging was perfect, no problems. I even tested with more than 24 channels and got expected results. I used the profiler window and the computer was at 90% the whole time. I did see the frames per second go to yellow for a short period but not like the crazy 8+ second delta from the dyno. This was just road tested not dyno. Drove the car as hard as possible.

I plan to return to the dyno to test the first car soon. Since I haven't fixed anything, I expect the same poor results. If I still have the logging problem, do you have suggestions. I feel that something is wrong with the PCM / OS combo, wiring / ground or maybe the PIDs that I have selected.

The dyno car has a PCM that is labeled GM racing (Operating System 09381344) and I have performed a full reflash to a 2002 OS 12221588 . I do not think there is anything special about the PCM and I believe that GM just programming the tune file for 31 degrees spark advance and slapped a sticker on it. After full flash, I see that the spark advance is not static at 31 degrees, it is want ever I program it to. I guess I will change the PCM if problem still happens unless you have any more advise.

[Blacky]

It won't be the PIDs you have selected, unless you have electronic throttle control and have selected some of the ETC PIDs. Those PIDs can take longer than usual to be transmitted.

I would first try a different PCM, if that does not solve the problem I would go looking for wiring problems.

Regards
Paul