Zaphod

@Brain and all MS3/MS2 users.

With the new code 3.0.3u they backported the boost control code from MS3 to MS2.

It is supposed to react better to changes in boost and to bring it to the target better.

I asked at MSextra.com how the tuning procedure changes and Ken answered the following:

You should still tune "P" first, but the number should be much lower than before... maybe 10% or so instead of in the 100's.

Then add enough I to get you to your target. However, when you add I, you'll probably start to overshoot. You'll probably have to double the P term once you start tuning I.

With P and I tuned, you should be overshooting slightly then leveling off right at the desired boost.

Does anyone (Brain?) have tried this already - (And what are your settings)? I have to tune my bost control anyway due to my new turbo so I would like to try the new code...

Thanks
Greets




Use a small amount of D term to compensate for that overshoot.

neogenesis2004

iTrader: (12)

I think Lars had Ken actually tune his boost control in person at the dyno day. And I'm pretty positive Scott has posted a datalog of his boost control since he upgraded. I remember it being flat as a table almost.

Braineack

iTrader: (62)

Shuiends got Ken (the developer of the code) to tune it pretty good. His final numbers were: P-23 I-33 D-20. He was confident he could get it better, but Shuiends thought it was good enough.

log targeting 170kPa: ftp://larssnead.com/megasquirt/dyno6.msl

For mine, I think I'm right around P-23 I-8 and I just starting tuning it yesterday. So I'm still fine tuning it, but I'm close:



that was on my drive in today, targeting 200kPa.

Braineack

iTrader: (62)

Im now at P-26 I-26 D-30



Hopefully I don't need to fool with the settings beyond this. had to keep adding a little I each time to remove oscillation and then alter the P setting to still reach the boost target, then when i was happy with that I added D until it stopped spiking.

WestfieldMX5

I wish my engine ran as cool as yours. I don't even get 198 cruising :(

Braineack

iTrader: (62)

I really don't have anything special but the reroute and a 1" aluminum radiator and really bad shrouding. It gets hot fast and stays steady around 187*F.

cardriverx

iTrader: (12)

Ohh, I need to flash that new firmware asap.

Zaphod

O.K. here are the results... of my first tuning attempts on the new boost control as posted on msextra.com

I just have the first (promising) results of this new boost control algorithm on 3.0.3u.

Settings:
P 23
I 27
D 8

Should be a 5th gear run from low rpm. Target 170kPa - overshoots a bit first but keeps within +/- 3 kPa.



3rd gear to target, shift, 4th gear



Nice flat line just a tad (~ 0.5 kPa ! ) under the target.

Just thought I share this as this thread came up.
Spool seems to be a bit slower than with open loop control last year.

Any suggestions what to change for a little better spool and a little less overshoot when boost is rising slow and constant?

Greets