TunerStudio VE Analyze Live! Custom Filter

[Rob]

Looking for a custom filter for Ve Analyze Live! I have a big cam and see lots of lean misfires, (AFR pegs high 19-20ish) on decel and takes a little to recover. I was hoping to mask any AFRs that are excessively high during the tuning process. I am wondering if these are affecting my VE table and how accurate it is.

[LT401Vette]

So you primarily just want to filter during decel? Decel Enrichment won't correct it for you?

A Datalog would help come up with a filter to accomplish that.

[Rob]

Actually I was hoping to filter any AFR reading above 19:1. Is that possible? I know the car is not getting that lean, and my problem is a result of the big cam. Plugs look good. I can't help but wonder how much that incorrect data going into my tune is affecting things.

[LT401Vette]

Yes, filtering over 19:1 is simple enough. This filter would do it:
afr1 > 19

But I figuring you can probably dial it in a bit more looking at other variables and factors.

Also I'm not sure how big your cam is, but that is often wrongly blamed for tune issues. Typically once you are over 1500 RPM the tune objectives should br the same until you get very big where I have seen a need to run lower afr's in most driving ranges. But big means very different things to different people. I have had a lot of people say they are running a very big cam only to find out they are running .550 lift and 230 @ .050 lash. When I say very big I am looking at cams over 285 degrees @.050 lash.
My current cam is 0.757"I/0.711"E and 253 / 262 deg @ 0.50 duration and that still tunes very nicely, with an 850-900 rpm idle no issues of lean out on decel.

[Rob]

Here are the cam specs. I should also say I have a 6/71 blower on the 392 hemi. I suspect I am seeing some extra fuel hanging around in the blower case that is helping to load things up like the old tunnel ram manifolds. I was incorrect to blame the rich conditions entirely on the cam.
valve lift: intake 0.485 exhaust 0.485
<R.A.R>
.050 duration: intake 238 exhaust 238
duration @ .008 intake 286 (effective) exhaust 286
lobe centerline: intake 107 spread 110 exhaust 113 advanced 3
.050" valve timing: intake open 14.0 BTC Close 44.0 ABC exhaust
open 54.0 BBC Close 4.0 ATC
I think between the blower an the cam, I am seeing alot of unburnt fuel going through the system at different times and this is what I want to filter. Here is a log file.

[LT401Vette]

That is some interesting data you are working with.
It looks more like you are getting the heavy lean outs during accel, more than decel. VE Analyze does filter any records where the accel or decel flag is set.
It seems it is the WB signal delay that makes it seem it is decel.

I would think there is not enough accel enrichment coming off idle, so VE Analyze is trying to compensate by cranking up the VE table, but looking at your PW's, it does look like it is getting plenty of fuel, is it more a case of being so rich it mis-fires and makes the AFR look lean.

It looks like you have some really big VE table values there 180-200 during those low rpm readings. It doesn't seem like your injectors are going to cut it under boost over 3000 RPM's.

Looking at the records where you were running a little more steady state, like around record 1244, you have some major lean outs there too with the ve close to 60, so it looks like it needs to go a good bit higher even at 50 kPa & 2700 RPM..

[Rob]

Phil,
I think the datalog I gave you was a little dated. Please see the one below and my current msq to see if anything jumps out at you. I am running fuel only at this point to keep things simple. I would like to get this figured out before I pull the msd dist. and lock it down. I have a msd6aL and a ready to run distributor handling the spark for now. I just can't seem to get the AFR to settle down. I think I understand what you are saying about the injector size, but I have stock pistons (10:1) and the blower wayyyyy underdriven so I run in a vacuum most of the time, I create 4lb - 5lbs of boost at 5500 rpm. I have (8) 42lb injectors at the top of the blower. Do you think I should go bigger with the injectors still?
Thanks,
Rob

[LT401Vette]

Your injectors do look to be on the small side still. They should be able to get you tuned at driving ranges now, and watch duty cycles as you start getting to heavy throttles. I would recommend you turn the Number of squirts to 2 instead of 4 as that gives your injectors more head room with only half as many opens.


Have you calibrated your AFR under the tools menu? Under the menu Tools --> Calibrate AFR

It just occurred to me, I don't think your WB is reading right. It seems to make a quick transition from rich to very lean with little changes to the fueling. It is odd that at about 80 kPa it is demanding double the fuel.

Do you have a fuel pressure gauge on it? Is the fuel pressure falling off?

[Rob]

I do have a fuel pressure gauge (mechanical) on the areomotive regulator and it holds very steady at 41-43 psi. I will take your advice on changing the squirts to 2 from 4. I used to run 2, but the car seemed to idle better at 4. I will change it back though.

As far as the WB goes, yes I have calibrated it under the tools menu. I will do so again and make a log to see what we get. I am using an Innovate WB LC1 and Xd-16 gauge. I have it setup for 10-20; 0-5v. Here are some screen shots.
Thanks,
Rob

[Rob]

Well I changed the setting from 4 squirts to 2 and recalibrated my AFR table in tunerstudio under the tools > calibrate AFR table.
I did not see much of a driving difference, but maybe you can tell from the log if the duty cycles or PW looks better. I do have a problem still sometimes pulling away from a stop sign and it feels like it loads up with fuel. You can see this at time period 418. The AFR goes to 10 and the log looks like things are going haywire. Then it clears out and drives somewhat decent. Also I attached the LC-1 settings in a word doc on the above reply if you did not see it yet.
Thanks,
Rob

[LT401Vette]

Have you been running VE Analyze on it? With the topic I assumed you have, but the table doesn't look like it, or something isn't set up right.
That log shows you running really rich in some areas and very lean in others., then looking at the table you still see a huge gap in VE between 70 kPa and 80 kPa.

I just ran MLV's VE Analyze on that msq and log and it made a lot of changes, the table looked much better. It is blending the 70-80 kPa gap out with just that short log.

If you are using TS's VE Analyze, are the heat maps showing good coverage of the table in the normal driving ranges?

[Rob]

Yes, I have been primarily running VE analyze Live while driving it around. Last year I ran MLV analyze with the same results basically.

I will go back to MLV VE analyze and run that and do some more logging to see if I can get it better.

You asked about the heat maps. Yes they are showing good coverage in normal driving ranges. I have logged countless hours driving until the heat maps turn dark green in the normal driving areas. Then I stop tuning and restart it to see if the areas that just showed green would change any more and they don’t. I would go out and tune some more a few days later and the whole table will want to change again, leaning out or richening areas I previously had tuned. I basically have always had the 3d shape you see for the VE table. I find that large bump to be very weird. I have tried to manually lean out the peak and raise the surrounding values to flatten it out as the manual says to do, but VE analyze live puts the old values back in, like that is what it needs. You earlier spoke about my WB delay, should I increase it or decrease it? I have tried to slow down the response briefly a couple of weeks ago and did not see a dramatic change in the table after tuning with a slower value. I set the WB delay back to the original value.

[LT401Vette]

At this point I wouldn't be too worried about WB delay, that is for when you are down to fine tuning. I'm trying to put my finger on what isn't working right. That table still needs a lot of work. But by running that short log and msq through MLV's VE Analyze, it is doing just what you would want. It is blending the values and enriching where it looks to be running lean and pulling down those really high VE numbers. Check out this picture of a before and after.



One thing that might help is getting another bin between 80 and 100. I don't think you would miss that 35 kPa bin much, and it would give you a lot more heavy med-heavy load tuning to have a 90 kPa in there.

Also your AFR table just caught my attention. It is set up with 14.2: at 80 kPa then jumps down to 12.9:1 at the next.

For starting out I would try an AFR target table that looks more like the one TS generates as the default for the TunerStudio Custom table. To see that look in VE Analyze.
-Click the Advanced Tab
-Click AFR Targets
-Select: "Switch to TunerStudio Custom"
-Click AFT Targets again
-Select "Edit TunerStudio Custom"

You could copy and paste that over to your VE Table 1.

[Rob]

Well, I added the 90 kPa line, and grabbed a new AFR table from VE analyze live like you suggested. I drove around and logged and ran the VE analyze in MLV. This msq I saved as current1a.msq. Then I drove around for much longer and logged and did the same VE analyze in MLV. Then I loaded that msq. This is my difference report between the two. It seemed a little better on the low end now, but the 3d table still has that huge bump at 2500 rpm and 100 kPa. Here are my log files and msqs from tonight.

[Rob]

And the logs: