Is the patch for SD only roms or does it apply to MAF as well?

And the obvious question, what was the cause?


BTW, I sent a request in to FIC for Small Pulse Correction data on a couple of injectors but have not received any reply as yet. FIC1100 and FIC1650.

 

Not sure of the cause of the lean spot. My current idea is that its some error in the load calculation. The AFR error is more than 15% though, so its hard to imagine the load calculation being off by that much and not being obvious in data logs. If its a load calculation error, the patch should only be needed for MAF.

I've got the data to build a PW correction table for the 1650s, and I've requested the necessary data for the 1100s. Probably won't have time to build the 1650 table until this weekend.

 

Thanks MrFred!!! I look forward to the 1650 patch. BTW, the table and settings worked great on ID2000 on SD and 91 octane. Thanks!

 

Mychaylo, if you send me the data I can/will build the table if you dont want to bother and post it up.

I did one for the ID1000 this morning, will post it up soon but its all on the other lappy.

Just realised, the table is so large it wont display the whole thing on my internet lappy, will have to do it on another one.
It worked, so here is a correction data set for the ID1000 injectors. Quite a bit different to what is in the table as stock, so hopefully this be be worthwhile.

 

Hate to hijack this thread, because I have ID2000s. Issue is, going back to my original post, is not with the idle.. it's with the very light throttle cruise, like going down a slight incline with your foot barely on the pedal. I actually like my idle on the ID2000s lol.

I have the minimum IPW patch applied and set at 1.032. Anything higher than that and the Idle was too rich and I couldn't tune it out elsewhere.

 

mrfred, a 15% AFR error can be corrected with the MAF AFR Matching Correction table, which as I understand it, you are setting to 0x80 ie 1.00 so no correction.

However you are doing your corrections via the MAF Pulse Constant table (the first of the two 21 element tables).

I have to say it, I do not think this is the best table to tweak to get these corrections.

 

The MAF Pulse Constant and MAF AFR Matching Correction tables both affect the fuel pulse width calculation in exactly the same manner, i.e., fuel pulse width = K*(MAF Pulse)*(MAF AFR Matching) + (accel/decel modifiers). The issue I'm having is that the MAF Pulse and MAF AFR Matching tables are both dependent on MAF Hz and not on RPM, but this lean spot peaks specifically at ~2400 rpm. It doesn't matter whether air flow is at 80 Hz or 300 Hz (or whether the fuel pulse width is 0.6 ms or 2 ms), so the MAF tables (or injector settings tables) are unable to fix it.

I spent a lot of time trying to determine a way to fix the lean spot without adding a new table, but in the end, a patch was the only solution. Must admit that I'm quite curious about whether this is somehow unique to my setup, but I've had no time to try to investigate on other Evos. What I have thought about though is that this seems to be about the same RPM where people are having stutter issues with the SD patch. I can't help but wonder whether these two issues are related.

 

Ok, and the acceleration Load v rpm table was no real use to you in fixing the problem either?

Is your patch a straight 2D rpm axis table with an 0x80 based correction?
What rpm resolution did you apply?

I do like the idea of what you have done, there have been times I have wished for such a table myself and thats on MAF.

 

The accel load vs rpm function is only transitory (I log its contribution) but the lean spot occurs under accel or steady rpm, so its not the answer.

Yes, its a 2D rpm table with an x/128 output that gets multiplied into the base fuel pulse calculation. Its an 8 element table that can be applied over any rpm range that is desired. I'm using it only for this lean spot, so I have the output set to 100% (128) from 0-2000 rpm. Then I create a linear increase up to 118% at 2375 rpm, and then linear back down to 100% at 2675 rpm and beyond. I could get more fancy with the shape, but I haven't seen a need for it yet.

A table like this can easily cause more trouble than it can help, so before I post the patch, I'd like to hear other people confirm that they have seen the same issue.

 

I have this same lean issue on my car mate.
However the weird thing is it only appeared once I changed over to a adjustable fuel pressure regulator.
Swap back to stock reg and issue vanishes.

I have had several other cars with the same issue on SD, different roms, different injector sets, all aftermarket regulators.

HOWEVER of the cars that had done that, ~20+ have had no issues at all.

 

^ Hose expanding maybe?

 

Nope, on accellerating throttle between ~2250 and 2750.

 

Try logging the fuel pressure regulator solenoid to see if it is opening or closing during that time matching the 2250-2750 rpm range.

Mr Fred, are you also using an aftermarket fuel pressure regulator? Same rpm range?

 

Yep, I am using an aftermarket FPR, but I don't see how that can matter for this specific scenario. There is nothing that ties FPR performance to a particular engine rpm range.

 

Pulled this from another evo x forum, but sounds like your issue..

Lack of a fuel damper. On cars that have both an aftermarket fuel rail and an aftermarket fuel pressure regulator, a clear and distinct lean-out issue is present between 2000 and 2500 RPM. This can be tuned out with moderate success, but often results in negative fuel trim swings around the problem area and never results in good drive quality IMHO. The reason for the lean-out on cars with these mods is the deletion of the fuel pulsation damper (which is a part of the stock FPR unit), resulting in a fuel delivery anomaly at that particular fuel delivery load. It is exacerbated by larger injectors. Solution: run a stock fuel rail, run a stock FPR (most people don't even need an AFPR if #5 is addressed), or add back in a fuel pulsation damper like from Radium.