part throttle VE tuning with WBo2

[PJG1173]

I have a question about ve tuning with a WB. when I put hedders on my engine I had to relocate the NBo2 which caused the readings to either be slow or way off. I thought since it was a heated sensor this wouldn't be an issue, but it made tuning using BLM's a nightmare. Since then I added a WBo2 connected to my autoprom. I'm still chaseing my VE tables around. The issue I think I am haveing is AE and decel skewing the numbers tunerpro is collecting in a histogram. As I understand it if I was using blm's, learns would be disabled while AE and decel were active. Is there a way to filter this data out? how do other people tune part throttle via WB?

[EagleMark]

Man this is going to be a good thread!

But would have to know what ECM/PCM/Mask your using?

Where is your narrow band O2 sensor mounted now? Where is your Wide Band O2 sensor mounted? This can play into getting good info from WB and NB telling ECM wrong stuff...

Even if ECM/PCM has not met qualifiers for BLM learn it still spits out BLM numbers in historygram in TunerPro. Decel Fuel Cutoff (DFCO) is usually MAP kpa lower than idle kpa and I've never adjusted them, just smooth a transition from driving range. AE does not mess up BLM in $42, if you disable it the motor runs like crap and does mess up BLM so I leave it, may add to it later, but I do disable PE when trying to get VE tables close, you can get more BLM tables filled in without PE skewing the number. You probably know everything is based off VE tables so getting them right first is key. Depending on ECM/PCM you can disable DFCO usually by TPS% or temp or?

HTH get you going! :rockon:

[1project2many]

Hah! Mark replied while I was typing. I guess I talk too much. :)

Adding headers should increase response times in the range of milliseconds. I've added headers and changed exhaust configuration on several vehicles without altering the values for O2 response time and still successfully tuned the engine. If you're seeing it take longer to reach a correction point the reason isn't because the O2 sensor takes longer to respond to the oxygen content in the exhaust, the reason is because there's been a change in the amount of unreacted O2. One possibility is that the combustion reaction, which often continues in the exhaust manifold, has been slowed or stopped by the headers cooling the exhaust. Another reason could be the headers have dramatically changed the fuel requirement. If the problem is truly due to slow response to exhaust pulse I'd love to see it happening.

AE and DE are very rapid and generally occur faster than the ECM can respond through INT and BLM. These modes are directed by fast change in TPS or MAP and are designed to cover the transition caused by this rapid change. In fact, they exist because the ecm cannot respond to them quickly enough through the O2 feedback loop or the VE table programming. Power Enrichment is long term and will generally force open loop. DFCO should in theory force open loop but I've seen a couple of GM cals which do not follow this thinking.

What I've done for tuning when the BLM and INT are making wild swings during load / unload conditions is to force min and max BLM and INT to 128 and use the O2 reading directly for my tuning. The engine will run badly if the VE is way off so you won't be able to skip adjusting the cells which require adjustment. VE and timing need to be addressed first so find a hill or a way to apply a load under steady state throttle then make corrections one cell at a time using data recorded on the scantool / laptop. If you suspect AE and VE are causing problems, zero them out or disable them. You'll stumble, spit, and cough when moving the throttle quickly but if the VE is correct, as you slowly move the throttle to a specific position or load value the engine will operate correctly and O2 should hover near .450 volts (I actually shoot for .48 - .50 V) during steady load operation. From this point tuning AE and VE is really a task of maintaining O2 voltage near .450 as the throttle is opened or closed. Remember that steady state VE is the main MAP and that AE and VE are in place to make up for air density changes which the ecm can't handle through the feedback loop.

HTH

[PJG1173]

Wow I wish I could have got this fast of a response on the other forums I posted on. I am running 16168625 $E6 mask. I have dfco disabled and I set the TPS % to 98% for PE. I have left AE alone so far. one of my big issues is this is my first time tuning and I changed almost everything on this engine. litteraly the only thing original is the crank, rods, and injectors. so yes the fuel and timing requirements have drastically changed from stock. Like an idiot I dove into this project as my first tuning experiance. I have hardly any logs from when it was stock.

The NB is mounted in the collector on the passenger side hedder and the WB is on the drivers side. thats about 8 - 10 inches further down the exhaust tract than stock. I am currently in OL and it runs decent. I have been solely tuning off the WB readings not the NB. since I am in OL there are no BLM's to compare to the WB. I have the WB simulating NB to the ecu but once I enable closed loop things start going down the crapper. I have been meaning to hook the NB back up to see what it reads now gas isn't pouring out the tail pipes anymore. my ve table looks like a mountain range. I idle at 45kpa and at cruise I'm around 80ish kpa. I tuned idle using the same theory I use for a carb. i increased or decreased fuel until i hit the highest vacuum reading. the WB reads 13:1 @ 650 RPM and it idles nice and smooth. to me that seems rich for idle, but the exhaust does not smell rich. How do I compensate for the hedders/NB location or should I even have to? could it be something else like too little of AE that is messing with my numbers/ve table?

May be (more like probably) I'm doing something wrong here.
included is a graph of my ve table from an old tuning session and the log it generated. I haven't worked on the truck in a couple weeks so I don't have anything more recent with me. I also attached a copy of my current bin if anyone wants to look at it.


Here's a quick breakdown of the mods.
93 chevy s10 ex cab
01 4.3 block bored .030 over
2 valve relief flat top graphite coated hyperT pistons
9.8:1 compression
edelbrock 2114 intake
stock injectors @ 15psi
adjustable FP regulator
TBI with ultimate tbi mods
TBI spacer
ported and polished vortec heads
back cut valves
comp cams 266hr
arp rocker studs
1.52 roller tip rockers
hedman long tube headers
cold air intake
built 4l60e
2075 stall converter
moates.net autoprom
Crane HI6 CDI ignition
4.88 gears
35x14.5x15 TSL boggers

[gregs78cam]

First off I would tell you to hook up the real NBO2, and disconnect the simulated one. I traced a few of my issues to the simulated NBO2 output. Then get some logs recorded in Closed loop.

[EagleMark]

Wide Band O2 simulators are not all they are cracked up to be... like Greg said, but your saying your open loop? So maybe you are open loop with WB/NB simulator?

But your doing good for a first tune on a built motor! :thumbsup:

If you are open loop and you ever want to run an O2 sensor I would get your WB AFR closer to 14.7 to 1 because that's where the Narrow Band O2 sensor is going to want to put them. This will get your VE table close. Then run closed loop off O2 sensor and finish VE with BLMs.

You may get a better vacuum at idle by bumping it up to 700-750. Or even run open loop idle. When you get idle up do a Minimum Air Adjustment on throttle blades and get IAC count to around 10 -20 at idle. More air over throttle blades pushing fuel in. Just give IAC a little control of idle.

Then use your Wide Band for WOT runs and get the AFR down to 12.5 - 12.8 to 1 AFR. Unless it's just for race then tune it all that way. Look in the ECM Information thread for a PE Delay zip file which explains the barrage of qualifiers needed to be met to enter PE including up to a 70 second delay?

[PJG1173]

I'll hook the NB back up and set the bin back so I will go into closed loop and see what happens. that was the idea all along just without using the NB. I just have been having a heck of a time getting the afr close to 14.7 the best I can get is 14.0 - 15.7. all I have to do is find time...

[1project2many]

Setting proper timing goes hand in hand with setting proper fueling. But timing can affect the observed O2 readings. More advance tends to show up as a richer engine while less advance shows up as a lean engine. This can be more pronounced on engines with a larger cam. The exhaust valve opens while combustion is still occurring so more advance means more O2 is reacted before the exhaust pulse passes the O2 sensor. One very good method I've found to attain best timing is to use an oil temperature gauge. As timing approaches best power you'll find a small range of settings that don't seem to make a difference in the seat of the pants dyno or in the O2 readings. But if the timing is slightly over advanced, then combustion pressure will fight against piston motion without causing knock. The wasted energy has to go somewhere, and the very first place it ends up is in the oil film between the rods and crank. Over advanced timing will cause oil temp to rise. Back off timing a degree or two and see if the temp stops increasing. It may not decrease unless you have an oil cooler, but at least it will level out and stop rising when timing is good.

[EagleMark]

I looked up his cam specs, it's fairly mild and ECM freindly. Actually without all the other work he could have probably just done a BLM tune.

Always thought the anaogy of oil temp and over to much spark advance was piston heat? Since oil is the only thing cooling a piston.

[dave w]

$E6 has two VE Tables and two Spark Tables; Near Idle / Off Idle. Typically, I use Excel to sort out the data into near idle / off idle. I also use Excel to sort out Open Loop / AE / PE data. TunerPro RT histogram is OK to use, and can get the tune close. In my humble opinion, the TunerPro RT histogram will not get the tune in the ball park, but somewhere out in the parking lot of the ball park. I'll use the TunerPro RT histogram to get the BLM's betweem 118 ~ 138, then I use Excel to get the tune down on to the playing field of the ball park.

dave w

[PJG1173]

DaveW,
do you have excel doing the calculations/filtering for you or creating a piviot table like the tunerpro histogram? Mind sharing if you do? This is what I tried to do with mine but I am having a hard time figuring out how to put it in a format that coinsides with my VE tables. I can filter all the stuff I don't want to look at out, and tried create a pivot table but don't really know what to do from there. I have a spreadsheet that I copy my off idle ve table to and the tunerpro histogram and it will do the calulations for me but its not filtered. I would love to be able to export a log filter the data and then reimport it, but it doesn't look like TP supports that. do you know a way? I've attached the excel sheet I made to do my calculations. I saved it to 2003 format since the forum will not accept office 2007 file extensions, so hopefully all the formulas still work.

[PJG1173]

Setting proper timing goes hand in hand with setting proper fueling. But timing can affect the observed O2 readings. More advance tends to show up as a richer engine while less advance shows up as a lean engine. This can be more pronounced on engines with a larger cam. The exhaust valve opens while combustion is still occurring so more advance means more O2 is reacted before the exhaust pulse passes the O2 sensor. One very good method I've found to attain best timing is to use an oil temperature gauge. As timing approaches best power you'll find a small range of settings that don't seem to make a difference in the seat of the pants dyno or in the O2 readings. But if the timing is slightly over advanced, then combustion pressure will fight against piston motion without causing knock. The wasted energy has to go somewhere, and the very first place it ends up is in the oil film between the rods and crank. Over advanced timing will cause oil temp to rise. Back off timing a degree or two and see if the temp stops increasing. It may not decrease unless you have an oil cooler, but at least it will level out and stop rising when timing is good.

unfortunatly I don't have an oil cooler or a way to check oil temp right now. timing is one thing I am going to tackle next. I have messed with it a little since it ran like doo with the stock SA tables so I copied the SA table out of a 98 vortec 4.3 to start with. the butt dyno is all I have right now the local dyno shop wants $125 a hour to run it on the dyno.

[EagleMark]

DaveW,
do you have excel doing the calculations/filtering for you or creating a piviot table like the tunerpro histogram? Mind sharing if you do? This is what I tried to do with mine but I am having a hard time figuring out how to put it in a format that coinsides with my VE tables. I can filter all the stuff I don't want to look at out, and tried create a pivot table but don't really know what to do from there. I have a spreadsheet that I copy my off idle ve table to and the tunerpro histogram and it will do the calulations for me but its not filtered. I would love to be able to export a log filter the data and then reimport it, but it doesn't look like TP supports that. do you know a way? I've attached the excel sheet I made to do my calculations. I saved it to 2003 format since the forum will not accept office 2007 file extensions, so hopefully all the formulas still work.Like this one!
http://www.gearhead-efi.com/Fuel-Injection/showthread.php?342-E6-OE-OD-31-42-VE-Fixer-Spread-Sheet

[PJG1173]

kind of. I saw that one a while back. I'd like to be able to take the raw logs where I could filter out stuff i didn't want based off of the flags and what not, then put it into that format.

[EagleMark]

TunerPro can export all or only what you choose to excel files.

[dave w]

DaveW,
do you have excel doing the calculations/filtering for you or creating a piviot table like the tunerpro histogram? Mind sharing if you do? This is what I tried to do with mine but I am having a hard time figuring out how to put it in a format that coinsides with my VE tables. I can filter all the stuff I don't want to look at out, and tried create a pivot table but don't really know what to do from there. I have a spreadsheet that I copy my off idle ve table to and the tunerpro histogram and it will do the calulations for me but its not filtered. I would love to be able to export a log filter the data and then reimport it, but it doesn't look like TP supports that. do you know a way? I've attached the excel sheet I made to do my calculations. I saved it to 2003 format since the forum will not accept office 2007 file extensions, so hopefully all the formulas still work.

I posted some information in this thread, http://www.gearhead-efi.com/Fuel-Injection/showthread.php?465-Help-with-VE-tables.-Only-been-tuning-for-a-week-or-so../page4 about page 4 for sorting the excel file into cells like the VE table. The ROUNDING formulas are my method.

If you have questions, please ask.

dave w

[PJG1173]

that is what I've been searching for thanks!!

[PJG1173]

that was the missing piece I needed. now to automate it...
thanks again!

[dave w]

that was the missing piece I needed. now to automate it...
thanks again!

Nice work on the Excel Filter. Automate the "IF < > Kpa vs RPM = Average WBo2?

dave w

[PJG1173]

actually I was thinking of maybe using MS access so I could import the logs into a table and set up a query to make a cross tab that I could paste back into tunerpro.

[dave w]

actually I was thinking of maybe using MS access so I could import the logs into a table and set up a query to make a cross tab that I could paste back into tunerpro.

I had not thought about trying MS Access. Basically I was thinking of automating the MS Excel steps show in the screen shot below. The pic below is for NB BLM, but the same process applies for WBo2 as well.

dave w

[PJG1173]

I think I amost have this working in MS Access. the round function works similar to excel. with access you don't have copy paste the formula around. so far I have a query that runs the round function against a table linked to the tunerpro export that feeds a cross tab query. the cross tab outputs a table in the same format as the ve table. I'll post a copy of it when I get more of it done.

[dave w]

I think I amost have this working in MS Access. the round function works similar to excel. with access you don't have copy paste the formula around. so far I have a query that runs the round function against a table linked to the tunerpro export that feeds a cross tab query. the cross tab outputs a table in the same format as the ve table. I'll post a copy of it when I get more of it done.

Very interesting! I am almost illerate with MS Access compared to what I know an do with MS Excel. I would greatly appreciate a copy of the MS Access file you firgure out!:thumbsup:

dave w

[EagleMark]

Dam! I could have taken courses on those years ago for free, but thought they were useless to a GearHead! :mad1:

[PJG1173]

davew this may help with some of the manual sorting of MAP in your excel sheet =IF(E6>40,10*ROUND(E6/10,0),5*ROUND(E6/5,0)) this will round any map over 40 by 10 and anything less by 5.

[dave w]

davew this may help with some of the manual sorting of MAP in your excel sheet =IF(E6>40,10*ROUND(E6/10,0),5*ROUND(E6/5,0)) this will round any map over 40 by 10 and anything less by 5.

Thanks! I'll give the "IF" test a try. :thumbsup:

dave w

[PJG1173]

here is what I have so far. there are instructions in the zip file. I'm still working some bugs out and not sure if I am going to keep linking the tunerpro logfile or just do an import. it seems like when you refresh the linked table it balloons the db file. There is still alot I want to do with this, like having it generate a new vetable based off the old one. also if you have a slow computer this may hog some resources when it is doing its calculations on a big logfile.

[EagleMark]

Thanks for sharing your work with us! :thumbsup: It's great to have new guys jump in with their skills!

Can't wait to try it out and learn some new programs (excel-access) you guys are way ahead on the learning curve there.

You are aware you can export the data log from TunerPro with just selected parameters? Acquisition, Export Log, then click selective and choose what you want.

http://www.gearhead-efi.com/Fuel-Injection/attachment.php?attachmentid=1545&stc=1&d=1328045564

[PJG1173]

yes, but it only allows you to select the item logged not values in that item matching a certain criteria. the SQL query that is in the database will do that. currently I have it filtering out AE (active), PE (active), anything less than 2% tps, idle ve flag (on), Dcel enlenment flag, DFCO, and idle spark flag. it leaves just the records that were recorded with no other adders active at the time.

[dave w]

yes, but it only allows you to select the item logged not values in that item matching a certain criteria. the SQL query that is in the database will do that. currently I have it filtering out AE (active), PE (active), anything less than 2% tps, idle ve flag (on), Dcel enlenment flag, DFCO, and idle spark flag. it leaves just the records that were recorded with no other adders active at the time.

That's how I "Sort" the data log with Excel. The tune is greatly improved when filtering AE / PE and the other parameters listed above.

[dave w]

I've been looking at the MS Access sheet posted above. I've got some learning to do.

dave w

[PJG1173]

so do I, learning as I go.

[PJG1173]

here is the latest. I would have used it to process my logs from last night but my laptop died before I could save the log:mad1:, so all was lost. what I did was set up a few more linked tables to copy the vetables out of tunerpro and paste them into the linked files in the working directory. Access then will calculate what your new values should be based off the logfile. I haven't set it to average the old and new tables together yet, but you get the idea where I'm headed with this. next I will work on getting it to do the BLM's and maybe chart PE & AE. instructions are in the zip file enjoy.

[PJG1173]

just an update, I found a typo in one of the formulas in the above file. I havent had a chance to work on it much. I've been pretty busy lately trying to get my daughters jeep running before she gets her permit. I did some datalogging the other day and noticed that I was at almost 8ms pulse width and over 72% DC at 5200 rpm with ve in the low 90's the WB was showing 14.7 and PE was not active. is this an indication that i need more fuel pressure or bigger injectors? I don't think that lowering the injector constraint will solve this, or maybe I don't understand how it will.

[EagleMark]

Not sure about your other questions but seems PE should be active at that RPM? Does your PE ever work?

[PJG1173]

not right now. I have it set to 98% tps. I was running about 80% throttle trying to hit cells up higher in the table.

[PJG1173]

so do you think this will be fine once PE is enabled?

[EagleMark]

That's a pretty vague question?

But if you got your VE tables in order then yes enable PE, turn off PE delay time, set other PE factors the way you want like TPS %, PE AFR and PE engine RPM. Makes a big HP/Tourque differnce to have PE!

[PJG1173]

Ok since I haven't been able to work on my truck because of the weather I've done some more work on my access 2007 based VE calculator.

there is an Instructions.txt included in the zip file. I would like someone else to try it and see if there are any bugs in the calculations. I've added the ability for it to calculate ve off of BLM as well as from WBO2 readings. I threw some other odds and ends in it too.

[dave w]

Ok since I haven't been able to work on my truck because of the weather I've done some more work on my access 2007 based VE calculator.

there is an Instructions.txt included in the zip file. I would like someone else to try it and see if there are any bugs in the calculations. I've added the ability for it to calculate ve off of BLM as well as from WBO2 readings. I threw some other odds and ends in it too.

I have a long weekend because of Presidents Day. I'll try to get one of my old .bin ~ .xdl files and see how it goes Monday.

dave w

[PJG1173]

dave did you get a chance to check this out?

[dave w]

dave did you get a chance to check this out?

I had to figure out that Win7 wanted me to be an administrator on the files that were in the .zip file. I enabled the Macro, and looked at the attached files in the .zip. I have not experimented with any log files as of yet.

dave w

[dave w]

I think I have firgure out how to get the Access spreadsheet to work. I have attached a .zip file that I've been working with. I pasted in a new tplog.csv file, but I renamed the original tplog.csv as tplog_01.csv. I'll need to verify the Access spreadsheet vs. the Excel spreadsheet to see if there is a difference between the "NEW" VE Tables. Basically, with WIN7 I had to make myself the administrator of every file in the vecals folder to get the Access Macro to work.
dave w

[PJG1173]

I run XP. I don't have anything with enough prossesor to run win7. with xp all I had to to do was add the folder in the access trust center.

[dave w]

Ok, Maybe better late than never to reply back to a post.

The data in the attached files are from an engine $OD currently being tuned. It's real world data! The EGR has been disabled, and as a result a good tune needs to be corrected. If there are any mistakes, well I guess we will learn.:yikes:

I have used the VECALCS and my Excel Macro on the same data log. The results / comparison can be found in the attached Access vs Excel spreadsheet .zip file. The results are very similar, but there are some difference. I think with the VECALCS doing the sorting and averaging, then using the attached E6~OD~OE~31 AFR VE Correction Table for TunerPro.xlsx spreadsheet would yeild a decent WBO2 tune. I think it would be a good idea to disable BLM update, because WBO2 data and BLM data with contradict with one another.

dave w

[EagleMark]

Run Open Loop with WideBand and use Lamda with a new spreadsheet?

[dave w]

Run Open Loop with WideBand and use Lamda with a new spreadsheet?

I'm thinking closed loop is OK, with a WBO2 tuned VE Table. The PCM will still use the VE Table, and adjust INT with the NBO2. The idea is to keep the BLM update from changing / re-learning the VE Table. The WBO2 data is more accurate, so calibrating the VE table with WBO2 data is more accurate than BLM update.

The challenge is being able to get enough WBO2 data, and how to sort and average the WBO2 data.

dave w

[EagleMark]

Also kill power to PCM to clear the BLM learn already done.

Disable should be easy "Minimum Temperature for BLM Update Enable"

BLM learn doesn't change VE tables, there is a BLM learn table which stores BLM. If you look at boundries you can see how the low end needs to be changed for a cammed engine or idle and off idle/low RPM end up in same cells. Ever notice BLM cell number in data logs?

Lambda is way easier then AFR. Lambda = 1.00. Rich is less then 1.00 and lean is more then 1.00 So if Lambda reading is .82 then VE cell X .82 = new VE.

[PJG1173]

Ok, Maybe better late than never to reply back to a post.

The data in the attached files are from an engine $OD currently being tuned. It's real world data! The EGR has been disabled, and as a result a good tune needs to be corrected. If there are any mistakes, well I guess we will learn.:yikes:

I have used the VECALCS and my Excel Macro on the same data log. The results / comparison can be found in the attached Access vs Excel spreadsheet .zip file. The results are very similar, but there are some difference. I think with the VECALCS doing the sorting and averaging, then using the attached E6~OD~OE~31 AFR VE Correction Table for TunerPro.xlsx spreadsheet would yeild a decent WBO2 tune. I think it would be a good idea to disable BLM update, because WBO2 data and BLM data with contradict with one another.

dave w

I think one of the differences is that VE calcs filters out any records below a certain temp. I would have to look at the query statement to see what I set it at. My phone dosn't do spreadsheets too well so I'll look at it when I get home tonight.

[dave w]

I think one of the differences is that VE calcs filters out any records below a certain temp. I would have to look at the query statement to see what I set it at. My phone dosn't do spreadsheets too well so I'll look at it when I get home tonight.

There are some filter differences in the Excel Marco vs. Access filters. The Excel Marco filters open throttle / closed throttle by 3 MPH / 2.7 % TPS. The Access spreadsheet is using the idle timing flag for near idle data. Closed loop data only in the Excel Macro (regradless of temperature set point), so the temperature setting is a possible difference in filtering in Access. The Excel Marco does not filter DFCO, but the Access spreadsheet does. Both Excel Marco and the Access spreadsheets filter AE / PE. Overall, I think either spreadsheet will filter / sort the data logs for a good tune.

I noticed an unexplained difference in the VE Table changes Access vs. Excel?

dave w

[dave w]

I've done another summary comparision between my excel macro vs the access spreadsheet. The results are very similar except for a few cells.:mad1: I can't step into access like I can excel to see the data processing.

dave w