Issues with HPX MAF on GTR ECU

[Sandell]

Hi,

I have a S13 with a RB30DET in it, it's based on a RB30 block and a RB26 head. I am running a standard Skyline R32 GTR ECU on it with Nistune of course. I have used a Q45 MAF before and that worked great with 406hp, but it was close to being maxed out. Since then I have upgraded to a HPX MAF in a 3" tube, and that's why I write this post. I also use 1260cc injectors.

I have downloaded BETA version 10.16 and the VQ map for the HPX MAF is by far better than the one in the latest "official" release, but I can't get it to work like I think it should.

When I use the quick solution to change the injectors and MAF through the drop down menu with halved values for the MAF, the car starts for half a second but doesn't keep the idle at all, it dies instantly. After some investigation I managed to get it to run quite well if I don't use the "halve values" option in the MAF resize menu, and at a quite ok lambda value as well. I have only tested it in neutral gear, revving, and also in some light load cruising situations. It works good so far.

My main issue now is that the VQ map reaches to 65536 instead of half that value like I think it's supposed to on a RB26 GTR ECU.

What will happen if/when I reach half of the MAF resolution and the ECU sees over 32768 in both MAF inputs? As far as I have understood, the GTR ECU adds the two MAF inputs together to make it a total max of 65536 as it is on ECU's with only one MAF input.

I have connected the MAF to both MAF inputs, to make that clear as well.

Other interesting info that might be helpful is this; on idle, the MAF puts out approximately 0.64 volts. At that point in the VQ map I need a value of about 420 to make the car idle ok. And I get that value if I use the built in MAF change tool without halving the values.

Am I correct in my assumptions as to how this ECU is supposed to work? Or is this a non-problem which I spend to much time thinking about?

Long post, but hopefully something good comes out of it...

[Matt]

When you used the Q45 MAF, was that basically the same setup? Q45 VQ map, using half tickbox and wired to both MAF inputs on the RB26DETT ECU?

The last stable release 0.10.13 has been just a re-release of the November version. Since then in 0.10.14 onwards the VQ maps have been improved using an updated cracked version of the HPX spreadsheet with the hidden polynomial equation correctly covering the 64 points now and tested against a customers vehicle (previously we tookl the HPX numbers for the 30 or so points from their sheet and spread them out over 64 points which didnt turn out so accurate)

You are correct that the total max will reach 65536 quite quickly, once it hits half way. I would think that you should be using the half version and making sure K is correct

If that doesnt work then next thing I would suggest trying is trying one MAF input to GND (0 volts) using the full size VQ map with the other MAF input. In theory it should only work from the one MAF and have the same output

[Sandell]

It was about two years since i installed and tuned the car with the Q45 MAF so I don't remember all the details. But when looking into the old bin file, I notice that the VQ map also reaches 65536 at the top. Which would mean that I didn't use the "halve values" option at that time either. But I have always had the MAF connected to both MAF inputs.

With the old setup I got 406hp at the crank at 0.7 bar as mentioned before, and the MAF voltage reached a maximum of just below 5 volts.
Unfortunately I don't have any saved logs from that time, I've lost them somewhere. But as far as I can remember, I didn't notice any limits regarding TP value or erratic behaviour when reaching full power and load in the fuel and ignition maps.

Do you think it will work if I half the VQ map values and double the K constant? Would I gain anything from this?

Are there different types of software in the Nistune boards that you have sold? I have had mine for quite some time now, thinking that there could have been a "firmware" upgrade of some sort?

[Matt]

You can only try it out and see how it goes

Firmware hasnt changed for a long time. Only additions since scratch were knock reporting for a few ECUs and PIN locking functionality

[Sandell]

Ok, thank you. I'll do some tests this weekend and report back what my findings are, I guess that there are others here as well that's interested in HPX MAF information.

[Sandell]

I have now tried both options, either not to use "halve values" and also to use "halve values" and double the K value. Any one of those two options give the same result, so far. I don't know how it will behave when the VQ map value will exceed 32768, so for now I'm using the "halve values" option and doubling the K value.

It works great by the way, I have only done the above changes to the standard map, and the car runs smooth as it should. On boost, AFR jumps around a bit between 10.5 and 11.5, but it works great considering that so few changes have been made.

Many thanks for the HPX support!

[Matt]

Excellent. Its great to finally sort out some proper curves for that HPX. Thank the guys on the Mustang forums which cracked open the original spreadsheet

[bardabe]

so let me get this straight, just for my sake. in order to use the HPX MAF with a GTR ECU all we have to do is solder both signal wires together to the single signal wire from the HPX MAF, use the "halve values" setting, and double the K Value?

-Juan