Discussion Forums

decel smoothing factor question

General questions and chit-chat about the ECU+.

Re: decel smoothing factor question

Postby tlcoll1 » Mon Jul 21, 2008 10:35 pm

Looks like your bigger fonts don't sit well with my ability to make screens, which look best with the standard Windows fonts. Can't you scroll up and down and left and right with the scroll bars to get to the real numbers?

On the MAP sensor thing: I should be installing a GM sensor in my car fairly soon, so hopefully I can reproduce the issue in my car.

Tom
tlcoll1
Site Admin
 
Posts: 1512
Joined: Sun Jun 13, 2004 1:01 am
Location: Odenton, MD

Re: decel smoothing factor question

Postby honki24 » Tue Jul 22, 2008 11:19 am

I cannot scroll through those windows at all.

I was thinking about it this morning too... while you're hopefully fixing the GUI on that one let's "philosophize".
Okay so I think the way I understand your fix is this:
Upon lift-off you're initiating a clamp on the MAS signal which is incremental based upon RPM.

So again... physically what happens is this:
You lift off, air vents which has already been metered and you get a rich condition because you now have less air going into the closed system than the ECU thought, therefore too much fuel.

The combination of the two:
If you lift off from heavy boost and blow off tons of air, approximately .5 seconds or so later you will have a rich condition where not enough air is in the closed system to balance out the fuel that is in it. This is slightly offset by the fact that your MAS frequency immediately drops to your clamped value (say 35hz) rather than slopes off back to you're idle MAS output.

The problem:
This is only activating when you totally lift off the throttle so see my picture below. Here is an example of in-boost lift-off, where I am using a DIVERTER VALVE. You can see that as soon as I snap off the throttle, my MAS HZ drops correspondingly. Now, it doesn't drop all the way down immediately but actually to a small "shelf" at ~130Hz. I believe this is because I'm recirculating and some of that airflow actually gets recounted when it gets burst into the intake tubing. (I've got proof of this as I've run a WI nozzle pre turbo inlet and when using windshield wiper fluid, found blue residue in my MAS and air filter, which proves this air is being recounted a little bit on lift-off) So lets say I was venting... It wouldn't have set at the ~130HZ shelf, it would have gone down to the ~85HZ level that you see it level out to. So lets say we use this feature. Instead of 85 HZ it would level out to 35HZ which would mean less fuel, but it would be too late. The extra fuel would have already been shot due to the blow off. THis would just lessen the following fuel added, which yes, decreases the likelyhood of stalling... but really doesn't "solve" the problem. You still get a rich condition and a small amount of ring wash. And in my application maybe a fireball... which is eating my turbos alive. I may acutally be able to use this to get rid of that little "shelf" you see from my diverted air by setting it to ~85HZ and 1 second for duration... but I'm not quite convinced that it will truly help for VTA.

True fix: You would have to anticipate a lift off and reduce fuel then... which I can't fathom being done with any real repeatability or level of safety.

Bottom line: Maybe this will work, heck I havent even tried it... but I don't think it will rid you of the rich condition all together.


Note: You can tell what I say being true about that reversion because when HZ hit that tiny peak after lift off you can see my AFR spikes up, which means it went lean right then which means not enough fuel, which means too much air was counted.
Attachments
example.PNG
example.PNG (73.07 KiB) Viewed 15101 times
Advoacate of the Theocratic Reign.
http://www.FlatlineMotorsports.net
03 Evo 8
honki24
 
Posts: 525
Joined: Thu Nov 17, 2005 9:59 am

Re: decel smoothing factor question

Postby honki24 » Wed Jul 23, 2008 10:26 am

Another development. I have not made the switch to the Beta for 2 reasons.
1.) I can't adjust that table I showed in my previous post
2.) You said MAS HZ may not be right for debugging purposes. I've got to be able to log all of my parameters correctly durring my race this weekend.

With that said I had the itch to try a couple of the new cool things that were supposed to be in there, results:
1) Smoothing: absolutely amazing
2) Trace current cell locations - problem...

I couldnt' get it to trace the current cells not because the function didn't work, but because this version seems to have scrambled my axes I think. I've attached a printscreen of my fuel map as the beta opens it. The regular v2.10 opens and reads it just fine. Both versions are pointing to the same ROM and the same definition. I do NOT have the "version 1.31 or higher" box checked in the beta b/c I do not have v1.31 or higher. Just thought I'd add this to your plate, and I was too lazy to start another thread.
Attachments
scrambled map.PNG
scrambled map.PNG (55.41 KiB) Viewed 15093 times
Advoacate of the Theocratic Reign.
http://www.FlatlineMotorsports.net
03 Evo 8
honki24
 
Posts: 525
Joined: Thu Nov 17, 2005 9:59 am

Re: decel smoothing factor question

Postby tlcoll1 » Wed Jul 23, 2008 11:46 pm

honki24 wrote:I was gonna try to test it today but I cant configure the screen. See the screenshot.

New EXE sent.

Tom
tlcoll1
Site Admin
 
Posts: 1512
Joined: Sun Jun 13, 2004 1:01 am
Location: Odenton, MD

Re: decel smoothing factor question

Postby tlcoll1 » Wed Jul 23, 2008 11:51 pm

honki24 wrote:I was thinking about it this morning too... while you're hopefully fixing the GUI on that one let's "philosophize". Okay so I think the way I understand your fix is this:
Upon lift-off you're initiating a clamp on the MAS signal which is incremental based upon RPM.

The way this works (assuming it does) isn't that it tries to correct the counted vs uncounted airflow, rather it just keeps the car from stalling. Normally, when you let off with a VTA BOV, your RPMs drop like crazy and the car stalls. This will effectively raise the drop-down RPM for a few seconds after you let off. Let's see how it works in practice.

Tom
tlcoll1
Site Admin
 
Posts: 1512
Joined: Sun Jun 13, 2004 1:01 am
Location: Odenton, MD

Re: decel smoothing factor question

Postby tlcoll1 » Thu Jul 24, 2008 12:04 am

honki24 wrote:Another development. I have not made the switch to the Beta for 2 reasons.
1.) I can't adjust that table I showed in my previous post
2.) You said MAS HZ may not be right for debugging purposes. I've got to be able to log all of my parameters correctly durring my race this weekend.

Ok, try it out next week.

With that said I had the itch to try a couple of the new cool things that were supposed to be in there, results:
1) Smoothing: absolutely amazing

Yea, I got something right!

2) Trace current cell locations - problem...

I couldnt' get it to trace the current cells not because the function didn't work, but because this version seems to have scrambled my axes I think. I've attached a printscreen of my fuel map as the beta opens it. The regular v2.10 opens and reads it just fine. Both versions are pointing to the same ROM and the same definition. I do NOT have the "version 1.31 or higher" box checked in the beta b/c I do not have v1.31 or higher. Just thought I'd add this to your plate, and I was too lazy to start another thread.

Looks like I got that check box backward. Fixed for the next version, but in the mean time just check the box and all will be well.

Tom
tlcoll1
Site Admin
 
Posts: 1512
Joined: Sun Jun 13, 2004 1:01 am
Location: Odenton, MD

Re: decel smoothing factor question

Postby honki24 » Thu Jul 24, 2008 11:09 am

tlcoll1 wrote:
honki24 wrote:I was thinking about it this morning too... while you're hopefully fixing the GUI on that one let's "philosophize". Okay so I think the way I understand your fix is this:
Upon lift-off you're initiating a clamp on the MAS signal which is incremental based upon RPM.

The way this works (assuming it does) isn't that it tries to correct the counted vs uncounted airflow, rather it just keeps the car from stalling. Normally, when you let off with a VTA BOV, your RPMs drop like crazy and the car stalls. This will effectively raise the drop-down RPM for a few seconds after you let off. Let's see how it works in practice.

Tom

Since I've got to wait till next week sometime to test this anyways, lets theorize a bit. So you're thought process is this:
VTA lift-off = rpm drop = HZ drop = not enough fuel and car stalls
But what I'm telling you is this:
VTA lift-off = artificially increased HZ reading = super rich condition immediately = unburning charges at no load and low rpm = stall.

I see there being a problem with clamping the lower HZ range because what you're really doing is adding more fuel by clamping, right? That is the opposite of what you want to happen.. and not only that... it's too late. The extra fuel is already in the cylinders before the RPM drop all the way... meaning anything you do "post-lift-off" may only band-aide at best. The best bet is to probably artificially lower HZ upon lift-off in order to put LESS fuel in so that hopefully all that extra fuel from the VTA and all that missing fuel from the clamp will even out and not kill the motor. Grillin' what I'm marinadin'?

Like I said, everything we do here can really only be reactive rather than proactive so here's the best way I can see it:
A. ECU+ has to recognize blow-off by quickly decreasing boost (which means it needs to be logging boost and needs to understand what is and isn't a "quick drop" in boost that is evidence of lift-off"
B. ECU+ then subtracts fuel in a quick burst at that moment.
C. ECU+ would need to use varying "burst intensities" for varying boost dropoff speeds. The steeper the dropoff or the higher the boost being blown-off from, the more ECU+ will need to pull fuel in that quick burst.

(note: The way I offset this to some degree w/ the OE ECU is this: I make the low load, high RPM fuel cells very lean so that when I blow off (I have the APS dual vent so some of it vents and some of it recircs) the ECU maps through those lean cells to offset the bit of extra fuel it wants to toss in from the blow off.) It works extremely well for me and my DV, but I am interested in getting rid of the recirc all together because it is hot air that is introduced back into my intake tract and as I've said before it reverts into the MAS a bit too.
Advoacate of the Theocratic Reign.
http://www.FlatlineMotorsports.net
03 Evo 8
honki24
 
Posts: 525
Joined: Thu Nov 17, 2005 9:59 am

Re: decel smoothing factor question

Postby tlcoll1 » Mon Jul 28, 2008 7:44 pm

What you say makes sense, and I guess one way to implement it is to use the clutch switch to indicate that "blow off is about to happen" and decrease fuel accordingly. Let's see how the current approach works first, though.

Tom
tlcoll1
Site Admin
 
Posts: 1512
Joined: Sun Jun 13, 2004 1:01 am
Location: Odenton, MD

Re: decel smoothing factor question

Postby honki24 » Tue Jul 29, 2008 9:02 am

Will do. I'm on vacation this week. I've blown up some IC tubing at the race this past weekend so I'll see when I can fix that next week... maybe I can do some testing.
Advoacate of the Theocratic Reign.
http://www.FlatlineMotorsports.net
03 Evo 8
honki24
 
Posts: 525
Joined: Thu Nov 17, 2005 9:59 am

Previous

Return to General Q&A

Who is online

Users browsing this forum: No registered users and 1 guest

cron