2JZ loss of sync at 6k rpm

[ab]

Update: The issue detailed in this first post has been solved! The second issue is outlined a few posts further down.


I've got the MS3X running pretty well on my 2JZ-GTE so far but I've not been able to maintain sync over 5000 rpm. In the datalog it shows sync loss reason 17 (second trigger not found when expected) and the composite log shows the cam trigger not being detected at all at that point. I'm using the factory 12 tooth crank wheel and single tooth cam wheel. The cam trigger is wired using the factory rear cam sensor and the factory wiring which is extended with another 20cm of shielded wire to the MS. The shield from the factory wire is connected in series with the shield from the extension wire then grounded at the MS3 card only. The VR+ goes to the MS3X and the VR- is grounded next to the shield ground on the MS3. Both the pots on the MS3X card (and MS3 for the crank trigger) are turned fully counterclockwise and JP7 is not jumpered.

I'm 90% sure the sensor is wired the correct way around. I tried wiring it the other way around but could not get the trigger wizard to align the spark close enough to TDC. With a trigger setting of 0° the spark was off by about 40°, with the trigger setting at +40 it wouldn't spark at all. From what I understand the trigger setting should be 0° with these engines set correctly, mine is about -5° with it wired what I think is correctly.

It looks like the next step is to apply some noise filtering. I've seen a few different methods mentioned, using a added resistor, software settings and adjusting the pots. I'm really not sure which one to start with or what the relative merits of one over another are.

Any help would be greatly appreciated. So far I'm very impressed with the MS3X system and can't wait to take it up to 8k rpm

[ab]

I'll have to take a new datalog, the only ones I have are larger than 1mb and can't be uploaded.

In the mean time here's a screen shot of the sync loss event.

[ab]

Looks like I've got it fixed thanks to this thread. The cam sensor needs to be wired the opposite way to how the factory ECU uses it and trigger edge set to falling edge. I also changed Tooth #1 angle to 0 and set the Trigger angle offset to 5 degrees after checking it with a timing light. I had thought that swapping the wires and changing the trigger edge effectively cancelled each other out, I guess there must be more to it than that. I took it up to 6500rpm without a hint of hesitation, I'll wait for a track day before testing it to 8k rpm. I also found my boost controller is definitely not connected right, MAP spiked to 340kPa, went alright though but probably not good for longevity

[ab]

Well the plot thickens. I'm no longer getting any sync loss 17 at high rpm but there is what feels like a spark cut rev limit at around 6500rpm. Spark cut is enabled but it's set at 7500rpm. It happens most of the time but definitely not all the time, when it does happen the tacho quickly drops to 0 before bouncing back up like you'd see with spark cut. Looking at the logs there's not much indication that anythings wrong, just a minor dip in PW and some roughness in the RPM. Sometimes it will come up with Sync loss reason 11, too few teeth before second trigger, but most of the time it wont. I'll try to get a small datalog and comp log to post up tomorrow.

[ab]

I'm still getting this sync loss at around 6000rpm. I believe it is a sync loss issue as both spark and fuel are cut and the tacho briefly drops to zero however the datalog still shows accurate rpm with no drops, I have no idea what to make of this. I've read a lot of 2JZ and 1UZ threads and this seems to be a common problem. The datalog will sometimes show sync loss reason 11 (too few teeth before second trigger), but not every time, and it never causes the sync loss counter to increase. The composite trigger log shows no sync loss spikes.

What I've tried so far is:
1. Tested all 4 combinations of crank and cam trigger as rising or falling. The only setting that works well enough to get the car driveable is ignition input capture on rising and second trigger on falling.
2. Adjusted the MS3X cam trigger hysterisis pot approx 6 turns clockwise which worked for someone else but gave no change for me.
3. Advanced the intake cam timing by approx 1.5 degrees which worked for someone else. I think the theory was the cam and crank triggers were overlapping at high rpm due to timing belt stretch and moving the intake cam (which has the cam trigger) solved it. Again no change for me.
4. Ran a new well shielded cable from the MS3 to the crank VR sensor with the shield grounded at the MS end only. Also no change after doing this.
5. Changed dwell from 3.0 to 2.9ms and max spark duration from 1.5 to 1.0ms as suggested in another thread. No change.

Before swapping in the megasquirt I could rev out to 8000rpm without any hesitation so I'm fairly confident all the other hardware is fine. I really don't have any idea what to do next. Any help would be very greatly appreciated!

The datalog is too big to upload here so I hosted it on rapidshare https://rapidshare.com/files/1687469743 ... .msl?bin=1 A screenshot of the log is below

[1uz_nissan]

Adjusting the hysteresis pot is not just 6 turns. I adjusted mine while the engine was running, looking at the composite log, and you can watch it change.

Previously you had a loss of sync, reason 17. This last post shows a reason 11. Get a short log on the composite logger, capture the error, and upload that here.

[ab]

Hey, thanks for the reply 1uz_nissan, as you might of guessed you were the one I got the hysteresis pot idea from

I tried adjusting it clockwise, starting at full counter clockwise, to the point of almost stalling while watching the composite logger and didn't see any change the whole time. I then wound it back counter clockwise enough that I could get it to start again without issue which gave me approx 6 turns clockwise in total. There was no change in the way it ran or in the composite log so I'm fairly confident we had different issues as I already had a very clean cam signal to start with.

Yes the reason 17 error has been resolved by swapping the cam VR sensor wires around and changing the trigger edge. The reason 11 error is still on going. Maybe I should have started a new thread for it. The composite log for this problem is posted in my most recent post above, just above the datalog screen capture. The frustrating part of this problem is the composite log shows no error or sync loss to suggest anything is out of the ordinary. Just by looking at the composite log you'd think everything was running fine. Even the datalog looks pretty good with no rpm dips and just an occasional reason 11 spike. But driving the car you definitely notice the fuel and spark are being cut. It's all quite confusing.

[ab]

I think I've found the problem, although I made the mistake of changing two things at the same time so I can't be sure which one it was just yet. First I disabled all the rev limits and boost cut just to be sure they weren't the problem. Second I noticed the MS is only seeing around 12.9-13.1V which was low enough to cause the dwell-battery correction to increase dwell to 3.5 ms. I've now set it to give 100% dwell at 12V and it revs out to 7k rpm although I think there's still some hesitation. It's still giving reason 11 errors but now only once it gets to 7k rpm and I can still keep revving with them.

So the plan now is:
-Run a new switched power wire to the MS so it sees actual battery voltage
-Do some more pulls at 3.0 ms dwell to make sure it can rev to at least 7500 rpm
-Turn the rev limiter back on at 8000 rpm, make sure it all still works
-Change the intake cam timing back to where it should be, make sure it still works

I'm still not sure what to make of the reason 11 errors. Timing % error increases considerably to about +/- 12% when they occur so I'll need to find a solution.

[3FUN]

HEllo guys.
Did you ever found a solution to this problem?
I have the same problem.....and maybe you caould help , if you have found one .
Thanks