Newbie looking for advise with first tune

[Dieselman]

Hello all,

I am new to this tuning game and would just like to run everything past you guys before I make to many changes. My car is a VE Holden Calais with a L98 and A6

I have started by following Swingtan's tutorial on VVE setup.

I did some logging to set up the IAT injector flow modifier table. The first log is factory and the second one is a modified tune. The final tune is running a lot flatter but is still a tad rich, Should I keep trying to get it closer to commanded or will it be OK???

Any help greatly appreciated

Ben


10090

10089

10091

[swingtan]

Running rich is not a big issue, at least not when setting the IAT correction. All you are trying to do is get the WB signal to indicate a constant combustion AFR over a wide range of IAT readings. You are pretty close looking at the second log, though there is a bit of a step at 66' IAT, but over all it's quite good. You can get the correction closer if you want, then go on to work on the VVE its self. This will bring the WB AFR back towards the commanded.

Simon.

[Dieselman]

Thanks Simon

I didn't know if the AFR had to match commanded or just be flat all the way though the temp range.

The step was when the Hi speed fans cut in :)

[swingtan]

Yeah, I noticed that, as the MAP and air flow changed. But the AFR line before that was on a steady richening up slope, not a lot though.

[Dieselman]

Is there anyway to adjust between segments ie in increments of 5 degrees instead of 10???

[swingtan]

Nope. I just did the settings at each 10 degree increment and the ECM will blend between them.

[Dieselman]

Ok that is easy

Thanks Simon

[Dieselman]

Ok did some logging tonight and I have now replayed the log but can't get the BEN's to come up on the map???

Am I missing something here or not :help2:


10100

[swingtan]

The BEN data is there. You want "BEN Factor Bank 1, Serial Wideband, E38 style". Make sure it's selected in the PID list and then you should be able to put it on a chart / map.

[swingtan]

I should also add...


Try to keep a steady throttle when logging for tuning the VVE.
If you need to add / remove throttle, do so in a gentle fashion. No fast movements!
Get as much varying data as you can, hold it in first and take it up to 6000 RPM with very little throttle. Hold it in a higher gear and load it up. change up / down gears when cruising to mix up the data a bit.
Don't forget to filter the transient states out!
Get lots of data and apply, then log lots of data again.


Simon.

[Dieselman]

I have the map up with the cell counts but when I go to the average value screen it is all zero's????

10101

10102

[Dieselman]

I should also add...


Try to keep a steady throttle when logging for tuning the VVE.
If you need to add / remove throttle, do so in a gentle fashion. No fast movements!
Get as much varying data as you can, hold it in first and take it up to 6000 RPM with very little throttle. Hold it in a higher gear and load it up. change up / down gears when cruising to mix up the data a bit.
Don't forget to filter the transient states out!
Get lots of data and apply, then log lots of data again.


Simon.

OK I will be nicer on the throttle :)

[swingtan]

It looks like you don't have the PID selections right and I think from memory that the default BEN map is incorrect. Try this...


Open ScanTool with the log file of choice.
Open the "PID's" tab or press <F8>.
Under the "System" drop down box, select Calculated.
Double click the "Ben Factor Bank 1, Serial Wideband, E38 Style" PID.
Allow the program to re-plot the charts and maps.
Open the MAPs tab or press <F11>.
Go to your BEN map.
Right click the map and then select "Map Properties".
Select the "Data" tab.
In the "Parameter" drop down box, select "(0) BEN Factor Bank 1, Serial Wideband, E38 style (factor ).
Open the TuneTool with your current tune and go to B8101.
Press <F2> to bring up the VVE table cells and then click on the grey box in the top left corner of the cell window to select all cells.
Right click and "copy with labels" or press <SHIFT>+<CNTL>+C
Go back to ScanTool
Select the "Column" tab
click "Paste Lables"
Select the "Row" tab
Select "Paste Lables"
Select the "Cells" tab.
Click on the "Constrain cell size" box so it is checked.
Set the minimum width + height to 30 + 10.
Select the "Empty" tab.
Note that you can check the "Hide Empty Cells" box to remove data from the MAP. This can be used to "hide" cells with only a small number of samples and helps to remove random spurious data that can adversely effect the VVE shape. It's good to hide cells with less than 50 "counts", it's probably acceptable to hide cells with less than 10 counts. You can turn off the hiding of cells if you are working on a specific issue.
At the bottom right, select "Save As" and choose a new name for your map.
Select "Save"
select OK



You should now have a working map for BEN factor correction. If you want to compare maps between enabling and disabling the "hide empty cells, you can make two maps up and just switch between them.

[Dieselman]

Thank you very much Simon :cheers:


The map works a treat. I hope you are ready for more stupid questions in the future though!!!

[swingtan]

always here to help ;)

[Dieselman]

OK time for another stupid question. When BBL how do you put BEN's in the PID selection???

I looked through the list but could not find it :confused:

[swingtan]

BEN's are a calculated PID, so don't work directly on the V2. The data to produce the BEN's are still recorded on the V2 and you can then select them in ScanTool. Actually, I think you can "preselect" them in the BBL setup so they are there when you open the log in ScanTool.

Simon.

I just checked ...

Open V8 Scan&Tune
Select <F5> BBX
In the "Selected PID Lists" window, select the controller / PID list you want.
To the right of the "Selected PID Lists" window, click "V7"
In the "Pre-Selected Calculated PID Names", add the BEN PID you need.
You can also name the predefined dash page you want to use.

[Dieselman]

Thanks Swingy :rockon::rockon: It is all good now

Cheers


Ben

[Dieselman]

Hello everybody,

What does the filter setup do in this screen shot???

Thanks in advance :grin:


10598

[swingtan]

My understanding is.... Nothing for serial Wide Band data.

For Analogue WB signals, you can average the analogue over a number of samples. The wording of the text is "Analogue/Digital" which reads to me (and many others) as Analogue <and / or> Digital, but it should actually be read as, Analogue "TO" Digital,

Simon.

[Dieselman]

My understanding is.... Nothing for serial Wide Band data.

For Analogue WB signals, you can average the analogue over a number of samples. The wording of the text is "Analogue/Digital" which reads to me (and many others) as Analogue <and / or> Digital, but it should actually be read as, Analogue "TO" Digital,

Simon.

Thanks Swingy. I changed it to 1 and it made no difference

I did some logging and made a few changes today. How do I upload a log bigger than 2mb??

[swingtan]

Try compressing the file with Winzip or 7zip. You should get it down to about 33% of the original size.

[Dieselman]

Ok went for a log today after changing my VVE slightly from a earlier log. This is the tune after the change and the log I got with it,

10606 10607



I then filtered andadjusted my VVE again. I went for another short drive and this is the result,

10609 10608


I don't know what I did wrong but the last log was going very lean :bad:

I checked the WOT and it appeared ok. This is doing my head in to say the least :help2::help2:

[swingtan]

It doesn't look too bad, Just remember to log for long trips to make sure everything averages out nicely.

[Dieselman]

I just got a bit worried when I saw the AFR heading to 16.5:1 in the second log :shock::shock:

Is the second log ok or did I do something wrong with my VVE adjustment??



10610


Thanks

[swingtan]

It's probably just the VVE co-efficients going a bit off. What you can do it not the values in thos cells, and you know they need to go up a bit. So make your normal changes and after regening, see which way they moved. If they lower, or not sufficiently higher, manually kick them up and regen.

Simon

[Dieselman]

No worries I will look at that next time I change my VVE. How do I get rid of the negative corner on the map??? I tried a few things but it just doesn't look right

[swingtan]

I just manually change the values to make them more sensible. Just continue the curves of the VVE and flatten it out up over 6800 RPM, unless of course you have the limiter set higher.... In that case you need to use real data.

Simon

[Dieselman]

Hey Simon just a quick one - does raising the VVE figures richen up the AFR????

Did you see my photo's of the plugs I pulled today. I might just make my whole table rich and work my way leaner. I don't want to fry my engine :(


Cheers

Ben

[swingtan]

Yes. Higher numbers in the VVE indicate that more air is flowing into the motor. This tells the ECM that more fuel is needed to keep the commanded AFR. so the ECM uses these figure to estimate the required fuel quantity for "stable" engine conditions. Varying load, RPM and throttle position will cause the dynamics to add/remove fuel and will alter the final AFR. This is why you need to run long "stable" periods to get the fueling correct.

Simon

[Dieselman]

Hello Simon,

Can you see what is causing my LTFT's to become active in this tune. I have set the tune like the tutorial but it looks like they have been active all along :shock:

This is probably why my VVE tuning has been difficult to get spot on

1089010891

I will be running this tune on the weekend and will get about 6hrs of logging

10892 10894

Any help much appreciated

Ben

[swingtan]

I'm going to guess that you never cleared the LTFT data after disabling in the tune. The E38 will stop learning LTFT data once disabled in the tune, but will keep applying already learnt data. a tune flash will not clear the learnt trim data and you either need to reset the trims using the DVT controls in "ScanTool" or power reset the ECM ( pull the fuses and turn the ignition on, then turn off and replace the fuses ). This will clear the learnt LTFT data.

Simon

[Dieselman]

I'm going to guess that you never cleared the LTFT data after disabling in the tune. The E38 will stop learning LTFT data once disabled in the tune, but will keep applying already learnt data. a tune flash will not clear the learnt trim data and you either need to reset the trims using the DVT controls in "ScanTool" or power reset the ECM ( pull the fuses and turn the ignition on, then turn off and replace the fuses ). This will clear the learnt LTFT data.

Simon

I cleared it a while ago but it has come back. Why would it keep jumping around??? It would just stay at one value if it is disabled wouldn't it??

Any clues on clearing up my knock in the logs :help2:


Cheers

[Dieselman]

Hello all :)

Finally got my cam fitted and have started the VVE setup again. I did a quick log last night and got my BEN's close to where they need to be. I then loaded a CLSD tune back in with the updated VVE table and went for another short run. I know the runs are not really long enough but it is all I had time for

Can you see why my STFT's are about +7-10???

I have attached the VVE and CLSD logs along with the tunes



11562 VVE 01 tune

11568 VVE 01 log

11563 VVE 02 tune after changes from VVE 01 log

11561 CLSD tune

11564 CLSD tune log



Cheers

Ben

[swingtan]

Did you take a log post VVE adjustment and pre CL tune? That would at least show if it was a VVE mishap that caused it.

Otherwise, You need to fix up that knock retard. Significant timing changes will alter the exhaust AFR as seen on the WB, so you want to eliminate that when getting the fuelling dialed in. The Idle looks very good, what cam specs did you get?

Simon

[Dieselman]

Did you take a log post VVE adjustment and pre CL tune? That would at least show if it was a VVE mishap that caused it.

Otherwise, You need to fix up that knock retard. Significant timing changes will alter the exhaust AFR as seen on the WB, so you want to eliminate that when getting the fuelling dialed in. The Idle looks very good, what cam specs did you get?

Simon

Hey Simon,

The CLSD tune has the same VVE table as the VVE 02 tune. It was what I changed it too after my VVE 01 log. My BEN's were fairly good in the data I logged so I was just wondering why the trims would be out 10%. I hope that makes sense. I have not logged my VVE 02 OL tune yet.

What do you do with spark settings?? Just take some timing out of high and low tables where the knock happens?? My spark tables are factory

I ended up just putting a factory LS3 cam in. This was going to work better with my future plans and keep it nice and stealthy :grin:


Ben

[Dieselman]

I will add that after I filtered my log map I copied with labels and filled and then opened the map in the Ad-Hoc conversion chart and got rid of any cells that looked too different from the surrounding cells. Then I pasted this table to my VVE.

Is this ok to do???

[swingtan]

OK, I was wondering if you had a log of the OL version of the tune. Take the current tune and just disable the STFT's and log some data. Just as a comparison to confirm that the VVE is still good.

For the KR, just highlight the area's where KR occurs in the log and go back to tune tool and remove an equivalent amount of timing from the high octane table. Once all sections are done, copy the high table to the low table and subtract maybe 8' from the entire table. Then add 4' back in on the idle cells.

The stock LS3 cam would explain the nice flat idle.....

[Dieselman]

Ok I will do some OL logging this week if I get a chance. I will go through the log and sort out that KR!!



The stock LS3 cam would explain the nice flat idle.....

Yeh it is not because I know what I'm doing that is for sure :doh2:


Thanks

Ben

[Dieselman]

Hello Simon,

Well after 4 weeks I finally got some logging done today :shock::shock: I have been a bit busy at work!!!

Here is the log anyway.

11845

11844

This is what I have changed my VVE too. Does it look ok??? it is hard to tell what has actually changed after the Coef and VVE blending. I have adjusted my spark tables in the new tune and will see how the KR goes now :)

11843

Cheers

Ben

[swingtan]

Looks like the attachments didn't work....

[Dieselman]

Try again mate I stuffed up the attachments :doh2:

[swingtan]

OK, I've had a look.....


The log is pretty short, making it susceptible to transient errors.
The first half of the log shows high IAT's and a very rich exhaust AFR. This indicates to me that the Injector vs IAT correction is off {B1213}. As a start,

Set -40' to 1.21
Set 120' to 0.70
linear fill between the two.


this should give a pretty good starting point for IAT flow correction. In fact, it's the theoretical correct values ( thanks to The_Alchemist for this ).

as a result of the above, I wouldn't have used any data for the fist half on the log.
The WOT sections, how ever, look pretty good and should have some good data there.


Moving on to the comparison between the tunes.


Most of the WOT data indicates that the fuelling is on the lean side. When comparing the VVE's between the initial and altered versions, it shows that in most of the cells, the altered version has lower VVE values, which will result in further lean-outs occurring. I can only assume that insufficient correction data was applied and the generation of the coefficients rounded the data down to a lower level than previous.
The idle cells indicate that previously, it was a little rich on idle. Checking the altered log shows that the VVE has higher values when compared with the original tune. This will result in richer mixtures being applied during idle. again, I'm going to assume that insufficient correction data was applied and the generation of the coefficients rounded the values back up.


To fix all this, I'd apply two additional steps to the injector flow correction one above.


Reset the RPM zone boundaries {B8020} and go with 1000 2200 3400 4600. This will give finer control in the lower and mis RPM zones to help shape the VVE. Follow the normal steps to regenerate the VVE after making this change.
If using BEN data, apply the BEN data at 120% to 150% ( use 150% if the fuelling is a long way off ). This should be done if the log sample is small, but you are confident the data is accurate. Once applied, generate the VVE and see how that goes.

here's a sample..... 11856

Simon

[Dieselman]

Thanks Simon,

The intake would have been heatsoaked I would say. The VE's take a LONG time to cool down as can be seen in the log I guess. AAT was 24-25C so it was a bit above that. I will add AAT to my PID list!!

With the VVE it really makes it hard to get the cells right when they flatten out so much during a re-gen. Is that why you say to go 120% and over correct slightly??? Is there any way for me to compare VVE02 to your modded tune. When I compare them now I get some funky shapes going on with original tune. This would be because the RPM zones have changed??

With the PE stuff is it normal for the WB to be lean for a bit compared to commanded AFR?? Is it because when you open the blade there is a in rush of air and it takes a bit for the fuel to catch up.

This is a screenshot of the filter I am using. Feel free to suggest any other parameters I need :)

11859

Thanks for the advice

Ben

[swingtan]

The filter looks fine.

When comparing things like the VVE, I'll open two copies of Tune tool and directly compare that way. If I show the full table and orient both to be very similar, I can use the Windows-7 "preview" option to quickly swap between windows. I don't do it very often though, only to check a few things here and there.

If there is little data in the log file, I wouldn't use a BEN factor MAP to make adjustments. I'd just look at the log and manually adjust the cells where the data was valid.

Simon.

[Dieselman]

OK that is how I have been comparing them if the zones are different :)

I will load your test tune in and see how it looks then. The main problem I have driving around here is the lack of hilly terrain to hit a few more cells. I might have to find a dyno and get some constant load to help me!!

Cheers

Ben

[Dieselman]

Hello Swingy,

Did some more logging today to try and tidy up my WOT zones. I did some logging on my way home from Melb last week and found my STFT's drifting a bit with IAT's so I tweaked B1213 to how I thought it should be. The only thing I noticed now was the AFR's are getting too rich as the temps get hotter. Maybe there is something else I need to change???

If you could have a look that would be great :cheers:

12041

12043 Image from VVE04 tune log

12048

12040

[joecar]

Hmmm... interesting graph.

[Dieselman]

Hmmm... interesting graph.

Interesting good or interesting bad??? I cant post the whole log because it is 22mb

[swingtan]

I've tried to respond a couple of times but the page keeps resetting.......

Anyway, I'll look at the logs tomorrow, but here's my take on the image.


The chart shows how the STFT's react with a climb in IAT.
The IAT is rising as warm air from the radiator is drawn in to the intake. The warm air is coming from the radiator, which is heating the air around it. This is a great example of why it's important to have the IAT sensor in the correct position.
As IAT rises, the STFT's correct more. This indicates that either the INJ flow vs IAT settings may need adjustment.
As IAT goes through about 90'C, the thermo fans switch on, drawing air through the radiator and pulling the warm air away from the intake. IAT's drop quickly as cool air is fed in and the STFT's react accordingly.
The change in rate of IAT drop changes, possibly as the fans drop off in power.

[Dieselman]

Thanks Simon :)

That image is taken from my return trip home on the VVE04 tune. I pulled up after being on the highway and just let it idle until the intake temps got to 50c and then I hit the road again and the temps dropped back to ambient. If you need a copy of the full log I can email it to you

I have adjusted the injector flow table further in the VVE05 tune and did a log with a temp rise section and it looks like my trims stayed a tiny bit flatter. The only thing is once the temps go this high my PE ratio's are out. This is ok in summer because they will go rich but winter they will go the other way :shock: Hopefully we have another cold day to do some testing.

I will spin the MAF around like we talked about and get the sensor away from the radiator hose and into some airflow!!


Cheers

Ben