Nissan CAS (RB25) firmware crash!

[jsmcortina]

This looks to be a serious issue.

A customer reported that his customer was having a nasty intermittent issue on an RB25 install. The engine runs great most of the time, but very intermittently it will cut out with the fuel pump and one injector stuck on.

I got hold of the MSQ yesterday and yes, there's a real problem. This is with 1.2.0 firmware. It can take a number of hours running on the stim before the issue is apparent. (I'm awaiting confirmation from the customer about whether I can share the MSQ.) Three tests took 2601 seconds, 10400 seconds, 3321 seconds before the firmware crash.

Presently I don't know if there is a specific set of conditions in the MSQ that triggers this bug or whether it is specific to 1.2.x firmware.

I've just started a test run on 1.3.2 to see if code changes in the last year have resolved it. (There were some issues with non-atomic writes that perhaps might have an influence.) With the very intermittent nature of the problem it could take a big effort to debug this, but I'm determined to do my best. It needs to be able to run for say 24 hours on the stim with no issue to be considered "ok".

Has anyone else seen this?

James

[AbatelliCristian]

you think the problem is only RB25 or also SR20?
I can make a test in SR20 next days

[jsmcortina]

Unsure. I'm early in my investigations at the moment, but can confirm that the MSQ I've been using triggers the same crash problem in firmware 1.3.2

James

[shaodome]

Just to add a datapoint.

I've done several installs now using the SR20 decoder and none have reported the above issue (latest stable release code). Some are daily drivers.

The only issue I've had with the SR20 decoder is what I'm thinking is related to dwell. Using that decoder method the spark is ALWAYS weak, regardless of the coil type used. For example, using LS2 coils, with plugs gapped close and blowing the spark out around 300-350whp. Same behavior on the Toyota 1zz coils as well as Nissan stock (crappy) coils. The solution was to swap to the DIY trigger disc and with no other change everything is fixed.

Just for what it's worth.

[jsmcortina]

I haven't fixed the bug yet (slow work) but I do have a firmware that resets in the case of the bug instead of crashing. This turns it from a "critical" bug into a "major"

I'll be road testing the modified firmware and aim to post a new beta in a number of hours.

James

[jsmcortina]

See here: http://www.msextra.com/forums/viewtopic ... 25&t=55072

James

[jsmcortina]

I'll repeat my tests with the SR20 mode later this week. That should help narrow down where the problem lies. I'll also check out the dwell, I was unaware of that.

James

[shaodome]

I'll repeat my tests with the SR20 mode later this week. That should help narrow down where the problem lies. I'll also check out the dwell, I was unaware of that.

James

I'm assuming it is dwell due to spark blow out as I don't have access to a scope to check it. The timing is solid all the way to 8000rpm as verified with an analog light on a plug wire placed between the coil and the spark. There was no sync loss reported in the logs either.

Also, ALL of the coil combos also experienced a strange missfire around 3000-4000 rpm under light load, but not while free revving the engine.

[mxsteven]

I think i have the same issue... Let me check when I get back next month.

The engine will just die just at anytime without any reason and it is impossible to restart... So i had to swap back to stock ecu to take a safe ride home, after testing at the bench test, everything was fine!! When plug back to the car it will start again....

After replacing the pull up resistor to 1k the problem solved for the moment but my friend says there is still some sudden jerk at low load sometimes. Later i will try to put 2.2k resistor and see will it improve the situation or i need to try the new firmware but the new firmware is not a fix yet? Right?

[jsmcortina]

The issue I'm reporting is unrelated to resistor values. I have tested a number of CASes on the bench using the 330R I show in the manual.

Yesterday I was able to get the crash on SR20 mode also.

Since then, the same firmware has stubbornly refused to crash in RB25 mode - I left it running for 10 hours and no crashes !

So this means that I have not been able to debug anything yet.

James

[mxsteven]

Thanks for the update.. anyway please update here if there is any!

[jsmcortina]

I found where the code was getting stuck and believe I have implemented a fix.

I'm going to be leaving the code running on the bench with a known erratic tach signal (that's what seems to trip the problem.) Hopefully it is still running ok in 12 hours.

James

[jsmcortina]

The code is running ok on the bench after 27 hours. I'll be road testing (regular mode, not RB25) tomorrow, then putting out a new beta.

James

[jsmcortina]

New beta released : http://www.msextra.com/forums/viewtopic ... 25&t=55156

All Nissan CAS or Optispark users are urged to upgrade.

James

[mxsteven]

Thanks!!

BTW the old firmware is there any issue for RB25?

New beta released : http://www.msextra.com/forums/viewtopic ... 25&t=55156

All Nissan CAS or Optispark users are urged to upgrade.

James

[jsmcortina]

Thanks!!

BTW the old firmware is there any issue for RB25?

Which old firmware?
The bug that pre-1.3.3beta2 fixes is present in all other MS3 firmwares.

James

[mxsteven]

sorry... i was refering to 1.2.4 firmware..

Thanks!!

BTW the old firmware is there any issue for RB25?[/url]

Which old firmware?
The bug that pre-1.3.3beta2 fixes is present in all other MS3 firmwares.

James

[timwds]

I used to have this issue. It would generally occur between gear shifts or decel and would "hydrolock" the engine with fuel. I gave up and converted to a toothed wheel.

EDIT: version 1.2.4 for memory.

[jsmcortina]

It is a shame that you didn't report that very serious issue so it could be fixed. I was able to put out the first fix within 24 hours of receiving an MSQ.

James

[turbostarlet]

Just some feedback.

I did a MS3+MS3X install on a RB25 that was exhibiting the same symptoms as you described.
I thought it was wiring issue until I read about the “RB25 crash” on the forum.
I did the firmware upgrade and the car run two days without problem.

The car is getting some upgrades and will not be drivable for the coming weeks. I’ll report the progress once the car is up and running again.
I think there are more cases out there. People are either giving up thinking it is a wiring issue or are enjoying their ride after firmware upgrade.

If this update helped you…please show your love and appreciation to James by writing some feedback on the forum.