HOW TO: Setup AFR / Lambda ERROR Graphs vs. VE Table (non Virtual VE Vehicles)

This is meant to serve as a basic how-to for setting up an AFR or Lambda ERROR graph (histogram) for vehicles with regular VE tables (i.e. not virtual VE).

This assumes you already have your wideband working properly....

---------------------------------

First go to the Graphs tab. If you don't have a Graphs tab, go to layout, add to layout, graph. Right click in the graph, select graph layout
http://i1117.photobucket.com/albums/...orials/601.png


We want to add a new table.. select the graph icon with the + on it and select Add Table
http://i1117.photobucket.com/albums/...orials/602.png


New blank graph is added
http://i1117.photobucket.com/albums/...orials/603.png


Click on "click to insert or change" beside parameter at the top. Go down to the Lambda & AFR list and find EQ Ratio error OR AFR ERROR. I'm not 100% sure on this one but I THINK it depends on what units your wideband is setup to read. Mine is setup to lambda and the EQ Ratio Error parameter works. AFR error does not. If one doesn't work, try the other....

Double click on EQ ratio error or AFR error
http://i1117.photobucket.com/albums/...orials/604.png


Back to the layout editor, click "click to insert or change" beside parameter for the column axis.
http://i1117.photobucket.com/albums/...orials/605.png

Couple steps at once here.. type manifold absolute in the text box at the top of the parameter selector, hit the gear icon with green check mark (filters out stuff you can't use) and double cilck on whatever MAP pressure parameter is left that you can use..
http://i1117.photobucket.com/albums/...orials/606.png


You'll probably get a box about using a generic sensor. Generic sensors are good. Hit yes
http://i1117.photobucket.com/albums/...orials/607.png


Alright, now we need to match the labels up with the labels in your tune. Go to VCM Editor and open your tune. Open the primary VE table. Right click inside it, go to column axis and click on copy labels
http://i1117.photobucket.com/albums/...orials/608.png


Paste these into the "values" section for the column axis
http://i1117.photobucket.com/albums/...orials/609.png


Next is the row axis.. very similiar process. Click on "click to insert or change" in the row axis section, type in engine rpm, hit the gear icon, double click on parameter that is left. If you get a generic sensor box hit yes.
http://i1117.photobucket.com/albums/...orials/610.png


Here's what it should look like at this point
http://i1117.photobucket.com/albums/...orials/611.png


Now you need the labels for the row axis. GO back to VCM editor and go to your primary VE table. Right click, Row Axis, Copy labels
http://i1117.photobucket.com/albums/...orials/613.png


Back to Scanner.. paste these values into the "Values" section for Row Axis
http://i1117.photobucket.com/albums/...orials/614.png


Ta-da
http://i1117.photobucket.com/albums/...orials/615.png


Next I usually setup some colors for different values to make it easy to see, maybe add some minimum cells required while you're at it. You can set this to 0 at first if you want then change it later after you make sure everything is working
http://i1117.photobucket.com/albums/...orials/616.png


And there you go.. you have a working wideband Error graph/histogram
http://i1117.photobucket.com/albums/...orials/617.png

BUT... what about boost you say? I didn't forget that part. There are two ways to do this. You can either have a "regular VE" chart and a "boost VE" chart or you can do one big one that includes both.


Go back to layout editor (right click in a graph, click layout editor). Go to the Graph you just created. click on the graph button with the "=" sign on it. That will clone your current graph.. Rename one of them to something like "WB BOOST VE ERROR"
http://i1117.photobucket.com/albums/...orials/618.png


Go back to VCM editor.. this time open up the BOOST VE table. Right click, column axis, copy labels
http://i1117.photobucket.com/albums/...orials/619.png


Back to Scanner..d elete out the "Values" for the column axis section. Paste the boost VE table values in here..

Now you have a regular and boost VE WB error chart. If you want to make one BIG chart instead of two small ones.. you can do that too
http://i1117.photobucket.com/albums/...orials/620.png


Go to the "boost" VE chart you just made. Select the "values" for in the column section (skip the first value). Right click and hit copy
http://i1117.photobucket.com/albums/...orials/621.png


You can either make a copy of your "regular" WB VE Error" graph or you can just change the original one. Here I used a copy. Go to the Column axis section and paste the boost VE table values in after the regular ones. You should end up with something like this:
http://i1117.photobucket.com/albums/...orials/622.png


And ta-da here's what you end up with. Looks like I'm going a little lean, oops! (side note: The really negative numbers are mainly from deceleration and DFCO, those can be filtered out but we're not going into that right now.)
http://i1117.photobucket.com/albums/...orials/623.png


Enjoy!

..good description!

excellent. good write up, thanks. i'm going to try this tomorrow....so...question....in my scanner, i've gone to the channels table, right clicked, "add channel", and added MPVI input 1...then right clicked on that once it was added to the table, and clicked "transform." under "oxygen sensors" i opened the "Lambda" table and clicked "NGK - AFX" (my wideband) then clicked "ok".....is that all i have to do to log lambda in the scanner and lambda error with the write up you posted?

Quote:

---

Originally Posted by 2000_SS

excellent. good write up, thanks. i'm going to try this tomorrow....so...question....in my scanner, i've gone to the channels table, right clicked, "add channel", and added MPVI input 1...then right clicked on that once it was added to the table, and clicked "transform." under "oxygen sensors" i opened the "Lambda" table and clicked "NGK - AFX" (my wideband) then clicked "ok".....is that all i have to do to log lambda in the scanner and lambda error with the write up you posted?

---

Yep, that will get you lambda

I log commanded EQ ratio and commanded AFR. I assume the commanded EQ ratio one is used in the lambda error math parameter, so at least add commanded EQ ratio too (I would add both)

But yeah, that's pretty much it. Log lambda, log commanded lambda and the built in EQ Ratio Error parameter (i.e. lambda error) will work

ok good deal...sort of like logging AFR Commanded Hi Res & Commanded EQ Ratio in 2.24?

Ive been reading through your write-ups and they're very helpful. I haven't had an opportunity to tune anything with 3.0 until a 5.3-swapped LUV was dropped off tonight, i'll be starting on it tomorrow. I've been pretty nervous and confused with the new scanner. I can see that it is better than 2.24, but the logic in setting it up is so different i didn't really know what to do for sure. i think i speak for 90% of this forum's members what i say THANK YOU for your write-ups.

Yep, exact same thing...

Yeah there's def. a learning curve going to V3 but I think most will prefer it in the end.. It takes some getting used to but making charts/graphs is so much faster in V3 once you get used to it. Being able to search for parameters and filter down to only usable ones makes a huge difference

And no problem, I think I would have been screaming bloody murder if I had to switch to V3 overnight.. I had been using the beta on and off so that made the switch a lot easier..

Hi, I've been reading this and have been going over everything many times however I still have some questions.

So far on making my 'virtual VE' graph, I have followed the info above and my parameter is Eq Ratio Error (math) and the proper MAP and RPM axis are there.
I added the following channels: mpvi a/d input 1(for my ngk afx/afr500), air/fuel ratio commanded, eq ratio commanded SAE (was already there but I thought I'd mention it's in there).

So high of 10 and low of -10 are good to use for the shading portion or is that an unrelated default that the above how-to shows as an example?

Is that it? Is that all that I need to get the correct lambda error data in my graph and adjust my VE table accordingly assuming all my sensors in the channels get the signals they require?

Thanks

RPM and MAP are two obvious ones you'll need to log also

Easiest way is to just start logging and see what happens. If you want to post the channel config/layout I'll take a look and see if you're missing anything else

I hope this is what you wanted.
Attachment 56674

Have you connected to the car with V3?

Yes I have but I don't have my WB sensor in yet as that's 2 months away when I get my new exhaust figured out. I know I won't be able to get any data before that, just seeing if I'm on the right track.

Are you going to need to calibrate the MAF table(s)? If so you'll need to add mass air flow sensor frequency to the channel config (if it isn't already in there)

Is my upload image not visible? I posted a screenshot of my channels

I'll be using the same 85mm maf which has been running perfectly this past summer after a mail order tune. So I'm assuming I won't have to calibrate it but yes it's in the list.

So if I were to do all the tuning based on the wideband signal when I get it, should I still use the narrowband o2 sensors for any part of the VE table or can extremely accurate results be attained for the entire VE by solely using the wideband?

I realize the NB o2 is for 14.7 afr tuning but have read mixed results on using the WB o2 for idle and part throttle tuning. Any clarification on this?

I'd bet improvements could be made to your MAF calibration but that's simple enough once you get everything up and going

I've had good luck using fuel trim values (i.e. not the wideband) in most of the drivability areas. You can use the wideband everywhere but have to put it into open loop first. In the past when I've tried to just use WB readings I always get some amount of error once I put it back into closed loop. I'm assuming this is because things like the EVAP system don't work while in open loop, but I'm not really sure

Keep in mind both types of oxygen sensors are actually measuring LAMBDA, not AFR. In closed loop the narrow band O2's and computer will always drive towards lambda=1 no matter what the stoich value is for the fuel you're running. Even if you put E50 in there right now fuel trim corrections will push you towards lambda=1 or whatever stoich is for that fuel

Same thing for the wideband.. it's reading in lambda then just multiplying that value times 14.7. You could put E85 in there and even if it were at stoich for E85 (say, 9:1) your wideband will read 14.7:1.

That's why I just use lambda, it makes my head hurt less

Ah ok thanks for clarifying as for some reason I kept associating lambda with WB o2 sensors. Ok that's even better then, I'll use my NB o2 sensors for idle and part throttle and wideband for WOT tuning when I gain a more experience.

For me to make this graph/histogram a success so I can copy and paste values into my VE table, would I choose afr error or eq ratio error as my parameter in the graph?

And will the NB o2 voltage automatically convert its signal for the afr error graph to get what it needs?

Sorry for chasing my tail, this is all new to me and everything I read takes a bit to solidify.

Until you get the wideband you'll want to use fuel trim values. Use LTFT + STFT (it's a built in math parameter) as the parameter for your graphs. The AFR/EQ Error ones require a wideband.

The stock / narrowband O2 sensors are only good at knowing two things. 1. you're richer than stoich, 2. you're leaner than stoich. They can't really tell how rich or how lean you are, thus the need for a wideband when tuning WOT.

The computer uses the narrowband O2 sensors by constantly cycling slightly rich and then to slightly lean (i.e. fuel trims). Unfortunately they aren't very useful once you want something other than lambda=1

That helps a lot and I appreciate your time. I'm amazed how good the support is on here, I truly haven't seen this much patience before and detailed explanations, greatly appreciated.

No problem. All of us have been in the "how the hell does all of this work" stage.

Hey Guys,
new to HP Tuners
My LC1 inputs are noisy
I've turned the settings in the innovate down to 1/3 p/s
need help
Attachment 56698