Setting up VVE to show lambda

[mowton]

I checked it and to me it looks like percentage instead of lambda. It looks like straight percentage. If so, at say 55kPa and 1600rpm, i am 22% rich and would half multiply??

I was expecting to see a rate of 1.xx. If I was say 10% rich it would be 1.10L or 10%lean being .90L

In the channel, 1.27 says 27% lean....you are thinking EQ where numbers greater than 1.0 are richer!!!!! numbers less than 1.0 are richer.

Where are your chart parameters for eq commanded, wb and Eq error??????

Ed M

Update -- the lean is more in decel after looking closer.....the values in the main field are with +/- 6%....not bad Post a log so we can see the actual data.

[Ghostnotes]

Been in the hospital with my wife the last 2 days. She had a severe appendicitis. Should be goi g home today though. Will try to post up one soon.

[mowton]

Been in the hospital with my wife the last 2 days. She had a severe appendicitis. Should be goi g home today though. Will try to post up one soon.

Hope she is feeling better...

Ed M

[Ghostnotes]

In the channel, 1.27 says 27% lean....you are thinking EQ where numbers greater than 1.0 are richer!!!!! numbers less than 1.0 are richer.

Where are your chart parameters for eq commanded, wb and Eq error??????

Ed M

Update -- the lean is more in decel after looking closer.....the values in the main field are with +/- 6%....not bad Post a log so we can see the actual data.

This is the only log i have so far and it's not set up for anything, running MAF only and closed loop. Yeah i see i said that bassackwards.

I thought the math handled the work. I have the EQ Commanded and WB in my channel list. I thought the EQ error was a function of the math i set up in the chart. Although i might not have had everything loaded in the channel list in when i did the initial log. This is how i have my layout as of now.

[Ghostnotes]

I still don't see this as Lambda for the VVE. Since 14.68 afr=1.00L. In the Log, it shows up to 20.4 Thats not possible if displaying Lambda since 0-5v =0.683 to 1.365L which is 10.00 to 19.98 AFR.

[schpenxel]

It's working correctly. The small, ~2% difference from 19.98 to 20.4 doesn't mean anything. That can easily be explained by ground offset error being at the extreme end of what both the WB can output and the HPT interface can read.

The colors you're using make it look like it's a total mess because you aren't understanding the units. The high numbers you're getting are during decel where the WB is maxed out lean. You can either ignore these or start filtering them out. I took the liberty of adjusting the colors a bit to make it more sane in the attached layout, but bottom line, it's working correctly. I also added a filter to try to help a little bit on the first one.

You can do the math yourself if you want. Here's the formula it's using:

100 * ([WB_Lambda] - [Commanded_Lambda]) / [Commanded_Lambda]

So take worst case of 1.365 WB_Lambda and commanded_Lambda of 1

100 * (1.365 - 1) / 1 = 36.5%. This would mean you're 36.5% lean, which is exactly correct.

Most of the times you're closer to WB_Lambda of 1.29 to 1.30 during decel, which obviously works out to 29-20% lean, which makes sense given your graphs. You also have a massive lean spike when you go into PE, which shows up correctly.

I also created some new graphs that put your wideband reading, commanded lambda reading and EQ Ratio error all on top of each other. That should make it very clear what is going on.

That's about all I can do--if you still don't believe it's working correctly then create a math parameter on your own that works instead.

Lambda VVE layout7-25-16..Layout.xml

[Ghostnotes]

I appreciate the help, all of it. I just cant make it compute in my head. When i did my MAF tune, and set up my offset with the formula above, when i go into PE it would read right around .85L (as it should) and i should never see a range higher than 1.25L ish in the cells. Seeing 20.xx is confusing because like my MAF setup, im expecting to see a range from .85 in PE to around 1.25 during DFCO. Trust me ,i understand that 20.8 is 20.8% rich,, I'd round up to 21.What makes it easier for me is if i see a reading of 1.09, I want to multiply that 1.09 to the table instead of doing the math of 9% of X.

When i started Tuning, I used the banish DVD's....still do. I followed the instructions on setting up the MAF/ Lambda calibration. He then went on to VE tune using the same technique. Do i really need to VE tune? I guess that's a question of perspective. My biggest reason is in case the MAF does fail.
When 3.0 came out i heard that a lot of brands were supported, mine included along with the lambda transfer function.


The lean spike is transitioning. The tune you see was done entirely by me. The cam I'm running is 232/242 .620/.620 110 +0, no stall and no buck. It still needs a little work. But not much.

I downloaded the above and i think i know where we differ. I rarely if ever use the chart, I go by the graph cells. The only time i use the chart is watching for KR and throttle position. I watch everything else in the channel list.

I will do a scan later today with the changes that were made and post it up here.

Thanks for being patient.

[schpenxel]

If you want to use WB / commanded like Banish does then you'll have to create a math parameter to do it; there isn't one built in. That is easy enough though.

The math parameter "Expression" would be [50119.238] / [50118.238]

So.. make a math parameter with that formula (or "expression" as HPT calls it) and then use it as the "parameter" in one of your VVE graphs and that should get you back to the units you want. Obviously you'll want to update the min/max colors and such then since you're going to be at a totally different range.

There is not a built in math parameter that will report out the way Banish does in his DVD's though--you'll have to add one, but that is easy enough.

Here is a math parameter you can use:
WB Lambda over Commanded Lambda.MathParameter.xml

[Ghostnotes]

Ok got it....finally.

When i heard/saw Lambda in the update, I assumed it would already be baked in.....sorry about that. Thats what i was actually looking for in the software. I figured since they had updated the software and automated a lot of in aftermarket sensors, that kind of function was put into the math already(which might not be a bad idea Bill...etc) especially for amateurs like me.For me at least, most everything else is easy to figure out...timing, trans settings, injectors. But for some reason this has been the most difficult by far, and I'm stuck on it.

I'm happy with the tune for the most part, especially since a few pro tuners said that cam will not work with my setup and will buck like crazy.I still get 22MPG if i stay out of it.

[Keith@HPTuners]

I checked it and to me it looks like percentage instead of lambda. It looks like straight percentage. If so, at say 55kPa and 1600rpm, i am 22% rich and would half multiply??

I was expecting to see a rate of 1.xx. If I was say 10% rich it would be 1.10L or 10%lean being .90L

You can just drag the Graph tab onto the Gauges tab so you don't have to smush the Gauges like that.

[schpenxel]

Ok got it....finally.

When i heard/saw Lambda in the update, I assumed it would already be baked in.....sorry about that. Thats what i was actually looking for in the software. I figured since they had updated the software and automated a lot of in aftermarket sensors, that kind of function was put into the math already(which might not be a bad idea Bill...etc) especially for amateurs like me.For me at least, most everything else is easy to figure out...timing, trans settings, injectors. But for some reason this has been the most difficult by far, and I'm stuck on it.

I'm happy with the tune for the most part, especially since a few pro tuners said that cam will not work with my setup and will buck like crazy.I still get 22MPG if i stay out of it.

There is a built in WB error parameter.. you just aren't liking the units it's spitting out, BUT that is the way most people tune. Most people don't do it the way Banish does in his DVD's. The idea for how most people do it is for the "error" values to be comparable to the way fuel trim values work. -5% mean you're 5% rich, so airflow needs to be reduced by 5% there.

HPT built in a "paste special-->multiply by %" option specifically for using the unit the EQ Ratio error math parameter (and fuel trims) spits out. You can copy the graph that is created while scanning, paste special/multiply by % and update a whole table at once. Obviously you need to do enough sanity checking to make sure you aren't factoring in junk like data during decel.. but you get the idea.

Either will work perfectly fine as the math is the same either way. I started off doing it the same way you're doing it, but it's not the only way and I would never go back to that way.

That paste special function WILL NOT work with the way you are doing it using WB/commanded for an error "factor". So that leaves you multiplying every cell or group of cells manually instead.

[Ghostnotes]

Thanks for the info

[Ghostnotes]

You can just drag the Graph tab onto the Gauges tab so you don't have to smush the Gauges like that.

OHHHHHH so now you jump in....J/K. I see how you can TAB it now.

So how about setting up a permanent math transfer function as described???
Seeing a factor of 1 and being able to multiply e.g. 1.06 to the table would be a nice option.

Speaking of which, will i just add the difference to the VVE table? Reason i ask is because i made a change to 1 cell(did not save). Because when i did, the coefficients became active.

I also see options for camshaft angles?

Sorry about being nitpicky but it is far better to ask instead of just assuming anything.

[Ghostnotes]

This???

[Ghostnotes]

Nevermind I got it.

I was still thinking like 2.25 in trying to find the user defined instead of going to the menu.....

I was looking for the configure user defined, but that does not generate until you create one.

Again I cant thank you enough for being patient through all of this!