MS3 pre-1.1 beta 20 released

Testing and development of Megasquirt 3

Moderators: jsmcortina, muythaibxr

MG four six eight
Helpful MS/Extra'er
Posts: 43
Joined: Sun May 17, 2009 10:36 pm

Re: MS3 pre-1.1 beta 20 released

Post by MG four six eight »

Are the extended (X2) injector trim tables still in the 1.1 beta releases? If I remember correctly, it seems like it was added in around beta 10, however searching through the change log I no longer see it listed?
I would like to change back to the 1.1 beta codes (for the extended trim range) from 1.0.6 once the new beta code is released for the WUE/ASE bug fix.

Bill

Edit: I just looked through the ini files for beta 20 and noticed that trim values are at 25.4!
Last edited by MG four six eight on Tue Apr 24, 2012 9:56 pm, edited 3 times in total.
y8s
Master MS/Extra'er
Posts: 478
Joined: Wed Jan 07, 2009 3:21 pm

Re: MS3 pre-1.1 beta 20 released

Post by y8s »

I've noticed using I think this beta and the test code you gave me for the AE fix that my idle-up adders are not adding what I have in the settings.

example: cranking idle duty this morning. CLT temp was 15°C. nearest points for cranking duty are 50.4% at 60C and 53.9% at 0C. actual cranking duty was 72.7%
I'm using voltage compensation. at 9.9V when cranking it's at 10% duty adder. Total settings were around maybe 63%--a full 10% below what the valve was sitting at.

edit/update:

it's fine during cranking, but as soon as the rpm exceeds the cranking limit, 10% comes from somewhere unknown. I'm not using the initial value table.
SvenCS
Helpful MS/Extra'er
Posts: 113
Joined: Mon Feb 08, 2010 10:33 pm
Location: Seattle

Re: MS3 pre-1.1 beta 20 released

Post by SvenCS »

The PID closed loop idle dashpot adder may contribute some. Although, 10% sounds high for this alone.
1972 BMW 3.0 CS, v3.0 board, MS3 + MS3X, external MAP, VR-sensor, 60-2 crank, CAM gear tooth sensor, wasted spark, sequential
y8s
Master MS/Extra'er
Posts: 478
Joined: Wed Jan 07, 2009 3:21 pm

Re: MS3 pre-1.1 beta 20 released

Post by y8s »

my fan and dashpot only add another 6-7% at most.
muythaibxr
Site Admin
Posts: 8228
Joined: Thu Oct 14, 2004 12:48 pm

Re: MS3 pre-1.1 beta 20 released

Post by muythaibxr »

Nothing has changed in that area, so not sure what you're seeing.

Ken
Megasquirt is not for use on pollution controlled vehicles. Any advice I give is for off road use only.
y8s
Master MS/Extra'er
Posts: 478
Joined: Wed Jan 07, 2009 3:21 pm

Re: MS3 pre-1.1 beta 20 released

Post by y8s »

I haven't been using voltage duty compensation very long.

I did two back-to-back startups this evening with it zeroed and with it enabled (linear 10V, 10% to 16V, -3%).

conditions were more or less equal.

with it zeroed, PWM frequency sits at a constant value from key on, through cranking, and until the post cranking timeout ends.
start_nvc.png

With the linear curve, it sits at about the same constant value with key on but then jumps up 10% during cranking. the duty only comes up when the engine starts turning (not when the voltage drops--which happens earlier). after exceeding the max cranking speed, duty jumps up another 10% before tapering down to normal PID levels.
start_vc.png
So I'm wondering--is the voltage correction being added for cranking and then AGAIN to the current duty? It sure seems that way. That's the only change above. If one of the 10% increases was caused by something else, it would appear in both graphs.

Matt
y8s
Master MS/Extra'er
Posts: 478
Joined: Wed Jan 07, 2009 3:21 pm

Re: MS3 pre-1.1 beta 20 released

Post by y8s »

Another thing I noticed while pouring over my latest logs...

tps Accel and map Accel (I thought) were supposed to display the currently added AE % from each category as set in the throttle pump AE graphs.

for example if I set 100 %/second TPSdot to add 10% and move the throttle at 100 %/second I should see "10" in tps Accel. Is this correct? (assume 100% tps based enrichment)

I'm actually seeing logged tps Accel numbers larger than have been set at various points in the curve which confuses me because I can't take those back to the curves and reduce them accordingly. In the above example, the log might show "15" for tps Accel. In some cases it's also undershooting. ie 300 %/sec set for 23 might show up as 18. It doesn't happen every time but as I was going through logs I was noticing this. And I am using 80% tps-map blend so I would expect to see values 80% of what I have in the curve, right?

Log and MSQ
http://y8spec.com/megasquirt/pre1.1preb ... .43.28.msq
http://y8spec.com/megasquirt/pre1.1preb ... .17.25.msl
muythaibxr
Site Admin
Posts: 8228
Joined: Thu Oct 14, 2004 12:48 pm

Re: MS3 pre-1.1 beta 20 released

Post by muythaibxr »

Are you logging with the SDCard? If not, what's logged might not line up with what actually happens during transients.

Ken
Megasquirt is not for use on pollution controlled vehicles. Any advice I give is for off road use only.
muythaibxr
Site Admin
Posts: 8228
Joined: Thu Oct 14, 2004 12:48 pm

Re: MS3 pre-1.1 beta 20 released

Post by muythaibxr »

y8s wrote: So I'm wondering--is the voltage correction being added for cranking and then AGAIN to the current duty? It sure seems that way. That's the only change above. If one of the 10% increases was caused by something else, it would appear in both graphs.
I'll take a look at this and see if that's what's happening.

Ken
Megasquirt is not for use on pollution controlled vehicles. Any advice I give is for off road use only.
y8s
Master MS/Extra'er
Posts: 478
Joined: Wed Jan 07, 2009 3:21 pm

Re: MS3 pre-1.1 beta 20 released

Post by y8s »

muythaibxr wrote:Are you logging with the SDCard? If not, what's logged might not line up with what actually happens during transients.

Ken
computer logs. is the delay between tpsdot changes and tps accel calcs that significant? not criticizing, just curious.
braineack
Experienced MS/Extra'er
Posts: 287
Joined: Wed Oct 17, 2007 11:55 am

Re: MS3 pre-1.1 beta 20 released

Post by braineack »

I'm pretty happy with prebeta21 right now.

I noticed the idle pwm voltage correction at cranking like y8s noted, but i didn't notice the "double-dip", only that I needed to dial it back a bit, and reduce the cranking table a little. I had the voltage curve too aggresive at first, as I drop to 9v during cranking. Seems to be fine after a bit of tuning and lowering the cranking duty a bit.

The accel-pump is amazing. I'm currently struggling with an odd grounding issue where my enrichments had to be completely retuned and my tpsDOT was very noisy and required a high threshold. Coming from alpha11, the filtering improvements coupled with the accel-pump have results in the best throttle response and smoothness to date. (consider that I've been on MsI starting in 2006, then MS2, and now MS3).

The idle improvements (intial duty, timing and voltage correction) have improved the idle quality significantly, better than the stock ECU was ever able to.

Here's the results of my idle after about 10-20 minutes of fooling around, going straight off my alpha 11 map.

http://www.youtube.com/watch?v=aeSrl9EockA
1991 Boosted Miata Running MS3 + MSX w/Full Seq.
https://www.trubokitty.com
ragepower
Experienced MS/Extra'er
Posts: 280
Joined: Fri Dec 17, 2010 9:13 am
Location: Leiria - Portugal

Re: MS3 pre-1.1 beta 20 released

Post by ragepower »

Last week load the beta 20 firmware. The engine seams to run better.

I notice a non-important "bug". The fuel economy l/km isn't working, i have the speed information on MS. Is anything more needed?

Another question in the accel enrichment. I noticed that i had this disable threshold to high, and started to play (trying to tune) with it. With the standard AE seams to a little confuse, so I tried the new accel pump feature and it seams more easy to tune. The accel enrich gauge doesn't move but is noticeable the enrichment, is normal?
And i didn't find this field in the datalog (done using computer), maybe i didn't search enough.

Now im testing CL idle initial values and it seams to work very well.
93white3400z
Master MS/Extra'er
Posts: 685
Joined: Mon Aug 25, 2008 11:37 pm
Location: Montreal, Quebec

Re: MS3 pre-1.1 beta 20 released

Post by 93white3400z »

ragepower wrote:I notice a non-important "bug". The fuel economy l/km isn't working, i have the speed information on MS. Is anything more needed?
In "Engine and sequential settings", make sure you have the "engine size" and "injector size each" filled up properly.
1993 Chevrolet Cavalier Z24
3400 Turbocharged Inter-cooled
MS3 v3.57+MS3X
ragepower
Experienced MS/Extra'er
Posts: 280
Joined: Fri Dec 17, 2010 9:13 am
Location: Leiria - Portugal

Re: MS3 pre-1.1 beta 20 released

Post by ragepower »

93white3400z wrote:
ragepower wrote:I notice a non-important "bug". The fuel economy l/km isn't working, i have the speed information on MS. Is anything more needed?
In "Engine and sequential settings", make sure you have the "engine size" and "injector size each" filled up properly.
I did notice that settings but didn't know what was they for. Thanks
kubeq_sq
Experienced MS/Extra'er
Posts: 157
Joined: Thu May 06, 2010 11:53 am

Re: MS3 pre-1.1 beta 20 released

Post by kubeq_sq »

hello, i am using beta 20 release and latly my engine started not smooth idle... i checked, ans in composite logger there is no trace of cam signal... everything is configured for seqential fuel and spark, but engine runs... i tried to disconnect the cam sensor and it made no diffrence... on earlier code versions everytnig was ok. i place here a composite log file.

what should i do?
Post Reply