dryad001

iTrader: (27)

Alright this has been boggling me for the past few weeks.

First off my trims seem to be in check and they never really move around. During idle it takes a few seconds for the AFRs to come back to stioch, so theres really no problem there. But during low speed (30-40mph) cruise it seems as if the car runs extremely rich (11.1-12.2). It actually seems to drive fine but obviously the spoolup is garbage. But then during a high speed cruise (55-70mph) the car runs extremely lean (16.5-17.7AFR). There is obviously something wrong in the tune.


Now I am running a blowthrough set-up with a Tial BOV with the 7lb spring. Could this be the chttps://www.evolutionm.net/forums/newthread.php?do=newthread&f=179
evolutionm.net - Post New Threadulprit? Im thinking if I move to a 9lb spring the valve would stay shut mostly all the time except for a throttle lift but then quickly close again. This might make tuning a bit easier. Also I am running 780cc injectors. Partial map is below.

I have a ton of logs of just driving normally around town and nothing looks different than usual. It's just the fueling I can't figure out. I have been trying to follow the MAF scaling thread but it has gone way above me and I now have no clue what they are talking about. Can someone put this MAF thing in layman terms? Or just try and help me out? Any info would be greatly appreciated.

l2r99gst

iTrader: (2)

From your screenshot, it looks like you changed your MAF size value already. Did you change the maf scaling at all? Do you know about the adder?

I don't think this has been learned from disassembly, but the DSM ECUs use the Maf size value for timing lookups and it uses the Maf scaling value and adder * Hz for the injector pulsewidth calculations.

You say that you are running like 11.2 at a low speed cruise. That sounds very far off. I assume that your LTFTs and STFTs are maxed out negative at that point?

What kind of blow-though MAF are you using and what have you done so far to calibrate it. I could probably point you in the right direction on how to use the Maf scaling and adder to calibrate it, but if you can use whatever hardware/software came with your Maf, it may be better, then use the scaling table for fine tuning.

But, it looks like you already did some kind of tuning, since your Maf size is changed. Who told you that value? Also, if you know how to change the scale in your Maf scaling map to uint8, do that. It will be a bit easier to work with, if you read my spreadsheet.

Your BOV shouldn't be an issue if you have a blow-through setup. The BOV is before the MAF right? If so, then the BOV won't matter even if it is open at cruise or idle, etc. The air would still be metered by the MAF since it is upstream of the BOV.


Eric

burgers22

Sorry to hijack, which scale do you change to uint8? The left one, which is marked Load or the right one which is marked g/s. For the left one, load, I added a new scale, which based on your work, was uint16 *6.25 or /6.25 for read or write, so the scale reads in Hz.

Ok, I've read the update in the other thread, so now have the info I need. Thanks.

dryad001

iTrader: (27)

Alright I put the MAF size back to the OEM spec, and resacaled everything to the uint8. Did some driving around with some very minor adjustments. Everything is pretty good so far. I only got in some low speed cruising in different gears and it looks to be close enough for now. All I need is to wait for my next off day to do some high speed runs and cruising and Im sure with very little manipulation of the scaling values everything will work out just fine. If you want to see some before and after logs i can send them out.

Thanks a bunch.

I have yet to play with the MAF smoothing table but I read what you may think it is. Very intresting after I read the entire thread a few times.

l2r99gst

iTrader: (2)

Yes, to be honest I don't know what the smoothing table does, so if you want to experiment with it, that's your call. Also, as far as your Maf size, I wasn't necessarily telling you to change it. I was just wondering where you got that number from, so I have an idea of what you have tried so far.

If your AFRs are off by as much as you stated in your first post, though, I would be careful no matter what you do. You don't want to break anything going WOT before you know what's going to happen. But I am assuming that this is only during closed loop fueling and your open-loop fueling is OK?

What blowthrough MAF do you have on the car...if it is something like the MAF translator, does it have hardware/software settings for tuning? How long have you had it? Have you tuned the open-loop fuel tables, but are just having troubles in closed-loop?

I just want to make sure that I completely understand what your troubles are before I try to make any type of suggestions.

I reread your first post again and you said that your trims are in check. If your lowe speed cruising AFRs are richer than stoich and your high speed cruising AFR are leaner than stoich, your trims can't possibly be in check if you're running closed loop. If you do any logging, make sure to at least include your short term and long term fuel trims, AFR, Hz, RPM, and timing.


Eric

dryad001

iTrader: (27)

I am running the MAF translator from full throttle using a ported, descreened 3" GM MAF. It does have dials for basic tuning to get the car running and then dials that are rpm dependent. This locks the BARO and AIR TEMP at a certian value.

I'll try some changes on the smoothing table tonight and see what I can come up with.

You are right, by going back and re-checking my trims I was loking at the wrong column, I completely forgot about the 02 sensor trim.

The only thing that is suffering is closed loop as you stated and I guess I should have thought out my initial post a little better. playing around with some of the lower Hz setting in the MAF scaling map, the fueling is a lot better than it was. On the way home form work this morning I ended up taking the freeway and did some logging but my laptop batteries gave out in the middle of it so the log isn't worth anything useful.

Now that I kinda figured out how everything works together although Im still learning, it's getting a lot easier to understand what you guys are talking about.

What I really need or actually waiting on is MJs new version logger. Im not a huge fan of Evoscan but this is the only one out there that can log AFRs for now. C'mon MJ, lets get er done.

l2r99gst

iTrader: (2)

OK, well that's good to hear then. I was just nervous that your WOT wasn't tuned and you were going to start doing pulls and end up with 15:1 AFRs and break something.

I actually used the MAF Translator on my DSM years ago. I don't know what has changed since then, but there were four dials in the translator box...one for injector scaling, and one for low, mid, high calibration. As you said, baro and intake temp were locked (the first version had the baro off by quite a bit which caused calibration problems, but snipping a resistor fixed that).

Anyway, back then, I had DSMLink, and people with DSMLink that had the MAFT just zeroed out all of the setting in the MAFT box and used DSMLink to calibrate the MAF. But, with DSMLink, we were able to calibaration airflow from 20% lower to 60% higher. You're basically trying to do this with ECUFlash at this point, but the actual maf scaling table won't give you this much range. You will get this much range if you make changes to the maf adder in combination with the scaling table, but I don't personally know if changing the adder will negatively affect anything else at this point. I haven't changed the adder in my ROM yet or did any testing, so I would advise against it.

This is what I would suggest...you don't have to take this approach, but this is how I would think about doing it:

1. Log your short term and long term fuel trims at idle, and a bunch of different crusing speeds (RPMs). Make sure that you have at least Hz values logged with these logs, since that will give your the area of the maf scaling map that you need to change.
2. If you are waiting for MJ's newest logger you should be fine, but if you use EvoScan for now, make sure that you change the scaling of the fuel trims to be the correct scaling, no 0-200 or 0-100 crap. Look at MJ's scaling for Mitsulogger...he has it right. You need the correct scaling to know what percentage of fuel the ECU is trying to add or take away. And you need both STFT and LTFT. I think EvoScan has these labeled wrong, too.
3. Based on your logs, see what areas have the fuel trmis maxed out negative or positive. At these areas, if possible, I would use your MAFT to make the coarse adjustments. (At this point, I would really recommend using the MAFT for course adjustments, and the maf scaling table for fine adjustments...remember, the maf scaling table doesn't have much room to go richer in the higher Hz ranges, so you only have fine control here. You *could* make the maf adder higher and lower all of the numbers in the maf scaling table for more room, but I don't know for sure if doing this would affect anything else. More disassembly is needed).
4. Once you have used your MAFT controls to start to get things a lot better, where the fuel trims aren't maxed out anymore, then you can use the maf scaling table as I have explained in the spreadsheet. Basically, if you need a 10% change in fueling, you take your maf scaling number, add it to 140, and take 10% of that number. At that resulting number to your maf scaling number to add 10% fuel. Subtract that resulting number to the maf scaling number to remove 10% fuel.

Hope this helps and you get everything working correctly.


Eric

dryad001

iTrader: (27)

Thanks for the advise.

Since Evoscan is garbage in my book Im going to wait with something I can actually work with. I looked at some of my STFT and LTFT in the Evoscan logs and they dont move. STFT is pegged at 140 and LTFT is at 146, they never change. Ill have to go back and log with Mitsulogger and see what the problem is.

l2r99gst

iTrader: (2)

Someone more familiar with EvoScan's default scalings can comment, but that may be the maximum values. If so, then your fuel trims may be pegged, meaning that you are very far out of calibration, needing to add fuel.

Eric