In my stuffing around with ROM fuel mapping tables, I had the following (incompatible) maps in front of me...


(1) GST Tune fuel table (with stock lean-spool enabled).

(2) My very limited "WOT pull" based fuel table (hammered out on a track straight, with lean-spool disabled).


I needed to start somewhere in merging this stuff, but I didn't want to drive on a pig-rich, "lean-spool" map for part-throttle data collection. I needed a mix of the two. Something sensible, safe and based on more than guesswork. I just don't have the experience to guesstimate AFRs!

So, what to do?

My approach was to build a simple .xls tool that takes a ROM fuel table, runs it through a "percent of lean-spool" calculation (via the stock "LS AFR v Fuel Map AFR" table mechanism), and spits out the actual fuel map value used by the ECU at that level of lean-spoolage.


This "should" show me what the ECU will actually use when, say, nailing the throttle hard after light driving - close to max enleanment expected. With a quick edit to one spreadsheet cell, what the ECU will use after a few minutes lapping a track course like Dick Dastardly (or perhaps Muttley).


Spreadsheet attached, in a zip.

Note: When I download zip files from EvoM, I need to first extract the file from inside the zip... then rename that file so it has a .zip extension, then unzip the extracted file again. The uploader seems to make dodgy zips-in-zips. Or maybe it's just me...!

Perhaps someone else will find this useful... or get a laugh, at least. Lots of n00b assumptions at work, no doubt!


To use it, put your ROM's fuel map data in the top table (GST RA 2.5 High Octane Fuel Map is in there now, I believe). Next, change the highlighted "100%" cell to whatever you want. In the bottom table, you'll see the adjusted map values, after they have been run through whatever lean-spool level you entered.


My idea was to enter between 0% and 100%, in order to get a feel for the dynamic range of values the ECU will use. Something middle-of-the-road would be a good starting point, I reckoned.


So for my map merging effort, I dialled up 50% lean-spoolage, and put the adjusted values from GST tune in one part of my "lean-spool-disabled" map (eg. low rpm, high load).

I then dialled up 40% - slightly richer dynamic adjustment - and put adjusted values in another part of my map (eg. high rpm, low load).

That gives me a little more confidence I can actually go test-drive this map without the transition areas being "too lean" for WOT, or "too rich" for light throttle. Then I can get busy logging loads of part-throttle AFRs, and refine the map further based on the results.

Hope that makes sense and is useful to someone...

Rich

 

I know I didn't come across clearly, sorry; what I was referring to was based on his other adjustments to Boost related tables, he was hitting his "limit", not allowing the curve to continue out. If for instance he were to raise the "limit", the rest of his adjustments could help push the trailing end of the curve back up....a bit
If that was your intention for a steep drop off then disregard. Just trying to help with what little experimentation I've done on my own

 

I think you might confuse people by saying that is a limit. I'd call it a target, which is how the ECU sees it.

Thanks for clearing your thoughts up though.

 

Yes target load, you are correct sir

 

I feel like I'm off to a great start with my new High Octane Fuel Map - been taking it out and pushing a little bit harder each time, then retiring to check the results.

Tonight's last run got to around load column 160. Here's my map tracer output...





Map tracers aren't perfect by any means, and this one has an inbuilt "wideband offset" to re-sync my slowpoke data link from the LC-1 unit. I've also got a limited lean-spool thing going on, with up to 0.3 points dynamic adjustment possible. So the cell readings here are always just an approximation.

But still, that's pretty smooth stuff in the 140 and 160 load regions.

I do have to see what the deal is with the relatively richer 120 column though. Off to dig through the actual logged data, and tweak cells that need it.

As always, if anyone has a particular issue with the measured data I chuck up, please feel free to point it out. That's kind of why I'm doing this...

Rich

 

That is actual AFR or fuel map value?

 

That must be actual, hence values of 18+ (foot off the throttle). Otherwise something go bang

 

I found another good approach to tuning for AFR:
Using raw values in EvoScan, create a Pivot Table using Load as Column, RPM as Row, and Average AFR for data points. For WOT it will create exact overlays of the "WOT section" of the fuel map (or you can use Timing, anything measured really); as well as cruising, etc. Adjust based on variance between actual and map, then hunt out knock and clean up.
Works pretty good for road tuning.... so far

EDIT:
Oh yeah, to save you from reading 100 lines of junk, set rows and columns to range (will look exactly like ECUFlash tables if you choose )

 

If you look at those specific data points that produced those AFR, you will notice that was when boost started (positive values). My thinking is it must be some sort of delay in the fuel system as the volume of air increases in the combustion chamber?
Is that somewhat right razor?

 

That's actual data from wbo2, run through an XLS map tracer. Consequently, it's kind of a broad brush approximation.

Obviously, the raw, graphed logs communicate a lot more - as you well know!

But the map trace is a great sanity check.

Rich

 

Running it a bit lean then....

 

I'd love to Rich, but I never have a darn clue what the h*ll you're talking about!! You're way too advanced for me, it's like reading Greek..... backwards!
Gas ain't cheap these days Bryan!! Maybe he's trying to save an Aussie buck or two at the pump.....

 

That's okay - I don't know what I'm talking about either...

Nah, just learning by doing. Without the feedback from knowledgable people on EvoM, I wouldn't have the confidence to try much of this at all. It's pretty awesome when you think about it - I can have an honest crack at a modified tune, take it for a conservative 10 minute drive, log real data, number-crunch it, present it here in a pretty format, and get feedback from those who have done it a thousand times. Feed that back into my approach, revise the tune... and avoid a lot of heartache!

Bryan, as usual, is spot-on. It was indeed leaner than the AFRs in my "target" table. The next step is to take the "new map AFRs" from those cell calcs and apply them. Then re-test. And so on.

Rich

 

Thanks, mate. I really appreciate you keeping an eye on this. Blowing up an engine wouldn't go down that well with my wife.

That data is from my very first AFR map. First of many iterations. Really pleased to have got meaningful maptracer data, smooth(ish) cell transitions, etc. as that gives me a sound base for the next revision.

Two short test drives so far. 40% throttle max - no more than 12psi boost. So trying to be sensible... hehe.

Rich

 

Have you adjusted the "Open Loop Load" crossover point? I haven't touched this map myself, and I'm not sure if its #1 or #2 that gets used under normal operating temps. Map #1 seems to keep it leaner longer under closed loop conditions. I'm lean in this area as well, but I thought it helped spool and wasn't a big deal until any "real" boost was made.

So what would be a good target afr for mid 100's load? 12.0?