PCM Diagnostics & Tuning - EFILive/Flashscan Guys - No More MAF Spreadsheets?

Incase you haven't been on the efilive forum lately, check this out:

post by black02ss (http://www.efilive.com/forum/viewtopic.php?p=13008#13008)

I'll be trying this in the next week or so

I just got my flashscan, :) Sounds like it might work, I might try that methode after I get my VE table dialed in. :)

HPtuners is supposed to be able to do this whenever 2.0 comes out.

Yeah...I tried to get them to do another interim release, but to no avail. I'll just have to wait for 2.0. Guess I'll see how my car does with Speed Density for now.

I'm just waiting on my WB so I can see how far off my VE and MAF tables are.

Good luck on the wait.

Yeah, since I have my wideband, I can at least get my VE table inline. From what I can tell though, it seems as if the MAF is what is really screwing everything up. My VE table didn't change much when I messed with it a couple weeks ago.

MAF can be a bitch. How are you tuning it?

For right now, I'm not going to. I'm just gonna run Closed Loop, Speed Density for right now. I tried doing the MAF at the same time as my VE last time, and the trims were way off (I was using RedHard's spreadsheet). So I just put everything back the way it was so I could drive my car to work.

This table is what I need to make it not so damn hard, especially since I'm gonna be running a descreened 85mm truck MAF starting this weekend.

You can't do MAF and VE at the same time.. Its like chasing your tail Steve. :lol:

Guess that didn't come across right...I was adjusting my VE in Open Loop SD, but still logging MAF Hz and Dynamic Airflow, and using RedHard's spreadsheet to calculate my new MAF table. That sound better? :)

The only time I have found that loging those items is only AFTER the Ve table is dialed in correctly.

i do MAF and VE at teh same time in SD closed/open loop, no problem. the reason you can do it is because while MAF is ignored in SD, it is still logged. so you gather up your MAF untainted VE values, but at the same time you gather MAF data. you do it few times, and when your VE is nice and smooth, then i combined all the maf data and run a huge calibration with lots of points, yielding a new nice calibration. very simple, very reliable, and you save time.

if you got problems with my spreadsheet, let's talk, email/aim/yahoo is the best way, they're all in my profile

I personally don't see how you can take the data from the MAF and use it after you have the VE table dialed in. When you alter the VE table, it will in turn change the values that are represented by the MAF. To me it sounds like circular reasoning. With the method I use, you dial in your VE table to where it needs to be. Its simple, you don't use any values except MAF Hz and the BEN Factor(commanded afr/Actual AFR). Then you take and apply your filter to remove any throttle transitions. This total process can be performed and give you a calculated MAF within 30mins. No need to log several times. The reason you don't log while you are dialing the Ve table is is because the values will change when you adjust the VE table causing the outcome to be off as well.

oh i know it's weird. but you know what? i did them together, checked it out, put stock tune back on, did them separately, and ended up with very close results. do i get it why? no. but it seems to work, i've done it on many cars, and they run fine. the only difference is that doing VE/MAF at the same time takes half the time.

if i was to guess why, i'd say that the data for both are so dirty anyway, that our margins of error are wide enough that the changes we get on the account of doing both at the same time result in changes smaller than our acceptable level of noise. since we can't verify it, (aka create a system with no noise) i'd say just go with it and enjoy the extra hour you saved.

Now, how do you go about filtering out the throttle transitions when using the MAF %Error histogram? I didn't think about that, but that would nearly make this histogram ineffective...

Now, how do you go about filtering out the throttle transitions when using the MAF %Error histogram? I didn't think about that, but that would nearly make this histogram ineffective...

One of many reasons to switch to EFILive! I'll be there in about a week and can't wait.

So EFI Live can filter the throttle transitions for you when using this Histogram? Is it already a part of this histogram?

Live has a filter that can filter out any fields that you want. If you want to filter out throttle transistions more then 5% per 100ms, you can do so. You can add as many of these together as you wish. You can make one for filtering out WOT data, KR, Part throttle, the choices are endless. You are also allowed to remove data in the scanner that "YOU" feel is unneeded. Weather it be 1 frame or 50, you can manually remove them. Its all in the hands of the end user.

Efilive can filter any criteria your mind feels is important and you can use the filter to add more than one pid

eg

Filter tps where tps changing more than 5% or kr >0 or rpm > 6800 (rev limiter) and the list goes on

You are also allowed to remove data in the scanner that "YOU" feel is unneeded. Weather it be 1 frame or 50, you can manually remove them. Its all in the hands of the end user.


Please tell me how to do this. I can't figure out how to even see the individual data points, for STFT for example, without going to dashboard b.

You will only be able to see the individual points in the dashboard using a scanner/dash. You can go frame by frame where as the MAPS show high/low/average.

Back to the the point at hand - Black - I appreciate your, and everyone elses, efforts.

Fedex and weather permitting, I will be "autotuning" my VE and MAF tables this weekend. I've also started a complete start-to-finish AutoTune document for EFILive/Flashscan. Once I get done with my tune, and proof read the document, I'll post it up.

Back to the the point at hand - Black - I appreciate your, and everyone elses, efforts.

Fedex and weather permitting, I will be "autotuning" my VE and MAF tables this weekend. I've also started a complete start-to-finish AutoTune document for EFILive/Flashscan. Once I get done with my tune, and proof read the document, I'll post it up.

The autotune works great. How are you logging your SD? Do you just drive around like normal or do you need to do anything special are far as stop and go or no WOT,etc?

Bill
:hail:

The way I use the AutoTune feature is as follows. I disable closed loop and change my Open Loop AFR table to 14.63 from 122* up and 70kPa down. From 122* up and 75kPa up the AFR is 12.96. I disable PE by changing all the values to 1.00(14.63) and a MAP to enable of 105. This makes sure PE isn't entered. I then do a log. I have open the MAP with my AutoTune PID and hit as many cells from 800 to 4000 rpms 15-105kPa. I try and get at least over 50-100 cells in each point before I am done. After this, open the log in the scanner. Apply the AutoTune/Ben Filter where it removes data where the TPA moves more then 5% per 100ms. I also enable the "low count" filter in the MAP itself and filter out all data where the value is less then 30. For WOT runs you will need to settle for something in the 10 range unless you want to do numerous WOT pulls. After that I copy the table with labels. Open the Tuner Ve Table and paste by factor with labels. Flash to car and re-do. Auto tune only works when you have a wideband, so don't try it without one. Also, disconnect your MAF from the car before attempting this and make sure the car is at full operating temp.

The way I use the AutoTune feature is as follows. I disable closed loop and change my Open Loop AFR table to 14.63 from 122* up and 70kPa down. From 122* up and 75kPa up the AFR is 12.96. I disable PE by changing all the values to 1.00(14.63) and a MAP to enable of 105. This makes sure PE isn't entered. I then do a log. I have open the MAP with my AutoTune PID and hit as many cells from 800 to 4000 rpms 15-105kPa. I try and get at least over 50-100 cells in each point before I am done. After this, open the log in the scanner. Apply the AutoTune/Ben Filter where it removes data where the TPA moves more then 5% per 100ms. I also enable the "low count" filter in the MAP itself and filter out all data where the value is less then 30. For WOT runs you will need to settle for something in the 10 range unless you want to do numerous WOT pulls. After that I copy the table with labels. Open the Tuner Ve Table and paste by factor with labels. Flash to car and re-do. Auto tune only works when you have a wideband, so don't try it without one. Also, disconnect your MAF from the car before attempting this and make sure the car is at full operating temp.

Thanks for the info. I have a LC-1 setup with FlashScan AutoTune. It does work great. Can you have numbers over 100 in the VE table? I have a few cells over 100. I did start loging before car was at full operating temp. I will do some more logging on Monday. Does stop and go hurt the numbers?

Bill
:hail:

max is 500 although no one should go that high. My ve table has values if 118 for the cam i am running

max is 500 although no one should go that high. My ve table has values if 118 for the cam i am running

What cam are you running? I am running a Torquer 233/233 .595 .595 112+2

Bill
:chug:

That is normal if you are seeing at at a high RPM and High MAP. It doesn't really matter when you start the log as you can just add a filter and remove all data where the ECT was below say 178*F. Start and stop driving is fine. That is what you want to be able to hit all of the cells in question. Also make sure you are using the Base Efficiency Numerator Filter. You could just add a filter to exclude all cells with ECT of 178*F and less.

That is normal if you are seeing at at a high RPM and High MAP. It doesn't really matter when you start the log as you can just add a filter and remove all data where the ECT was below say 178*F. Start and stop driving is fine. That is what you want to be able to hit all of the cells in question. Also make sure you are using the Base Efficiency Numerator Filter. You could just add a filter to exclude all cells with ECT of 178*F and less.

Thanks. I did check out the filter and it works great.

Bill
:hail:

Black - Throw me a bone. I can't seem to figure out how to adjust the low cell count filter. Mine wants to hide cell counts of 0.

edit the map properties and click on the empty cells tab

Okay, here's a poser for all y'alls...and this is for those that have Long Tube headers only.

Does anyone else see the inherent problems with tuning your MAF table with a wideband? If you are running in closed loop with the MAF, then you are using narrow band O2 sensors for feedback. In fact, that is pretty much the only feedback your engine has. It is using these to maintain stoich.

So if you are tuning your MAF sensor off of a wideband, you are giving the PCM conflicting information. It is kinda like when I went back to closed loop with my speed density setup. I know the car should be running right at 14.62:1 (since my VE table was spot-on in open loop), but it doesn't. It runs right around 15:1. But the PCM thinks it is running at stoich, because that is the only feedback it is getting (from the narrow band O2s that is).

Anyone have any thoughts on this?

edit the map properties and click on the empty cells tab

Duh.

Thanks

Okay, here's a poser for all y'alls...and this is for those that have Long Tube headers only.

Does anyone else see the inherent problems with tuning your MAF table with a wideband? If you are running in closed loop with the MAF, then you are using narrow band O2 sensors for feedback. In fact, that is pretty much the only feedback your engine has. It is using these to maintain stoich.

So if you are tuning your MAF sensor off of a wideband, you are giving the PCM conflicting information. It is kinda like when I went back to closed loop with my speed density setup. I know the car should be running right at 14.62:1 (since my VE table was spot-on in open loop), but it doesn't. It runs right around 15:1. But the PCM thinks it is running at stoich, because that is the only feedback it is getting (from the narrow band O2s that is).

Anyone have any thoughts on this?

Don't confuse OL or CL with SD. You can run CL without your MAF and still have feedback from your O2's, which allows you to use fuel trims. That is SD - no MAF.

If you force the car into OL by flooring it or by disabling CL with your software, it can operate with the MAF or without it (SD).

That being said, you can tune the MAF based on LTFT/STFT's for part throttle application where the PCM is commanding stoich. But, you'll need the wideband to tune the MAF for OL situations.

If you're seeing false readings from the NBO2's because you have LT headers, try upgrading your NBO2's to the rear vette O2's. They run a little hotter from what I've read on here.

The only thing that still confuses me is you say you're at 15:1. IIRC, LT headers tend to create a richer AFR due to the O2's being moved further back to a slightly cooler location. If that's the case, you should be seing 14:1, not 15:1. So, I'd suggest double checking your wideband setup. When is the last time you did a free-air calibration? Are you sure the voltage being reported from the sensor is being interperted correctly by your software?

I'm well aware of the differences between OL SD, CL SD, and CL MAF. My point is, the PCM is getting false readings from the NBO2s. I do have rear vette O2s on there, and they are still errant. I just think it is something that can't really be done well because of the obvious shortcomings of NBO2s when running headers.

I'm pretty happy with my compromise...it runs lean at cruise, but is dead nuts on at WOT. I was happy enough with my results to not consider running a MAF again...I'll just have to keep an eye on my tune as the weather changes.