New official code 029q released

suberimakuri · Post by **suberimakuri** » Sun May 21, 2006 4:38 pm

Anders: Are you gauge and MS taken from the same point on the manifold?
Maybe you have a leak on the MS feed?

snillet · Post by **snillet** » Mon May 22, 2006 12:12 am

Rollagti16 wrote:hello, my 029q2 is also changing the ego correction step timer from ign counts to milisec every time i connect my laptop. My we table are also very odd at idle.. How could that be

I know that one

,i´ve had that happening every now and then to since i switched to 029q (not running q2). that the ego correction timer switches from counts to ms. and at the same time my switchpoint gets set to ZERO.

slappala · Post by **slappala** » Mon May 22, 2006 5:42 am

Some more problems I noted with 29q2 code..

Tuning the priming table does not make any difference on prime pulse widths..it´s always 5ms no matter what I do.

Then there is strange reset/halt thing sometimes after start and sometimes when the unit stays on for a while at steady rpm. What I see is injectors not firing at all and engine dies + box goes into non communication state.

This happens sometimes after a while when engine is stopped and can´t communicate with the box until power has been switched of and on again.

There is no logic what so ever in this and it happens randomly..

I´m using EDIS 6 on BMW m20b25 supercharged engine with normal map sensor etc.

Would there be more stable code? I have IAC valve from bosch motronic 3 pin. so the code should be able to control that too.

So not quite steady code yet what I reckon..

cybo · Post by **cybo** » Mon May 22, 2006 7:38 am

slappala wrote:Some more problems I noted with 29q2 code..

Tuning the priming table does not make any difference on prime pulse widths..it´s always 5ms no matter what I do.

It has been always like that.

If u choose Power up and priming table it´s allways 5ms: you can choose standard prime then it uses value what you choose.

After 2 sec works like it should be.

Twise: first pulse is 5ms and second is from pw table.

Post by **jsmcortina** » Mon May 22, 2006 10:14 am

snillet wrote: I know that one 8) ,i´ve had that happening every now and then to since i switched to 029q (not running q2). that the ego correction timer switches from counts to ms. and at the same time my switchpoint gets set to ZERO.

What do I need to do to replicate this issue?

James

boost junkie · Post by **boost junkie** » Mon May 22, 2006 10:54 am

It's happened to me too. After I hit alt-b in the VE wizzard. It doesn't happen every time though...

snillet · Post by **snillet** » Mon May 22, 2006 12:36 pm

jsmcortina wrote:
snillet wrote: I know that one ,i´ve had that happening every now and then to since i switched to 029q (not running q2). that the ego correction timer switches from counts to ms. and at the same time my switchpoint gets set to ZERO.
What do I need to do to replicate this issue?

James

For me it has happened MOSTLY (not always) when i´m driving around and start to fiddle with the VE bins in the fuel VE table and hit "burn" while engine is running (on idle those times though).

Could be a communication bug between laptop(MT) and MSmaybe ?
Have an old Toshiba portege 610CT (90Mhz pentium) with REAL com port. MT2.25 non beta.

evolotion · Post by **evolotion** » Mon May 22, 2006 12:58 pm

jsut guna throw this in, when im messign with the closed loop idle settings, every now and again (its pretty random) as i change a variable (and not necesarily burn) the megasquirt "crashes" the engie stopps, obviously, and the fuel pump and middle LED stay on constantly.

im fairly sure i can repeat this and capture any info needed. what should i be logging/saving.?

edit also, some of the closed loop settings require a reset to be recognised properly. but im sure they never used too in erlier 29 codes.

captain16vgti · Post by **captain16vgti** » Mon May 22, 2006 1:07 pm

snillet wrote:
Rollagti16 wrote:hello, my 029q2 is also changing the ego correction step timer from ign counts to milisec every time i connect my laptop. My we table are also very odd at idle.. How could that be
I know that one ,i´ve had that happening every now and then to since i switched to 029q (not running q2). that the ego correction timer switches from counts to ms. and at the same time my switchpoint gets set to ZERO.

The same thing happens to me all the time. Also, if you click on required fuel all the settings are back to default. And the last one, If you change any of the constants like injector opening time the car will immediately die and the fuel pump will stay running. Hope this helps.

-Chris

Keithg · Post by **Keithg** » Mon May 22, 2006 2:02 pm

and I had been thinking it was not apparent here, but it just happened while I was driving. this is with Shawn's hires version, but it did just that, ego switch at 0 and ms from cycles.

KeithG

Posted by email.

Anders Karlsson · Post by **Anders Karlsson** » Fri May 26, 2006 2:32 am

Karl_Skewes wrote:Anders: Are you gauge and MS taken from the same point on the manifold?
Maybe you have a leak on the MS feed?

Yes, the gauge and MAP-sensor take vacuum from the same point. I have a T-connector on the hose to split it to both the gauge and MS.
I have not changed any hardware outside the MS since i changed to 029q2 code. With the old 021u3 code the gauge and MS read the same MAP-values.
All the other gauges seem to match, CLT, IAT, RPM.

Keithg · Post by **Keithg** » Fri May 26, 2006 4:44 am

Anders,

With the new version of MT (2.25), it requires a different custom.ini and settings.ini. If I were you, I would move all the stuff in the car1 directory to a different place and make sure that the most current verison of these files are in the car1/mtcfg directory as are the inc files for your installation.

Make sure that you go through and edit the settings.ini file to match your setup. Make sure that teh MSQ was not corrupted somehow or a value otherwise overwritten. The 029 code moved a lot of variables around, so there may be a messed up variable for the map sensor on the engine constants page. I noted that the MT fires up with a 115kpa sensor here in this bin. You probably have a 250 kpa sensor and this could be why your readings are ~2x.

KeithG

Anders Karlsson · Post by **Anders Karlsson** » Sun May 28, 2006 11:32 am

I had my engine running again this friday, and now the MAP-gauges where the same. I took a look at all my settings, made a new req-fuel setting and then it worked. I probably had something set up wrong before.
In req-fuel i had 250 kpa map-sensor by default, so nothing wrong there.
I need to tune the coldstart and cranking settings a little bit, because it was a bit difficult to get started at first. Air-temp in the garage was 16 C.

PeterCip · Post by **PeterCip** » Thu Jun 01, 2006 11:15 am

So is this code rock solid yet? I'm not really feeling the Cycle count changes.. but i'd like to be on the general release. I'm also dreading having to change my fuel table..

newtyres1 · Post by **newtyres1** » Fri Jul 14, 2006 2:21 am

Somewhere between 029i and 029q it looks like constant baro correction in alpha-n mode got broken. I can't download any in-between codes to find out where it happened but I see in the asm that some tweaks on baro corr were done in 029n.

Ian.

Edit: Found 029m and 029o in my computer of all places. Baro corr (for alpha-n) works in the m version but not in the o version.

Keithg · Post by **Keithg** » Fri Jul 14, 2006 4:53 am

There was a comment that const baro responded to increases in altitude, but not decreases (IIRC). James made a tweak that supposedly fixed that. I am running constant baro as well and am curious. What is the problem and how did you find it?

KeithG

Posted by email.

Post by **jsmcortina** » Fri Jul 14, 2006 5:28 am

I tested realtime baro, but I did not test in alpha-N mode where the standard internal map sensor is used for realtime. My baro changes may have broken that in error.

James

Post by **jsmcortina** » Fri Jul 14, 2006 5:37 am

I have had a look at the code and it certainly appears try to make a correct when alpha-N is selected using the internal baro sensor (which must be left open)

To test this, run the motor and hook a hose and syringe up to the internal sensor and simulate high altitude and see how the PW changes.

James

newtyres1 · Post by **newtyres1** » Fri Jul 14, 2006 2:31 pm

I was testing on the stim with a syringe as you say, in 029m the PW would increase slightly and steadily with decreasing MAP, in 029o or 029q there was no change at all. The power needs to be cycled after selecting constant baro in alpha-n. Will double-check the tests again later this morning and report back if I find any more info.

Ian.

newtyres1 · Post by **newtyres1** » Fri Jul 14, 2006 6:06 pm

Ian wrote "Is it not getting switched on? There is simply no response."

Yes, it doesn't get switched on if you don't enable it in the constants page Ian. However 029m didn't require this. It also helps Ian if you don't limit the baro values for processor resets, which were greyed out in 029m, but do affect the range of baro correction in later codes. Have you got it now Ian? Yes thanks Ian, it's all working fine now, sorry about that.

Ian