MS2/Extra 2.0 beta 16 (20071122)

SimonKho79 · Post by **SimonKho79** » Mon Dec 03, 2007 7:55 pm

My Daihatsu 3 Cyl prediction doesn't work. What can I do? Without prediction its even worse.

With the 1st Derivation enabled. When I try to accelerate the engine loses power and air fuel all gone. No matter how much Fuel VE i increase, its all same.

Any solution?

Cheers,
Simon

msoultan · Post by **msoultan** » Mon Dec 03, 2007 10:59 pm

Ok, I think I see what's causing it. In this datalog, I held the throttle at around 2k rpm and every so often, DECEL is getting triggered, even though I have the slider all the way to the left and mapdot thresh set to 255. If I remember correctly, a DECEL event would cancel EGO correction, right?

Attached is a datalog. Any ideas what's up?

Thanks,
Mike

msoultan · Post by **msoultan** » Tue Dec 04, 2007 8:32 pm

Did some more testing

I put the accel slider bar all the way to the left, maxed out both map and tps thresholds (255 and 25.5 respectively) and set decel to 100%. And even though the slider is to the left (map), it is still using the TPS threshold for decel. Also another bug is that when I have the decel set to 100, the PW goes up upon a decel event! It takes a value of about 96% for decel to have about no change, but it's still reporting there is a decel event.

This bug is also in 1.0.2. I think the reason I wasn't see this problem as much with 1.0.2 was because I had the TPS lag set to 30, but I'd like to bring this up if possible.

I tested the msii v2.881 code and it also doesn't allow the TPS threshold to be increased past 25.5 and also sets the DECEL status even if the slider is to the left. The only difference is that it doesn't increase the PW when set to 100%.

msoultan · Post by **msoultan** » Tue Dec 04, 2007 9:54 pm

also, this problem is more pronounced at high rpms.. dunno why..

msoultan · Post by **msoultan** » Tue Dec 04, 2007 10:18 pm

Sorry for the 20 gazillion posts on this, but I looked in the code (ms2_extra_main.c) and on line 1562 there is a 1500RPM threshold for decel hard-coded. So that's why it only happens at high RPMs. I'm gonna set it to 8000 for the time being so I can continue tuning...

Thanks,
mike

Post by **muythaibxr** » Wed Dec 05, 2007 8:23 am

I'll have to take a look to see if Al changed any of that in 2.8xx. We're using the code as it was from 2.6xx, and didn't know of any bugs listed there.

I believe I did make it so the TPS value is limited to 25.5, I'll have to check the MAP value later.

Ken

hassmaschine · Post by **hassmaschine** » Wed Dec 05, 2007 12:24 pm

I've had a couple issues with PWM closed loop idle - mostly during warmup on colder days.

The first issue is it seems to completely ignore what I have in my target table - I don't know if this is a settings thing or what, but the highest target I have is 1200rpm, yet when it's warming up the idle is consistently 1300-1400rpm.

Here you can see my coolant temp is 50c, my warmup target is ~1080, but my idle is currently over 1300:

it does this consistently during warmup, but it's usually pretty close once it's up to operating temp.

The second issue, which I don't have a log for yet, is the PWM valve will seem to close and my idle will drop to ~300rpm (almost stalling). if I use the throttle it will come up, but the idle valve doesn't seem to want to activate. I never had this issue until beta 16, but it was warmer before then too. This usually happens when the coolant temp is halfway between cold and fully warmed up. if I can I'll get a log on the way home from work, but it doesn't always happen.

In case either of these is a settings issue, here are my PWM closed loop settings:

I have made some changes to the engine that require me to retune my PID settings, but it drives fine once it's warm and I haven't had the time to mess with it yet.

Post by **muythaibxr** » Wed Dec 05, 2007 4:00 pm

hassmaschine:

Please PM me your email... I've made significant changes to the PWM idle code which I think will address your issues, and could use a third person "alpha-testing" them.

Basically, you'll have to completely retune your PID settings, but the end result will be much better control. KeithG has already tried the changes and had some success with them.

Ken

msoultan · Post by **msoultan** » Wed Dec 05, 2007 4:15 pm

muythaibxr wrote:I'll have to take a look to see if Al changed any of that in 2.8xx. We're using the code as it was from 2.6xx, and didn't know of any bugs listed there.

I believe I did make it so the TPS value is limited to 25.5, I'll have to check the MAP value later.

Ken

could you change it such that a larger value can be used? You should be able to disabled accel enrichment. Right now both map and tps thresh use an unsigned char, but the map threshold isn't dividing it by 10 later...

also, you can't enter a larger value than 25.5 in the MSII code either, so the bug is carried over from Al's code...

Post by **muythaibxr** » Wed Dec 05, 2007 5:22 pm

I can at least see how difficult it would be. IT will probably require some flash rearrangement in that page, which is sometimes easy (if there's any extra space) and sometimes not.

I could easily make the TPS threshold not divide by 10 later. I doubt that it's really necessary to set tpsDOT to .1%. MAP will probably have to be made an unsigned int.

It does seem a bit silly to have it so low.

Ken

msoultan · Post by **msoultan** » Wed Dec 05, 2007 6:47 pm

Thanks, that would be really cool! For now I slid the slider over to the MAP side and that seems to effectively disable it as well. I posted something about this on the MSEFI side so I'll see what Al says.

Thanks!
Mike

Post by **muythaibxr** » Wed Dec 05, 2007 7:19 pm

Oh BTW, I don't think the 25.5/255 was a bug... I think Al did that on purpose.

tk_tech · Post by **tk_tech** » Wed Dec 05, 2007 8:08 pm

i'll try your dec 1 alpha tomorrow let you know my results

sean

msoultan · Post by **msoultan** » Thu Dec 06, 2007 12:37 pm

muythaibxr wrote:Oh BTW, I don't think the 25.5/255 was a bug... I think Al did that on purpose.

He wrote back and said it might be a simple INI change. I guess it all depends on what kinda resolution you guys want for the threshold (I asked that as well). I report back on what he says.

Thanks!
Mike

msoultan · Post by **msoultan** » Thu Dec 06, 2007 3:50 pm

Hey Ken,
Lance posted a reply saying the error was in the INI. There is also an illegal limit for MAPTHRESH so here is the corrected code for both:

Code: Select all

tpsThresh       = scalar,  U08,    584,             "%/sec",    1.00000,   0.00000,  0.00, 255,      0 ; * (  1 byte)
mapThresh       = scalar,  U08,    585,             "kPa/sec",  1.00000,   0.00000,  0.00, 255,      0 ; * (  1 byte)

Here is Lance's reply:

http://www.msefi.com/viewtopic.php?p=185125#185125

I will test to see if a value of decel=100% still causes an increase in PW.

thanks,
Mike

Post by **muythaibxr** » Thu Dec 06, 2007 5:15 pm

Alright, I'll still probably look into increasing them as 25.5 and 255 both seem a bit low to me if you want to use those variables to disable AE.

Ken

msoultan · Post by **msoultan** » Fri Dec 14, 2007 3:13 pm

Hey Ken,
Just wanted to make sure you saw this bug. Easy fix...

[edit]

helps if I post the link:

http://www.msextra.com/viewtopic.php?p=175550#175550

Philip Lochner · Post by **Philip Lochner** » Mon Dec 17, 2007 11:52 am

My MS2/Extra Beta 2.0 causes the car to stumble every time (possibly not ALWAYS) I do a burn.

Is that normal?

msoultan · Post by **msoultan** » Mon Dec 17, 2007 11:59 am

Philip Lochner wrote:My MS2/Extra Beta 2.0 causes the car to stumble every time (possibly not ALWAYS) I do a burn.

Is that normal?

Yup.. happens w/ the 1.0.2 release as well.

hassmaschine · Post by **hassmaschine** » Mon Dec 17, 2007 2:53 pm

Philip Lochner wrote:My MS2/Extra Beta 2.0 causes the car to stumble every time (possibly not ALWAYS) I do a burn.

Is that normal?

it does this with all versions of MS, as far as I know.

Megasquirt Support Forum (MSEXTRA)

MS2/Extra 2.0 beta 16 (20071122)

closed loop

EGO correction bug