MS3 pre-1.4 alpha 1 released

[franksidebike]

Hi
I have found same Bug in 1.4alpha1

In Rev Limiter when I switch Progressive fuel cut on, the Fuel cut no!

Rev. no fuel cut.msl

Rev. no fuel cut.msq

In Over run fuel cut:
Progressive fuel cut to On, I can no see "Progressive".

O. r. Progressive fuel cut.msq

Progressive fuel return to ON, no "Progressive".

RPM re-engagement (rpmdot), work no.

Fuel adder works sometimes.

I make many Test on the Stim and running engine, no succes.

Frank

[franksidebike]

O. r. Progressive fuel cut.msl

O. r. Progressive fuel return.msl

O. r. Progressive fuel return.msq

[franksidebike]

O. r. RPMdot.msl

[jsmcortina]

In Rev Limiter when I switch Progressive fuel cut on, the Fuel cut no!

Rev. no fuel cut.msl

Rev. no fuel cut.msq

It sure works on my stim - the injectors shut down progressively. I tested an earlier version on my engine too.

James

[jsmcortina]

I think you need to do more testing. All seems ok to me. See attached datalog.

James

[franksidebike]

I think you need to do more testing. All seems ok to me. See attached datalog.

OK!
In your log:
1245s Fuel came back by: 1740rpm/-70rpmdot/3.5PW
1254s Fuel came back by: 1750rpm/-170rpmdot/3.5PW
1263s Fuel came back by: 1748rpm/-30rpmdot/3.5PW
what is "Progresive"?

Frank

[jsmcortina]

what is "Progresive"?
Frank

The injectors turn on/off one by one for a smoother activation and de-activation of the fuel cut. On my car it is very smooth and gets rid of the "clunk" that used to exist on the re-application of fuel.

James

[franksidebike]

The injectors turn on/off one by one for a smoother activation and de-activation of the fuel cut. On my car it is very smooth and gets rid of the "clunk" that used to exist on the re-application of fuel.

I have understood it.

And what is this:
In my Log O. r. RPMdot
1906s Fuel came back with adder rpmdot-340
1909s Fuel came back without adder rpmdot-2790

When my rpm falls rapid the adder comes no?
I do not understand?
Frank

[jsmcortina]

The RPMdot settings control when fuel is re-applied. If RPMs are falling rapidly, then fuel is returned sooner to prevent a stall. If RPMs are falling slowly, then it is ok to keep fuel cut until the lower RPM.
The "accel enrichment" type pulse is showing up ok in my datalogs.

James

[wes kiser]

Since I have been forced to use the RX7 as a DD the last 2 weeks, I have been playing with the closed loop boost control. It certainly behaves like a "normal" loop now.

I have been slowly creeping up on the "P" gain (other 2 terms still 0), and wound up maxing out the gain at 200 today. I assume if I increase the closed loop sensitivity this will effectively multiply that gain?

My test methology has been to set my bias table a couple of % shy of adequate to reach target, and to set my target at 215 kpa up until 5000 rpm, and 222kpa at 5200 rpm. I do 3rd gear sweeps from below the min rpm I can achieve target, until 5500 rpm, looking to get close to the 215 kpa target, and feed duty in to at least approach 222 at 5500. So far I am certianly adding to the bias table, but not really increasing to get to the 222 (it takes more duty anyway as rpm goes up, particularly when warm). It also takes at least 4-5% more duty (will post cold verses warm logs, with turbo speed when actually tuned) on a 75 degree day than a 40 degree day.

[muythaibxr]

Yep, increasing the sensitivity slider should allow you to lower the PID gains.

[wes kiser]

Excited. After increaseing sensitivity this morning (to around 300, with a 190 P gain), I was able to get the "classic" underdamped "P" only controller slightly rippling overshoot, settling slightly below target. Now I have added a little bit of "D" and killed the overshoot, with small constant osscillations.

On way home I will kill a little "P," and add in "I" slowly to get to target. It now works like anyone who has ever tuned one expects a PID loop to behave.

I have a feeling I will be wishing for a setting to clamp the contribution from the "I" term.

[muythaibxr]

It's already clamped using a hard-coded value. I guess I could add another setting if it becomes necessary.

To be honest I'm a bit surprised that much PID tuning is needed... The bias table used with the delta setting should mostly get you to the target.

Ken

[wes kiser]

To tune the PID so that it functions usfully the bias table is intionally "wrong." Not out in left field wrong, but guaranteeing the PID routine has to do its job.

[Chiburbian]

Excited. After increaseing sensitivity this morning (to around 300, with a 190 P gain), I was able to get the "classic" underdamped "P" only controller slightly rippling overshoot, settling slightly below target. Now I have added a little bit of "D" and killed the overshoot, with small constant osscillations.

On way home I will kill a little "P," and add in "I" slowly to get to target. It now works like anyone who has ever tuned one expects a PID loop to behave.

I have a feeling I will be wishing for a setting to clamp the contribution from the "I" term.

I am no super expert but I thought that the NEW PID tuning method for this is to start with PID at 0,0,0 and increase I until you get oscillation etc...

From the thread:

How to tune/test:
- Turn on closed loop boost control
- Turn on the initial values table
- Set your delta as low as you can (around 10-15 % load probably, make it larger if you need to in order to avoid overshoot)
- Make sure overboost protection is configured and working.
- Set control interval to about the same as valve frequency (so 19.5 Hz should be ~51ms control interval).
- Set tuning mode to "advanced"
- Set PID gains to all 0
- Tune the initial value table until you are spooling the way you like and reaching the target at various target values and RPMs
- Tune the I term to hold the target after PID is enabled.
- Tweak the initial values table to hit the target you want initially
- Add P and D term if needed to avoid overshoot and react to changes in throttle position after PID engages.

Or, are you just further along in the tuning process?

[muythaibxr]

He is further along in the tuning process.

[Chiburbian]

Ahh, ok.

Btw, I finally got my speed correct within a mph or two and my gear calculations are now correct but I noticed that something broke in the mpg computations.

Can someone check to see if they are getting reasonably accurate mpg readings? I did notice that when I upgraded firmware my pulses per mile was multiplied by ten. I manually reduced it and I think that is when my economy meter stopped working. What can I post that would help diagnose the problem if it's on my end only?

[hardline]

Problem noted in MS3 1.4 Alpha1

With the Zetec VCT wheel decoder, MS3 will only obtain half sync. Reverted to MS3 1.3 Beta12 and I have full sync again.

[jsmcortina]

How about the current stable release?

James

[hardline]

I have not tried the stable release yet. I'll flash it over the weekend and let you know