Page 2 of 3

Re: Low tooth count wheels - beta testers required

Posted: Mon Jul 04, 2011 10:10 am
by juansh2385
Can I test the code with a 420a ? I am having trouble with a miss fire and i have run out or ideas maybe this can help?

Re: Low tooth count wheels - beta testers required

Posted: Mon Jul 04, 2011 10:17 am
by jsmcortina
juansh2385 wrote:Can I test the code with a 420a ? I am having trouble with a miss fire and i have run out or ideas maybe this can help?
420a should be unchanged.

James

Re: Low tooth count wheels - beta testers required

Posted: Mon Jul 04, 2011 4:09 pm
by richyvrlimited
jsmcortina wrote:That misfire is exactly the kind of symptoms you might get.

James
Thats exactly the same symptom MX5/Miata owners suffer from....

Re: Low tooth count wheels - beta testers required

Posted: Tue Jul 05, 2011 2:55 am
by jsmcortina
I tested the existing code on the bench with that wheel mode and wasn't able to observe the symptoms I've addressed in this code change.

You are welcome to try the code if you wish though.

James

Re: Low tooth count wheels - beta testers required

Posted: Tue Jul 05, 2011 2:57 am
by jsmcortina
Attached is a technical explanation of what situation the code change addresses.

James

Re: Low tooth count wheels - beta testers required

Posted: Tue Jul 05, 2011 8:50 am
by piledriver
Loaded it up this AM after work, couldn't get the Linux loader to work for some reason (perhaps malpractice on my part), loaded fine in WinXP.
(loading up from 3.1 pre-rc, old msq showed 303Y sig)

Started fine with the last saved msq, but after about 30 seconds under load it started to have a weird cyclic (on/off ignition?) issue, cycled between target AFR and dead lean every few seconds when attempting to drive, accel cured breifly, cycling pwr reset for another ~30 seconds, ran fine... for ~30 seconds.
have burst datalog but laptop and I are both dead.
Tried turning everything off EAE/overrun/rev limit, no dice. (tried a herd of different MSQs, all same issue, jacked with trigger filters etc)

After looking at the logs, Looks like the injector PW was going to zero every ~250 ms.

Short: Loaded 3.1.1 loading same 3.1-pre msq, zero issues. (drove it around the block, at least, was ~undrivable with the beta)

Need to reload the beta from scratch tomorrow afternoon when I'm 100% with it and see if it was just user error somehow.
ms2-3.2.1-beta-stutter.tar.bz2.msq

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 07, 2011 8:16 pm
by piledriver
After a good day of sleep, loaded the beta code, loaded the MSQ that tran fine in 3.1.1... I did it all over from scratch but only on XP.

Gave exactly the same results, with the fuel PW cutouts, idles fine.

Back to 3.1.1 again, same msq loaded, still runs great.

Unless something got garbled in the email, it's broken on my setup, and I jacked hard with it trying to get it to work, turned off overrun, EAE, second baro, set everything i could to defaults. Even tried loading the supplied msq, tweaking only the required settings to get running.
No dice.

Re: Low tooth count wheels - beta testers required

Posted: Sat Jul 09, 2011 9:01 am
by jsmcortina
I've replicated your problem on the bench. It is due to EGO. We'll have a look soon and will get you a new test code.

James

Re: Low tooth count wheels - beta testers required

Posted: Sat Jul 09, 2011 9:38 am
by piledriver
jsmcortina wrote:I've replicated your problem on the bench. It is due to EGO. We'll have a look soon and will get you a new test code.

James
Groovy, it's good to know I'm not crazy :D
Looking forwards to testing the new code.

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 3:38 pm
by jsmcortina
Greg G wrote:Ken said the code he sent me with the PID changes had this code as well. FWIW- my car wouldn't start when I flashed it
That's because it gives a config error - you need to fix that.

You have spark B (PM4/D16) set to come full on > 80degC. Surely that will melt the coil?

James

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 3:44 pm
by Greg G
jsmcortina wrote:
Greg G wrote:Ken said the code he sent me with the PID changes had this code as well. FWIW- my car wouldn't start when I flashed it
That's because it gives a config error - you need to fix that.

You have spark B (PM4/D16) set to come full on > 80degC. Surely that will melt the coil?

James
:o

Sorry James- spark B? Kinda lost me there. Am running a DIYPNP, where do I find those settings? Can't find anything relating spark to temperature, and I'm using Fahrenheit units. :?
Car runs fine with the release 3.1.1, and the mario b versions. (Mazda Miata NA6)

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 4:09 pm
by jsmcortina
Output port settings PM4 is enabled and giving a config error, although whether that is accurate is a different question.

James

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 4:35 pm
by Greg G
Oh that! PM4 (ALED) is wired to my auxiliary fan :)

Whew :D

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 4:49 pm
by racingmini_mtl
Greg G wrote:Oh that! PM4 (ALED) is wired to my auxiliary fan :)

Whew :D
And where is your spark B connected? PM4 is where it should be connected so you can't also use it for something else.

Jean

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 5:16 pm
by jsmcortina
Seems Jean and I have missed something here... spark B is PM4 on MS2, but NOT on Microsquirt.
Did Ken send you a Microsquirt compiled s19 ? i.e. ms2_extra_us.s19

James

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 5:48 pm
by jsmcortina
After much fiddling with wires, I hooked up my Microsquirt module to a stim and I get some rpm (EDIT) and no config error with the Microsquirt version of the code.

How to see which is loaded? Look in the title bar and after the (C)KC/JSM/JB you will see either "u" or "MS". That's short for Microsquirt or MS2 code.

James

EDIT: it's late now, I need to validate further tomorrow.

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 6:20 pm
by Greg G
jsmcortina wrote:Seems Jean and I have missed something here... spark B is PM4 on MS2, but NOT on Microsquirt.
Did Ken send you a Microsquirt compiled s19 ? i.e. ms2_extra_us.s19

James
Yes, he sent me a microsquirt s19 file. Spark B? Is that the sequential mod? That uses PT 6&7.

From the DIYPNP documentation:
IG1 - Logic level spark output 1.
IG2 - Logic level spark output 2.
WLD - Warm up LED output, which can be used as the 3rd spark output or as a relay control.
ALD - Acceleration LED, which can be used as the 4th spark output or as a relay control.
Does that mean I'm OK? The car works on all the previous firmware.

I'm afraid I'm a little out of my depth talking about the hardware. Paging Braineack! :)

Re: Low tooth count wheels - beta testers required

Posted: Thu Jul 14, 2011 8:13 pm
by racingmini_mtl
Yes, you're ok. Both James and I made the same mistake: on MS2, ignition 2 (spark B) is on PM4. However, this is not the case for the Microsquirt module in your DIYPNP.

Jean

Re: Low tooth count wheels - beta testers required

Posted: Fri Jul 15, 2011 2:46 am
by jsmcortina
The spark A,B,C,D are referenced in the manual: http://www.msextra.com/doc/ms2extra/module.html

The code seems to work for me, but I don't have the hardware setup very well for the stim to work with the module.

James

Re: Low tooth count wheels - beta testers required

Posted: Fri Jul 15, 2011 10:17 am
by econ
Hi. I could test the code in next setup:

- Engine: Suzuki GSXR750 SRAD Turbo
- 8-1 cam wheel
- 1 pulse/revolution cam sensor
- Microsquirt module based engine control
- 3.1.0 extra code
- full sequential injection
- wasted spark ignition

I have a triggering problem with 3.1.1 code during cranking. If crankshaft angular speed changes rapidly (when engine is trying to start), ECU looses sync. Sync loss reason is usually 2 or 1. If I somehow manage to get the engine running, it runs perfecly. The problem occurs only during cranking.

Triggerin works during cranking too if sparkplugs are not fastened and angular speed is fairly constant.
Could this development of the code make a difference for this problem?