doc

OK, I have a MAF sensor guts transplanted into a 3.5" dia tube. I want to tune this MAF for the correct air flow. I have done some logs and best I can tell the MAF in under-reporting by about 35%.

With the MAF table scalars currently set at 17, 30, 53, 87, CF, FF for tables 1 thru 6 respectively, what do I change these values to, so that I can edit the max value up 50% more than what is now in MAF table 1? What does the 17 change to, so that I can change 22 gms/sec to 29.7? Right now with CatTuner, I am limited to 23 gms/sec.

I have been using TunerCat as my editor, but I can not change the MAF scalar values, no capability in CatTuner, but in checking out TunerPro, I can finally do what I need to do.

TIA.

Grim Reaper

It has been a LONG TIME since I re-scaled my buddy's MAF on his highly modified engine (and MAF). So I will quickly answer what I can remember...and hopefully others will pipe up. If I didn't address a part of your question, it's because I can't remember (at this moment).

First, contact TunerCat to add the parameters for changing the "max" MAF values. I did this myself "manually" way back a couple years ago, but it is only necessary for the last table.

You should have a WB to properly scale the MAF and best to do this in REAL-TIME. It will take forever without a Real Time tuner

You can get away without a WB and do a reasonable job by just locking the min/max on the BLM to 128/128 and tuning off the INT. But a WB does help. You will need a driver if you are doing the tuning in Real Time with a WB. I found some parts needed a nice long flat road while other parts needed a nice long hill. You need to hold the car at a particular "air flow" level while playing with various entries. As I recall, the two lower MAF tables and the upper most required the most changes.

Some how, I recall that there were a couple of discrepencies in the "cut-offs" from one table to the other. But, this is function of the bin you start with a base. I found at least 5 different versions. We eventually started with the MAF scalar tables from the ARAP as our base.

Somehow, I recall there is a setting (not in Tunercat) that tells the program when the max value of a particular Scalar table has been reached (that's those 6 values) and to go to the "next table".

Hope this helps.

doc

Thanks for the reply.

I do have a WB, but have had little experience with it. Concerning the BLMs at 128 and using the INT values to tune the MAF, I read this technique in another post last nite. I would think that tuning to the long term fuel adjustment (BLM) would work out best. Since I do not have any tuning capability in real time, the best I can do is to record WB for certain FTCs or engine RPM and take on-road data logs, then come home and disect them and then decide on the next EPROM modification.

Can someone explain why tuning to the INTs is better than tuning to the BLMs? Please,,,,, I had tuned my LS1 PCM by using the LTFTs after I installed the much larger GMAF. LTFT = long term fuel trims (3rd gen BLM).

Grim Reaper

The reason is the INTs can be directly co-orlated to an adjustment needed (especially with the VE table on an SD car....but it also works for MAF).

For example, what adjustment is needed between these examples: BLM=132/INT=124 and BLM=124/INT=132? In fact, is any adjustment needed at all?

The readings are being affected by the Block Learn routine and you need to consider BOTH to make the proper adjustment. I just found it too cumbersome to calculate an adjustment. With the BLMs locked, I could see the amount I was out at a glance. (This is especially necessary if you are driving at the same time...or you just might have an accident while you are trying to figure out if you need to lean or richen the mixture).

It's been a number of years but on one of my old computers (which is dead and not available), I had derived a very complex but quite accurate algorithym (prior to going WB). It was quite good at calculating the amount of adjustment I needed to a particular VE value based on an INT reading. With only the NB I could develop a very good closed loop tune in only a couple of readings for an SD car. I am sure with a little experimentation, a similar algorithym could be developed for MAF cars.

This is why I started to lock the BLM and tune by the INT. I only unlock the BLM once I have a tune that I like and I want the Block Learn to start working properly.

However, with the BLMs locked at 128. Then any INT > 128 needed more fuel and any INT < 128 needed less fuel. The only risk you run is, if your tune is WAY OFF, you may take the INT to it's lowest limit (or highest limit) at which point you'll encounter an O2 High/Low SES condition. But then you are either needing to add/subtract a LOT of fuel depending if you are above/below 128.

Sorry folks, but that SD algorithm is lost. After I went WB, I had no need for it anymore, so I didn't bother trying to recreate it when I remembered it. So please don't ask me for it, as I no longer have it.

But if you are creative and spend a little time, you might be able to derive an algorithm yourself. Some where in my old posts in TGO in 2001-2003 I discussed the components in an early version that was the base of that algorithm I ultimately derived, and lost. So if you dig you may get a starting point to derive your own algorithm to calculate the amount of change you need for a particular VE reading based on the INT value to get you close to 128/128 in Closed Loop.

Grumpy

I'd wire in a second ecm and datalog both the stock and new MAF at the same time.
Then, using that data, and an ecm bench work out the tune.

dimented24x7

iTrader: (2)

One thing to keep in mind about the INT is that in many cals. its not an actual multiplier, but added pulsewidth IIRC. If this is the case in your cal, then the INT cant be used directly to provide the needed changes. Instead, it can only guide you in the direction of needed change. To get the actual corrections, youd have to back calculate the needed change in gms/sec based on the added/removed pulsewidth. With the $6E, though, the injector PW is based on duty cycle, so dividing the INT by the total available injector firing time will give you the needed change in DC, which should also be the needed percentage change in gms/sec airflow.