Working with my new best friend gVE to smoothen out the VE table works wonders!

 

Just to relate all these graphs to reality, I uploaded a video. This is still with 3.1.1 mariob v5 code (D term not yet corrected). Spark and table switching with AC switch, and the AC compressor is on a time delay relay (here shortened a bit from the graphs posted). It's relatively stable with AC, but has some issues with headlights & AC on at the same time. I need to smoothen the VE table in that area.



Lots of noise in the background. There really is a vacuum cleaner running in the background, it's not my supercharger

 

if I turn my lights and a/c on at the same time my radio turns off from the voltage drop.

 

Does anyone else here with an ms2 want to try the D term changes I made? Seems Greg won't be able to for a while due to weather issues.

Ken

 

What should we expect? I have fixed this in my own firmware, and with a D term of about 20 I didn't notice anything for the better or the worse.

 

I had the best results with zero D term. I'll try it when I get back into town.

 

I'll try it out. My idle control on 3.1.1 sucks currently

 

Car didn't start with the test code. Will wait for a new one.

In the meantime, I continued smoothing out the VE tables looking at gVE, and got the lights + AC to idle fairly well!



I was curious if I could get away with no tableswitching. Here's how it is with spark & table switching off:



It's actually not too bad. The dips are bigger- the instant spark bump from table switching must help out there. But perhaps putting a small slope in the spark table towards lower RPM (my spark tables are flat because I'm using table switching), plus the corrected D term in the next version will be enough to minimize the dip.

All 3 videos: P 25, I 12, D100

 

Mariob v14, P 25i10d60. Looks bumpy, but note the scale. It's actually pretty smooth. Initial dip still not dead, but tolerable.

 

Idle duty seems to hardly need to change?
And any D seems to be a bit weak. Did you try turning it up from 60?
How much did MarioB make D stronger? 2x? 4x?

 

I believe mario just fixed the sign, didn't amplify the effect. Will edit the ini to allow crazy D again, and see what happens

 

Any luck? I have a couple of days that I can test new firmware if someone wants to send me the file. AND instructions on how to upload it if it hasn't been prepackaged. I can swap s19 files and direct tunerstudio to the new ini file necessary.

 

Are you still running Mario_v5 or did you finally move to v14? If you are having good results with the v14 version, can you upload the microsquirt s19 file you are using?

 

V14 works well for me.

Why do you want gregs .s19 file? Just grab the V14 zip from MSExtra.com and your away!

 

The latest v14 upload doesn't have a microsquirt s19 file.

 

I thought the uS and MSII .s19's were merged now?

 

Can someone please confirm? I guess worst case it wouldn't burn to my ecu if it wasn't merged.

 

Haven't had any garage time to fool around with the car this week :P

Arrrg. s19 is not a valid file type for uploading (!) Could we get that fixed?

Anyway I posted it in MSExtra:
http://www.msextra.com/forums/viewto...289497#p289497

Have fun!

Greg

 

I really appreciate all of the extra work for MS2, especially since I can't upgrade the diypnp to an MS3. I'd be more than willing to test new releases.

 

No no no and you'll get all sorts of weird problems if you use the wrong .s19. The INIs have been merged, not the .s19.