Tuning Boost Control With Bias Table

[aidandj]

Yes, just basic mode. I will play with advanced later this week.

[muythaibxr]

That is actually great news. It means the idea of using the slider is working as I intended.

Ken

[aidandj]

Yeah, its not 100% when I do multiple gear pulls.

Hopefully with PID I can tune that out.

Here is a log about what I mean.

[aidandj]

Also, because I logged so many values, I can close in around my favorite slider value and try to get it even better before I use advanced.

[muythaibxr]

Yeah, its not 100% when I do multiple gear pulls.

Hopefully with PID I can tune that out.

Here is a log about what I mean.

That actually looks pretty good... Some advanced mode tuning will probably get it the rest of the way there. Make sure it's close to perfect in setup mode first though. That dip in boost in both gears makes me think the bias table needs some work at that RPM.

[aidandj]

By not 100% I mean its not perfect. The log I posted is in setup mode. See the small spike, then dip in 3rd?

On another note in what firmware version is this slider "bug" prevalent. I am helping someone on another forum tune closed loop on ms2 and the slider isn't fully working for him. MS2 firmware version 3.3.2.

[nathaninwa]

Sure, I'll give it a shot for sure. I'll down load it tonight and get it installed during lunch tomorrow and make some pulls tomorrow evening

[muythaibxr]

By not 100% I mean its not perfect. The log I posted is in setup mode. See the small spike, then dip in 3rd?

The slider does not do anything in setup mode. That is all down to bias table tuning.

On another note in what firmware version is this slider "bug" prevalent. I am helping someone on another forum tune closed loop on ms2 and the slider isn't fully working for him. MS2 firmware version 3.3.2.

ms2 still uses the same method as firmware rev 1.3. That should not have any issues.

[muythaibxr]

Sure, I'll give it a shot for sure. I'll down load it tonight and get it installed during lunch tomorrow and make some pulls tomorrow evening

Cool thanks.

[aidandj]

By not 100% I mean its not perfect. The log I posted is in setup mode. See the small spike, then dip in 3rd?

The slider does not do anything in setup mode. That is all down to bias table tuning.

On another note in what firmware version is this slider "bug" prevalent. I am helping someone on another forum tune closed loop on ms2 and the slider isn't fully working for him. MS2 firmware version 3.3.2.

ms2 still uses the same method as firmware rev 1.3. That should not have any issues.

d*** it. Typo, I meant basic mode. I will try in setup mode.

[nathaninwa]

I did not mess with PID at all. Just the slider

Car was still in boost by gear so target was 270 rather than 280

pulls in third gear. 3 runs of the 5 are good data

slider at 50 is 16333
slider at 475 is 16441
slider at 250 is 16545

I like the 250 postion in the log the best

its cooler out today and the tires are not warmed up fully, so I experienced some tire spin (25-27psi is near 780 to 800whp) and it feels like the race tank is getting a little thirsty so I need to mix up some more fuel later.

[aidandj]

I also did some tuning today. I played with the bias table and got it pretty close. Need to take some more logs and test in the morning then I will post. Safe to say the new slider works a lot better though.

[nathaninwa]

I agree. The new slider technique is a bit easier and not so over responsive

[muythaibxr]

OK, that and the other change I made will be in the next beta. Thanks for your feedback!

[nathaninwa]

I guess to be a little more detailed, with the old slider with me, just like the others would oscillate bad. I worked from low to high and found smaller numbers to be not so sensitive and that's when I worked with pid to help dial it in

I'm thinking my bias table needs a bit more duty out the top

The new slider is way less sensitive and allows some fault in the users settings and not kill boost control. I'll work with just the slider and see how that affects boost control

I can see from a devolopers point getting a code to work with all setups ca be challenging, I'm seeing just making a spring change how different the setup is. My last spring made 300kpa around 80%, now I'm in the 60%'s for the same boost pressure, so I lost resolution in the system which makes tuning on my part more crtitcal than before

I'm happy to open a log and see a little overshoot and boost flatlined within 10kpa of my target!

Thanks Ken for all your support on this

[AbeFM]

A quick question before I take this as a sign and try the newer code - are you making changes only to the boost control, or is it all PID (ie will idle start/stop working or be the same as before if I change it)?

[jsmcortina]

This is ONLY boost.

James

[aidandj]

I'm still struggling with temperature changes. I will do some more logging, but anything specific to check for would help. Mornings, ~60 degrees, I'll hit over boost (set at 207kpa), later in the same day I will struggle to hit 190kpa. Its like the software is taking into account the bias table too much. Is there a way to lessen the effect of the bias table on the output and let PID do more of the work?

[muythaibxr]

Does this happen in setup mode?

[aidandj]

From what I've seen yes, I will try tomorrow morning and get logs.