MAF Calibration (No Spreadsheets!)

[Black02SS]

It was suggested on here but I had to try it for myself. I am not to up on worksheets anymore and wanted something that was pretty easy to use. I used a custom MAP/Histogram that is based on MAF Frequency and AFR(posted in the other thread. The tables are setup so the difference from commanded and actual is displayed in a percentage just the same as LTFT/Wideband. I plotted the VE table just the same as I would if I was tuning for VE, but instead used the % error from the MAF table and after two passes was done. The ability to make our own custom MAP's is great and no more excel!!! Not to mention the guy is tickled pink now with his car.

[405HP_Z06]

It was suggested on here but I had to try it for myself. I am not to up on worksheets anymore and wanted something that was pretty easy to use. I used a custom MAP/Histogram that is based on MAF Frequency and AFR(posted in the other thread. The tables are setup so the difference from commanded and actual is displayed in a percentage just the same as LTFT/Wideband. I plotted the VE table just the same as I would if I was tuning for VE, but instead used the % error from the MAF table and after two passes was done. The ability to make our own custom MAP's is great and no more excel!!! Not to mention the guy is tickled pink now with his car.

That is too awesome! I need some cash to buy EFILive! :P

[Black02SS]

Once you switch, you will have wished you would have done it sooner. I know I was.

[bink]

It was suggested on here but I had to try it for myself. I am not to up on worksheets anymore and wanted something that was pretty easy to use. I used a custom MAP/Histogram that is based on MAF Frequency and AFR(posted in the other thread. The tables are setup so the difference from commanded and actual is displayed in a percentage just the same as LTFT/Wideband. I plotted the VE table just the same as I would if I was tuning for VE, but instead used the % error from the MAF table and after two passes was done. The ability to make our own custom MAP's is great and no more excel!!! Not to mention the guy is tickled pink now with his car.

Very Cool! 8) :lol:

Did you create a PID for % error - I can't find one??

Can you post the actual steps you used to create the map in EFILive - uhh...a method for the 'tards, like me? :D
( yes, I know that's not pol correct :P )

[Nox]

i would like to see this as well.

[Black02SS]

Here is the MAF % Error MAP that I used. This is for my LM1 using the Tech Edge BEN. Again, this works awsome on my car. Just copy and paste with a factor back to the B5001 Table. You guys need anything else, let me know.

[SinisterSS]

9375 Hz MAF map for PLX Wideband O2.

8)

[SinisterSS]

If your MAF table goes to 12000 like mine, use this map -

12000 Hz MAF map for PLX Wideband O2.

8)

[SinisterSS]

By the way kids, this calibration method works extremely well.

Step by step:

0. Assuming a calibrated VE using speed density mode and "AutoVE"...
1. Put the tune in closed loop and all normal PE, etc.
2. Disable LTFT {B3801} and STFT idle trims {B4108}.
3. Log a bunch of data - normal driving, wide open (to hit the high Hz, do some bonzai 4th gear 100+ mph runs :wink:).
4. Save the log file.
5. Filter the throttle transients out using the BEN filter.
6. Copy and paste the MAF map as a factor into your MAF table in the tune.
7. Repeat steps 3 through 6 and you'll have the MAF dialed.

My MAF is dialed in to within 1% from 2375 Hz to 11k+.

Thanks black02ss; nice work! 8)

[Black02SS]

Nice write up. I edited my attachment to represent 12000hz.

[Nox]

thanks sinister! this is greatly appreciated.

[TAQuickness]

Now that's good Schtuff!

[BowlingSS]

Should I keep the MAF after SD or no? That is the question???

Bill
8)

[ZL1Killa]

you can keep it or trash it. if you calibrated it in with this map then i would say plug it in and leave it.

My buddy nick williams knows (and I'm sure that you also know) people that do and don;t have the MAF. Nick is running the MAF and his car is awesome.

[MSGHUFF]

.
6. Copy and paste the MAF map as a factor into your MAF table in the tune.



When I do this I paste and multiply with lables right?

[Dirk Diggler]

Yup

[Bruce Melton]

When I do this I paste and multiply with lables right?

You will have to change the column from 8000 to Value,Value
993

I have been fooling with this for a while and once changed table I could Paste and Multilply but it seems to take my AFRs lean (PLX).

Let us know how it works out for you -_-.

[SSpdDmon]

In my experience, there's not a direct 1:1 relationship with the MAF and AFR. I usually make partial corrections, which is where Excel does come in handy. That way, I can turn a BEN of 1.05 into 1.025 (50% correction) easily for each frequency. This helps dial things in a little closer and really doesn't take much time at all.

[Bruce Melton]

I assume that increasing the value in the MAF table increases the amt of air added?
In that the AutoVE tunes for no MAF when you throw the MAF back in with it's air correction the MAF air should be ~0 given identical conditions.

Kinda hard to imagine.

[TAQuickness]

Bruce
yes - increasing the value in the MAF table effectively tells the PCM there is more air for the respective Hz value read from the MAF. The MAF calibration is a look uptable similar in nature to the VE table.



In my experience, there's not a direct 1:1 relationship with the MAF and AFR. I usually make partial corrections, which is where Excel does come in handy. That way, I can turn a BEN of 1.05 into 1.025 (50% correction) easily for each frequency. This helps dial things in a little closer and really doesn't take much time at all.

Have you seen this across the board or down low where the VE and MAF are "blended"?

[SSpdDmon]

Bruce
Have you seen this across the board or down low where the VE and MAF are "blended"?
I'm pretty sure I noticed this at WOT too. It's been a while since I've played around with this, but I'm pretty sure that's the way I remember it. The funny thing is, applying the full correction down low where the VE is referenced actually over corrected the MAF.

[Bruce Melton]

Yes, on over correcting the MAF. I sent two hours udating AutoVE then immediately made MAF runs and applied to stock MAF table. The result after plugging this all into a fully MAF/CL enabled tune was AFRs generally above 15 to 16 except in VE where they were 12.5 to 13.4.
I see the problem as adding MAF air where it is not needed. We tune for no MAF then add table for MAF to fill in where no gap exsists.

How wrong would it be to AutoVE with stock MAF table enabled?
Would the resulting VE MAP be adjusting VE (+ MAF) = final AFR? If so, that sounds like a winner for those of us who are MAF challenged.

[Tydriver]

Dhayang !! Thats good stuff.. I need to try it out..

Hey, Chad, you ever get a chance to look at that TUNE I sent ya ?

[SSpdDmon]

Bruce - You can't AutoVE with the MAF enabled. That's the inherent problem. You kill the MAF to get an accurate VE...then re-enable the MAF and get an accurate MAF with the assumption the VE is still good. I guess you could kill the VE with B0120 set to a number below idle while tuning the MAF. But, it might not be accurate enough at low airflow scenarios. I plan on experimenting with this at some point (if I ever get my car on the road again).

[Bruce Melton]

It seems as though (in theory) that once a good VE is AutoVEd that the VE table should be reduced by some percentage to allow more room for a more normal amount of correcting MAF air. The MAF is a major player in the factory tune's ability to lock in all-weather AFRs. Seems we are limiting it's positive influence.

Solving backwards from desired AFR is the purpose of AutoVE but we are forced, it seems, to leave out the big equalizer.

Ok, I am done.:rolleyes:

[SSpdDmon]

It seems as though (in theory) that once a good VE is AutoVEd that the VE table should be reduced by some percentage to allow more room for a more normal amount of correcting MAF air. The MAF is a major player in the factory tune's ability to lock in all-weather AFRs. Seems we are limiting it's positive influence.

Solving backwards from desired AFR is the purpose of AutoVE but we are forced, it seems, to leave out the big equalizer.

Ok, I am done.:rolleyes:
My understanding is, airflow is airflow. The PCM will calculate fuel based on what we tell it. Kill the MAF so only one variable is in the game (VE). Once VE is dialed in, it's no longer a variable assuming it's dialed in correctly. Then, the only variable left should be the MAF once re-enabled. Tune the MAF and it should be able to compensate appropriately just like it did from the factory. Reducing the tuned VE table will most likely lead to lean transient fueling according to what I've read on the boards. So if anything, I'd think you'd want it increased slightly.

Then again, I talk to other tuners (who seem to know some of the engineers here at GM Warren Tech Center) who say you should have the MAF flowed for your setup and leave it. Then, adjust the injectors for appropriate fueling. That trends back to the older way of doing things. But, they seem to be making all of the power. The last car he just tuned was a Mustang with a twin-turbo kit and upgraded fuel system. The damn thing made good power and 25mpg on the 1000+ mile trip home. So, I don't know which way to tune anymore... I kind of like the idea of getting the AFR dialed in for the most part in less than 4 hours instead of 4 weeks. However, it does seem frowned upon with a lot of the do-it-yourself tuners.

[Bruce Melton]

IMHO the results achived by AutoVE are the foundation for a lifetime tuning hobby. The results can be great but reqiure "tune of the day" (TOTD) for optimum drivability and peak power.

I have done it both this ways and "old school" and with diddled IFRs, was much more repeatable but the transitions were poorer than a fresh AutoVE. It also did suck up Injector capacity, but then again I am fighting loafing low end 42s now.

What VE table does one start with for a home brew 402 for instance?
For a lighter head and cam car that is not an issue but when you do bigger or more radical motors it is murky to me.

In the end a pretty PE enhanced, WOT AFR wins races and makes dyno queens?
How to get out the the TOTD hobby?


Did I say I was done last time?

[1998ws6]

The OEM aproach of having the intake track flowed as an assembly and using that MAF calibration is probably correct, but how many people have a flow lab where they can have this kind of work done?

For tuning that mustang, depending on what kind of tuning software was being used, they were probably provided with correct injector flow data from the company that provides the tuning software, so that would be the value that I would take more faith in, and adjust the maf.

At the end of the day, you need to ask yourself, what factor do I have the least confidence in being correct. On a stock LSx car, I would say the injector flow data is correct, as it came from a flow lab as well.

Just my 0.02

Ryan

[98SS2836]

Does anyone have this map it seems the attachments are too old to down load.

[SSpdDmon]

You can make it fairly easily. Open the scanner and the tuner. Create a new map in the scanner. In the tuner, right click in the upper left grey cell (so the entire MAF table turns blue and a menu pops up) and select "Copy with labels..." In the scanner, select MAF Hz for the rows and click on the Paste Labels button. For the column, select RPM, set the number of columns to 1, and then type ",8000" for the labels (yes - type the comma too). The data values you want to use depends on how you're tuning - BEN for your WBO2 if you're using one or LTFT/STFT's if you're using the stock NBO2's (not recommended for anything other than normal, closed-loop operation). Simple enough... :)

[Bruce Melton]

I emailed you the 12K file with repaired columns. Let me know if you need the 8k instead.

SSpdDmon, your help has been greatly appreciated. It seems many have sworn off MAFs which does not work for some of us. Cheers!

[98SS2836]

I emailed you the 12K file with repaired columns. Let me know if you need the 8k instead.

SSpdDmon, your help has been greatly appreciated. It seems many have sworn off MAFs which does not work for some of us. Cheers!
thank you:D

[98SS2836]

I emailed you the 12K file with repaired columns. Let me know if you need the 8k instead.

SSpdDmon, your help has been greatly appreciated. It seems many have sworn off MAFs which does not work for some of us. Cheers!

That's exactly how i have it set up thanks again I wanted to verfy that i have it correctly.

Does anyone leave it is OL with the maf. Or does every one go back to closed loop.

[SSpdDmon]

I'm running an open loop MAF tune in my Z right now because I'm sick of the NBO2 issues in closed loop. Only time I run into an issue is when the IATs/ECTs get hot enough to lean out the mixture. I just haven't had time to get into the custom OS stuff to fix that.

[Bruce Melton]

[quote=SSpdDmon]You can make it fairly easily. Open the scanner and the tuner. Create a new map in the scanner. In the tuner, right click in the upper left grey cell (so the entire MAF table turns blue and a menu pops up) and select "Copy with labels..." In the scanner, select MAF Hz for the rows and click on the Paste Labels button. For the column, select RPM, set the number of columns to 1, and then type ",8000" for the labels (yes - type the comma too). The data values you want to use depends on how you're tuning - BEN for your WBO2 if you're using one or LTFT/STFT's if you're using the stock NBO2's (not recommended for anything other than normal, closed-loop operation). Simple enough... :)[

Ok, I guess I do have a problem on closer examination-

I found with the MAF table Sinister offered above I could not copy with labels and paste and multiply because of "8000" col label so as suggested, I changed the column label to ",value,value" --That fixed it.
Except-
For some reason that limits the scale to 1500-9375 which is less than the 1500-12000 needed to match the MAF scaling with 125 increment steps.
And-
The MAF table (B5001) is---- HZ > col and Grams/sec> rows
The BEN_correction MAP is--- MAF freq > col and RPM > rows

I am losing it :Eyecrazy:

[SSpdDmon]

For the column tab, I have always logged Engine RPM, set the # of columns to 1, and used the label ",8000" to capture the data. You don't need to paste with labels on the column tab - just the row tab. If the MAP has reverted to 1500-9375, simply copy the table in the tune and paste the labels once again. Then, save it.

I've yet to try the ",value" label you mention. I'll have to sometime. For some reason, it only wants to remember the first 64 rows for the MAF Hz though. Maybe this is a glitch that can be fixed before the next release???

[Bruce Melton]

But don't you get these factor values that need to be copied and multiplied to the B5001??
http://forum.efilive.com/attachment.php?attachmentid=993&d=1159619998

[Xtnct00WS6]

Here is the MAF % Error MAP that I used. This is for my LM1 using the Tech Edge BEN. Again, this works awsome on my car. Just copy and paste with a factor back to the B5001 Table. You guys need anything else, let me know.

You're the shit :)

[SSpdDmon]

...The MAF table (B5001) is---- HZ > col and Grams/sec> rows
The BEN_correction MAP is--- MAF freq > col and RPM > rows...

Then, this MAP was built wrong. The MAF freq should be rows and RPM should be the col.


But don't you get these factor values that need to be copied and multiplied to the B5001??
http://forum.efilive.com/attachment.php?attachmentid=993&d=1159619998
Yup, I get those BENs (as long as supporting pids are logged). Then, they get filtered and take a pit stop in an Excel sheet so that only 33~66% of the correction value is applied.

[Bruce Melton]

Aaah-HA
I missed the pit stop no wonder I am out of gas (lean)!

[98SS2836]

what's this pit stop in excel. ?? :Eyecrazy:

[SSpdDmon]

I set up a spread sheet so that I could apply a percentage of the correction and not the whole correction. Shoot me an email to SAJ79 at aol dot com and I'll send you the sheet if you want. It might be a few days til I get to it though...it's on the other computer.

[Beer99C5]

I am using earlier logs, and just dabbling (stuck offshore)...
Thanks to TAQuickness for showing me the trick of being able to add calc PIDS if you happened to log the parameters needed for it AFTER the FACT :)

OK, does this look right?

http://www.retrorelics.net/jim/Boards/EFILive/MAFFreqben.jpg

So I would do something like this to modify the MAF FREQ table when I have enough samples...This is halfing the Ben Difference per SSpdDmon

http://www.retrorelics.net/jim/Boards/EFILive/excelMAFben.jpg

I would apply the new values and relog til I was satisfied...

[Bruce Melton]

Beer,
There are three things you may encounter:

* The MAP col label 8000 will prevent you from direct copy with labels. Using "Value" instead allows paste with labels.

* There is currently a 64 row limit for MAPs which is an issue for 12k MAFs

* My experience has been that adding the MAF back into an optimized VE MAFless tune tends to lean out AFRs.

Let us know how it works out?

[SSpdDmon]

One other thing to be careful about...

Make sure you have enough resolution on your log. 7.2 and 7.24389 are two totally different numbers when it comes to the MAF. Other than that, it looks like you're on the right track. :)

[Dirk Diggler]

No matter the precision you see the software always copies out to 6 or so decimal places. This can be verified by copying and pastign in Excel

[joecar]

...
* There is currently a 64 row limit for MAPs which is an issue for 12k MAFs
...Top get the required 85 rows, you have to use the Create Labels button, and specify start 1500, stop 12000, step 125.

[Dirk Diggler]

Top get the required 85 rows, you have to use the Create Labels button, and specify start 1500, stop 12000, step 125.



And everytime you edit that map you will have to re-create the rows

[Beer99C5]

Thanks for the additional info guys, I get paroled Thursday, have a short two weeks home, and the driving season is drawing to a close in New England. I got 42's # injectors to put in and I am gonna play with the VE tune then roll into this MAF tuning (need the MAF for dry Nitrous).

With any luck I will have that done and get the opportunity to see what the tuning did N/A on the dyno and then get some dry nitrous dyno's/logging above the 50 shot I was limited to before I put the car up for the season.

Thanks for the help will post my results :cheers:

[Xtnct00WS6]

Could someone host that .map file? Those attachment links don't work.

...when I try and download it, I end up downloading attachment.php.

[neil]

Try right clicking and rename the file " maf_error_206.map ".
You should then be able to import it.

Regards,

Neil.

[heavyfoot]

SSpdDmon, Do you still have that spreadsheet for the MAF % correction avaliable?

[MoTruck]

Ok, this is probably a dumb question, but here goes. When you disable the LTFT's and STFT's, does this eliminate the O2 sensors? Will I have to have a separate bung in the exhaust for my LM1 sensor to do the MAF calibration? Thanks!

[TAQuickness]

MO - disabling the trims is not necessarily the same as eliminating the NB's. By disabling trims, you are no-longer adjusting fuel from the NB feed back. A few more steps would be required to eliminate the NB's all together.

But, once the trims are disabled, you can physically remove the NB's and install WB's for tuning. If desired, you could go a step further and use one of your WB outputs to simulate the NB's as well.

[MoTruck]

So, once you disable the NB, you just pull it out and leave it plugged in to the harness while you use the bung for the WB? Thanks.

[TAQuickness]

I wouldn't leave it plugged in. In fact, all 4 of mine have been sitting in a box in my garage for close to 2 years now.

[Bruce Melton]

Careful TA, he his "up North" where every day is not the same.

[MoTruck]

If I understand correctly, I need to do the following to accomplish the MAF tuning.

1. Disable the LTFT's and STFT's
2. Remove a NB to place the WB in it's place.
3. Disable the DTC (MIL) for the NB, so it can be unplugged.
4. Drive and log the MAF ben that is in this thread, like the Auto VE tuning.
5. Paste the correction factor into the MAF table.
6. Drive, Log, paste again. Repeat this step until satisfied.
7. Re-install NB, turn fuel trims back on.

Is this basically the process? I am trying to avoid having to install a separate bung for the WB. Thanks.

[TAQuickness]

that's it in a nut shell.

Bruce - I hear ya. Winter time here is rather dynamic. One day we have a high of 80, the next 50.

[SSpdDmon]

SSpdDmon, Do you still have that spreadsheet for the MAF % correction avaliable?
Yes...pm me your email and I'll shoot it over to you.

[Beer99C5]

OK that was interesting:Eyecrazy:

I made the mistake of trying to MAF calibrate using CL and I guess the trims were messing me up. Seems like I was chasing them and getting nowhere, the bens were 1.04-1.05 I would make apply the ben to the MAF table and I chased (again they were 1.04-1.05) them til I threw the Fuel Trim Rich codes, then I came back and reread the thread.

So I did a couple of logs yesterday OL with MAF enabled (using the stock MAF table) what I was expecting to see was Bens of 1.00 or so since thats where OL (no MAF) had them and what I ended up with was Bens that were .85 on the MAF Freq MAP. I averaged the readings from the two logs, and applied the .8X bens (copy w labels, paste and multiply w labels) to the MAF table. Seems like the MAF has a huge part in CL (I thought it was used 4000 RPM and higher).

I am hoping that my MAF Bens start doing what I think the should and be closer to 1.00

Am I on the right track now?

[SSpdDmon]

You can tune the MAF in open or closed loop. Closed loop, you use the trims assuming your NBO2's are somewhat accurate around stoich. But, you are limited to CL operation for tuning accuracy. Outside of CL, I don't trust the NBO2's at all. The trick is to disable LTFTs and tune off of the averages of the STFT percentages. Filter out fuel trim cells above 19 to get rid of open loop/decel, fuel trims where STFT's=0 (-0.01%<STFT's<0.01%) to get rid of laziness, and cells where TP% is changing too much. This will at least get you close.

If you want to tune with the WBO2 and BEN's, then you need to command open loop, log the appropriate pids, and again use the right filters.

Either way will work assuming two things...
1) You're doing it right...no user error.
2) Everything mechanical is working right....WBO2 or NBO2's, no intake/exhaust leaks, etc.

If your car suffers from the side effects seen on many C5's with LT headers, tuning the MAF in CL might become troublesome and cause you to chase your tail a little. If not, I'd say the tail chasing you were doing is more related to user error (maybe you didn't use your filters right?).

[Beer99C5]

Using the VE filters in the tutorial, but I am accepting less than 50 hits (I think 10) because my test circuit is limited in what I can do traffic and time wise (limited laptop battery life). No headers, pretty much stock, have a new LS6 intake vice the LS1 and catbacks. Using an LC-1 WB for logging.

[SSpdDmon]

Using the VE filters in the tutorial, but I am accepting less than 50 hits (I think 10) because my test circuit is limited in what I can do traffic and time wise (limited laptop battery life). No headers, pretty much stock, have a new LS6 intake vice the LS1 and catbacks. Using an LC-1 WB for logging.
What I did for those is look at what the surrounding cells are doing. Data points with 10 hits are useless. The stock MAF curve is fairly accurate and does well when progressively scaled. If you use the compare feature to look at your current calibrations vs. the stock table, you can see the delta %. If 3000Hz is roughly 3% over stock and 3500Hz is 3.5% over stock but you don't have that many readings for the 3125~3375Hz range inbetween....increasing them 3.25% over stock is a relatively good educated guess and will get you close. If you use the spreadsheet I built, it's easier to see the percentage change and make adjustments for the cells you didn't quite hit enough times. Ironic I push the spreadsheet given the title of this thread??? Yes...but it's extremely easy to use and it's virtually dummy proof.

My MAF table increases at .5% intervals over stock starting at 3000Hz until it reaches 6.5% over stock at 4500Hz (ie 3000 has a .5% increase, 3125 has a 1% increase, 3250 has a 1.5% increase, and so on until I get to 6.5% at 4500). Above 4500Hz, I carry the 6.5% increase all the way to the top because that's what I know the car likes at WOT. I used these settings based on watching the trims and using the WBO2 for WOT. Now my fuel trims fall into a -5%~-2% range. It's not 100% perfect nor will it ever be....but they're in a good spot. That, IMO, is the trick to tuning. If you drive yourself nuts sweating the details trying to get everything in line 110%, you'll never be able to enjoy the vehicle. In my case, the -5% LTFT's are well within the limits of the PCM for adjust during CL operation and WOT is right where I want it...so, I'm done with this tune until the next mod. :)