Chiburbian

iTrader: (1)

I am just trying to understand what is going on internally in the megasquirt that causes my VE map to not look like most that I have seen.

What started this journey was the realization that pulsewidth at a given "VE current" was not consistant across RPM and load values. For example, a VE value of 112 at 1250 rpm and 25kpa might be 1.4 milliseconds but a VE value of 84 at 3,500 rpm and 80kpa was 1.8 milliseconds. These numbers are not exact but I have observed that sort of behavior.

I have since corrected my tune to "include AFR target" which seemed to smooth out a lot of the mid map values so I didn't have these crazy peaks and stuff but I am still seeing what looks to be oddly shaped VE maps compared to most on here.

Obviously if my AFRs are correct who really cares right? But still, I am feeling like a gremlin is hiding in here somewhere and this is just evidence and if I don't figure this out I will have problems later down the line. I certainly don't want to be upgrading to built motor 300hp+ land until I have everything working like a top.

Looking at my maps and settings, is there anything that jumps out at you? Anything you have seen before?

For the ride home tonight, are there any specific things you would like me to log to include in this thread?

BTW, the injectors are ID1000s, stock motor, stock NB2 compression, VVT head, etc.

18psi

iTrader: (76)

are you talking about the part down low where its dumping a ton of fuel? you don't tune that with autotune, you gotta dial that in by hand and then set parameters for autotune not to touch it. otherwise it will decel fuel cut and think its going lean at the same time, resulting in same fuel values at max vacuum as 200kpa. LOL

Chiburbian

iTrader: (1)

what if I turn off decel completely for now?

18psi

iTrader: (76)

I used to still get really weird results down there, so I would set threshold at 30ish kpa and just dial those in by hand.

But try it, maybe you'll get better results.

Chiburbian

iTrader: (1)

True but it also seems strange to me that my idleVE numbers are in the same range... 100-108 or so.

It just doesn't make sense to me why my IDLE VE numbers are higher than my VE numbers in boost at 3800 rpm...

duffbuster243

iTrader: (1)

That bottom row isn't your idle cells, those are high vacuum decel.

Chiburbian

iTrader: (1)

You beat me to posting...
idleVE:


AFR Targets:


My target at 3,800rpm and 160kpa is roughly 11.5:1. It takes a VE value of 105 to reach that AFR.

At IDLE it takes a VE value of 112 to reach 12:1 or so. I will see if I can pull a log real quick.

EDIT:

I just did a quick look for a log...

At 153kpa and about 5k rpm and at a VE value of 127 my resultant pulsewidth is 6.598ms.

At 1k rpm and at 20.3kpa (sitting at a stop light) and a VE number of 125 my resultant pulsewidth is 1.451ms.

18psi

iTrader: (76)

are your injector dead times correct?

*edit: you're using "include afr" so nevermind, this all makes sense. include afr should make your ve table super smooth and the values really really close to each other since it uses the afr target table a lot more for calculating fuel

let me see if I can dig up a thread where we discussed this

hornetball

iTrader: (4)

Are you matching your AFR targets? If you are, move on and don't worry about it. Also, some cells are much harder to hit/tune than others. It looks to me like you started rich and the cells you haven't or can't hit are still rich.

Chiburbian

iTrader: (1)

I am not so concerned about the mid load, mid RPM part of my map so much as why my VE numbers act the way they do. (not corresponding to PW)

You may hate me for this, but I am running alpha code from a couple months ago and I noticed that it contained a new "page" called Fuel Pump and Pressure.

Here are my current settings for that page:



I tried changing "Pressure regulation/correction" from "Vac referenced" to "Fixed" and I don't recall exactly what it did but it certainly didn't fix my problems, but then again that was before I switched to "include AFR targets."


I also have no idea how to set up the rest of that wizard to work with my setup. Maybe I need to revert to a release version of the firmware to see if that fixes my problem?

I mainly installed the "alpha" code to play with the new closed loop boost control settings but haven't gotten them to work either.

Chiburbian

iTrader: (1)

I just looked at my build thread. When I set "Fuel Pressure regulation/compensation" to "fixed" I got a new modifier called "fuel pressure compensation" which pulled fuel at low RPMs/loads which only compounded the problem.

I realize that you shouldn't tune low load/RPM with VEAL or trust the O2 sensor, but in general my car runs really poorly above 15.2afr indicated at idle. In order to get it to idle (especially when it's hot out, AC on etc) I need to be a bit richer. I could probably go richer than 12.2 but for now that is the only way I can keep it stable and not hunting.

DNMakinson

iTrader: (3)

On the first selection, "FUEL PUMP MODE", you have "ON/OFF" selected. Is there a plain OFF you can choose? If so, that would be what you want.

Otherwise, yes, a released version of FW would be a good idea.

If that table is in the fueling equation, then you would be reducing the idle area by a 0.25 factor, so then you would have to be running 4X the VE in that area to compensate. Interesting that it would apply the factor when you have no pressure input (set to NO).

Why does it matter? Compensating for errors is never as good as doing it correctly. And your comment about understanding what MS is doing is right, also.

Last, in a log, there may be an annunciator at the bottom with G(something) that represents this correction in the fueling equation.

Ben

iTrader: (33)

On/Off is the default setting. It means that the MS3 only switches the fuel pump relay on or off, such as the case with the NA and NB Miatas. This menu is to set up fuel pump control for cars that run multiple speed fuel pumps, with a few different strategies for cars with 2/3 stage pumps or those with continuously variable pumps (like Fords that don't have fuel pressure regulators, but instead have a fuel pressure sensor in the rail and vary DC on the pump to vary fuel pressure with load).

That aside, if the mode is set to on/off and pressure regulation is set to vacuum referenced, this table is not applying any corrections. It's doing nothing.

DNMakinson

iTrader: (3)

Thanks, Ben. Seems like I need to go into the mode of talking less, listening more.

What do you think of Chiburbian's VE table? Does it seem as strange to you as to others?

Chiburbian

iTrader: (1)

^^ +1