Sync loss appearing at the same point in time?

[SwedCharger-67]

I'm having sync losses during heavy load over 4000 rpms. MS3X fully sequential with 1.5 FW, 36-1 on crank with VR-sensor, cam signal using Hall sensor on old distributor.
Attached you can see that the sync loss seems to come at same time...11 teeth after the "Sec Level Flag" goes low...is that the cam signal?
What do you think, any valuable hint in that observation?

Sync losses 2.png

[jsmcortina]

Please post the composite log files.

James

[SwedCharger-67]

Here is a file showing the 2 sync errors, recorded as sync error log.

Also attaching a logfile showing part of the run with with a point in time with a bad sync error. Not sure how it correlates with the sync error log. I was struggling a bit with the error logging, TS says one file contains no data, but it does and can be loaded into a spreadsheet. I need more time to figure out what's the matter with the files. Hopefully the 2 attached files can give you some clues. Originally I had a log file with 3 instances of sync errors, but it was too big for this forum and the current file is severly stripped...

Martin

[jsmcortina]

There's a false trigger on the crank.

complog-false-trig.png

James

[SwedCharger-67]

Yes, there is a "double tooth" causing the sync error.
What is the best way to fix it? I'm currently preparing to measure the R56 zero crossing voltage and possible adjust the voltage a little higher. Is that a good way to fix it or is there some better ways, like series resistor, noise filter etc?

[billr]

James, how did you open that CSV file? It came up for me (in MLV) as "no data in file".

[jsmcortina]

Just opened it??

James

[billr]

Oh well, one of those mysteries of life. I've tried it many times now, always "no joy". Using MLV MS4.1.17

[SwedCharger-67]

OK folks, thanks for your help and interest.

I adjusted the zero crossing ref voltage. The R56 was all the way CCW as the manual suggests and the voltage was 0.662 Volts. I increased the R56 two turns CW and the ref voltage is now 1.231 Volts.

Car started and run normally, but I cannot go for a test drive due to rain. Tomorrow I will do the test drive.

Martin

[SwedCharger-67]

Did a test drive with the new R56 setting described above...unfortunately no change, sync errors appear just above 4krpms.
Next is to try a 10% noise filter, then add an inline resistor, then put the tooth wheel in a lathe...

[billr]

Hmm, I figured out that the CSV file was only "no data in file" on the first page, not the whole file. One "mystery" solved.

How about posting a tooth-log?

If you really get stuck here, I suggest filling in the missing tooth area to make it a "missing gap" wheel. I have found those to be a bit more reliable, but you would need to call it a 36-2. Electromotive has been filling that gap about half-way for years now; I don't know whether it is for balancing or if they, too, have found it to be more reliable like that.

[Brian H EFI]

There seems to be an entry "MARK 000" at the start of the data which is throwing MLV off. I removed it from the file and it loads up no problem.

#Firmware: MS3 1.5.0 release 20170116 19:37GMT (c) JSM/KC ********
PriLevel,SecLevel,Trigger,Sync,ToothTime,Time
Flag,Flag,Flag,Flag,ms,ms
MARK 000
1.0,0.0,0.0,1.0,0.409,0.409
1.0,0.0,0.0,1.0,0.41,0.819
1.0,1.0,1.0,1.0,0.019,0.838

#Firmware: MS3 1.5.0 release 20170116 19:37GMT (c) JSM/KC ********
PriLevel,SecLevel,Trigger,Sync,ToothTime,Time
Flag,Flag,Flag,Flag,ms,ms
1.0,0.0,0.0,1.0,0.409,0.409
1.0,0.0,0.0,1.0,0.41,0.819
1.0,1.0,1.0,1.0,0.019,0.838

[SwedCharger-67]

Here is the latest Sync Error Log showing the MS3X with potentiometer R56 turned CW 2 turns, voltage reset from 0.7V to 1.2V.

Seems that both TS and MLV HD are struggling to show some really usable info, or I just don't know how to use them. Too me it looks like I have to load the sync error log into a spread sheet and manually analyze the recorded data.

For next test drive I'll abandon the sync error logging, and just prior to the test plug in the memory card and record a composite logger, which seems more usable than the sync error logger. Immediately efter the test I will unplug the memory card, to keep the data size as small as possible.

EDIT: Ok, I maybe got it mixed up...composite/sync error etc loggers are logged into the laptop, and the drive data itself on the memory card...

[jsmcortina]

The sync error should be what you need - it stops recording when you got a sync error. Otherwise you may not capture a random intermittent problem. However, the data you just posted doesn't make a lot of sense! (In part due to how MLV renders it.) The one you posted before was easy to see the false tooth.

James

[SwedCharger-67]

Yes James, I agree regrading the files. Don't know why the latest log looks the way it does...
However, the spread sheet seems to show the false tooth again, but I need more time counting the teeth to see if there is any pattern/clues to why it happens...

I have however just ordered a Hall-effect sensor to replace the VR sensor at the crank...might do some difference...?

[SwedCharger-67]

Ok, the sync error problem seems to be solved. Today's test drive shows the engine pulling happily past 5k rpms like it has never done before.

The simple solution was to take my time in setting up the noise rejection filter on the crank sensor. I've tried it before but must have done something wrong because at that time it didn't work well.

Now the engine works so well that I even discovered that rev limiter was set too low!!