Synch loss/ Random Misfire

[PatrickLewis]

Hey everybody, I am having some synch loss issues and i was wondering if somebody could point me in the right direction. I have a 1987 Corolla with a 20v 4age, I am running the threaded hall sensor from DIYautotune, as well as the 36-1 Trigger wheel. The car randomly misfires at any rpm, here is a screenshot of the synch error logger and a datalog:

synch.csv

SynchErrorPatrickLewis.png

I have inspected the wheel and cleaned up the edges with a fine file. The wiring is all Shielded wire from the sensor to the megasquirt and all connections are soldered. I pulled the harness and inspected it. I have attempted adjusting the pots, nothing seemed to change there so i returned them to the recommended settings in the manual. Just curious if anybody has dealt with this before, any help is appreciated. Thanks.

[Matt Cramer]

Looks like a noise pulse in between teeth. What sensor gap are you running, and is this the tabbed or threaded Hall sensor?

[PatrickLewis]

Threaded, I have tried adjusting the gap from as close as possible without it hitting, all the way to where it wouldn't run. No change in the miss.

[PatrickLewis]

I should also mention that the "extra" tooth always appears in the same spot in relation to the missing tooth. It is always between the 7th and 8th "real" teeth.

[Matt Cramer]

Do you have a tooth log? I find these a bit easier to read for missing tooth wheels.

[PatrickLewis]

I will get one tomorrow, thanks for taking a look at this for me.

[turbostarlet]

I just want to tell my experience with these two sensors.

Engine:Toyota 4A-GE 16V
Trigger wheel:4" 36-1 modded to 18-1 mounted on the exhaust cam pulley.
ECU: MS3

I first used the threaded hall sensor. The engine started perfectly.
But as soon as the engine warmed up, it started to miss fire/dropping rpm.
So I had a feeling the higher operating temperature might be effecting the sensor.
I then compared the operating range of the two types of sensors.

Threaded hall sensor:temperature range -40 to +85 degrees C
Tabbed hall sensor:temperature range -40 to +125 degrees C

Then I ordered a tabbed hall sensor, modified the sensor bracket and problem solved.

My question to you(PatrickLewis): do you have sync loss/random misfire as soon as you start the engine cold?
Or does exhibit the same symptom as in my case?(with warmed up engine)

If does losses sync only at higher temperature, than you might have to switch to a tabbed sensor.

But this was my experience.....so send your tooth log so we can have a look before swapping sensor.

[PatrickLewis]

Ok here is the tooth log:

PatrickLewisToothLogger.csv

And a screenshot of the misfire event:

PatrickLewisToothScrnShot.png

To answer your question turbostarlet, my car misses cold or hot, it makes no difference. Thanks for the input

[Matt Cramer]

You might be able to filter this out with a 50% tach period rejection in the noise filter menu - let me know if that works.

[PatrickLewis]

thanks ill try it tomorrow

[PatrickLewis]

Well I tried changing the noise filtering setting, no change. Guess ill keep looking. I am very suspicious of the trigger wheel, becuase the extra tooth always appears in the same place in relation to the missing tooth.

[billr]

Can you post some pictures of the wheel and sensor? What is the smallest air gap you can use between the sensor an wheel? I'm wondering about how well the sensor is "centered" over the wheel, how much run-out (both axially and radially) there is on the wheel, how far the wheel teeth are from something else ferrous (like crank pulley). The wheel teeth do not need to be perfect; the tips are usually machined to allow minimum air gap, but the rest is often a bit rough from the cutting process. (Talking in general, haven't seen a DIYautotune wheel in-person)

[PatrickLewis]

I gave up on the 36-1 wheel and am now using my factory distributor. Nippondenso 24/1 dual wheel. Here's what ive got so far.

LewisNippondenso24T.png

LewisNippondenso24T_2.png

LewisNippondensoTune.msq

the pattern for the cam tooth is clearly not right, I've messed with the pots a bit, but nothing has changed so far. Anybody had this problem before?

[PatrickLewis]

Ok, the problem i was having in the previous post was just a matter of changing settings in the ignition options menu. turns out i needed both triggers on the "falling edge" to get it to run. However i am still having synch loss issues, sometimes synch error logger doesn't even catch them (which i find weird), but my Synch Loss gauge on my dashboard shows it when the car misfires. When synch error logger does catch it, i get an image of what appears to be a normal pattern. I've counted the teeth, made sure tooth 1 on the 24 tooth wheel is always after the cam tooth, and looked for noise pulses. This is the only weird looking screenshot I've gotten so far, it may be nothing.

weirdcam.png

This is what I normally see from synch error logger:

Nippondensosyncherror.png

If there is anything i am leaving out let me know and i will get it asap. Thanks for all the help so far.

[PatrickLewis]

http://s237.photobucket.com/user/patric ... 4.mp4.html

I scoped the inputs at the db37 connectors (both mainboard and 3x card), this shows the cam and crank signals coming in clear while the ms logs synch errors and misfires, even stalling. I have concluded the problem to be in the megasquirt itself as i eliminated everything else by going directly to the connectors on the unit. I have also spent countless hours over the last nearly 3 years trying to diagnose a problem i cannot fix regardless of how much i work on the megasquirt or adjust its potentiometers. What should i do? I have also changed my sensor setup and re run wires at least twice per component, with no fix. I feel as though i have recieved a defective megasquirt unit, this is the second mainboard i purchased to get this thing working, it is a 3.57 board with 3x card and ms3. All components were preassembled and the only mod on the board is a jumper wire to power the mid current outputs. Any suggestions as to what i should do now?

[Matt Cramer]

The snippets of log you posted didn't capture the sync errors - can you post a complete composite log with the sync errors?

[PatrickLewis]

5_14_2014.csv

Here you go, this logger caught 3 synch errors according to the "Lost synch counter" gauge, which were each accompanied by a nasty misfire. I used the lost synch gauge because I can't use both synch error logger and composite logger.

[Matt Cramer]

Looks like you've got a case of this problem.

[PatrickLewis]

Went throught that already.

Ok, the problem i was having in the previous post was just a matter of changing settings in the ignition options menu. turns out i needed both triggers on the "falling edge" to get it to run.

[PatrickLewis]

Ok I took some logs while experimenting with the settings.

This is with both triggers on the falling edge (the way I have been running it):

bothfalling.png

No synch loss in this log, but it does occasionally lose synch with these settings.

This is crank rising edge and cam falling edge:

crankrisingcamfallinglog.png

This is the only other configuration that works. The engine synchs and runs, but still loses synch.

Both rising, no synch:

bothrisinglog.png