No, I haven't touched that setting - still factory.

As for target AFRs mid-100s load, my current target table (got to start somewhere) has about"
12.2:1 at 4000rpm / 150 load , enriched to...
11.7:1 at 6000rpm / 150 load , enriched to...
11.3:1 at redline / 150 load

Like so: (ignore the yellow WOT highlighted path)...




If that's a target only a complete donkey would aim for, now would be a really excellent time to say so.

That will give me the chance to make the requisite EEH-AWW! noises before I use said maps in anger, instead of after.

Rich

 

Hmm, I had lean-spool on for my first AFR map. It's a bit early for that. It'll be off now, while I get my AFRs properly dialled in to suitable "most conservative" style targets.

That should drop most open-loop readings by a bit... Then I'll make sure those targets are good, and revise the map accordingly.

Rich

 

Could you elaborate on your understanding of lean-spool? I don't really have much of a clue how it works. I had heard that is helps to not richen up through the gears. That seems to be true, but odd to me since the turbo doesn't spool at every gear shift. After exploring areas of my map tonight that you inspired me to log, I had NO success at fine tuning it. I was quite lean when I stomp it from about 4000rpm, and by the time I had that sorted out it was way rich once it spooled up. Also, would you happen to have the xml for 53600010 that address the lean-spool parameters?

 

From what I gather, it's a timer-based system that takes the relevant value from the High Octane Fuel Map 3D table, and then applies a certain percentage adjustment from the LeanSpool AFR v FuelMap AFR 2D table.

The "certain percentage" depends on how much you've been thrashing it up to that point (hence the timers). The more you've driven it like you've stolen it, the less enleanment adjustment will occur.

The idea is (apparently) to keep the car perky in "quick squirt" conditions - leaner - but protect the engine with richer mixtures if belting around like a nutter.


Yeah, ain't it fun! There's plenty to this whole area, that's for sure.

I can go into a bit more detail a bit later, if it'll help.


Afraid not. Is it not in the definition from goldenevo.com ?


Rich

 

That's a great site! I had no idea it existed. But, I didn't see a definition for lean-spool stuff in my rom. I looked in 53610010 too (i think that's yours) and also didn't see it. Am I missing it?

I think I'll tun lean spool off. At least until I get the xml. It is making me MUCH too lean, and it too slow to come back down. Yes I do think tuning is fun, But I don't appreciate the lean-spool leprechaun adjusting my tune if I don't know the rules he plays by!

Thanks Rich

 

Yeah, I'm on 53610010. The XML definition I have is...


For scaling...


For table...


When addresses line up with ROM reality (check!!), it appears as:




Rich

 

Ah HA!!

That gives me the exact same map that you posted. Must be the same address.
That sure explains the lean readings. Asking for 11.8 and getting 13 is a big difference. For me, it holds that for quite a while. Now I see why I had to enter into the 9s in some areas to get it decently rich. According to this post , there must be more maps that define additional attributes. I'm not sure if you've read it.
I turned it off for now. Much easier and predictable.

Thanks for the info

 

The more I work on this stuff, the more I'm hitting my target AFR table with the enrichment stick... especially around the 2500/3500rpm mark.

Latest target table, for reference...




Because the ultimate aim is to bring back a (more limited) lean-spool adjustment table, I need to keep asking myself, "am I okay with this area being 0.3 points leaner?". If the answer is "not really", then the target needs to be a little richer.

Taking it slow - still haven't taken it over 5500rpm and 150 load. zzzzzz. But more thinking and less doing can't hurt when getting to grips with AFRs.

Rich

 

Best luck. I don't know if your engine management will act like mine, but I had a really steep incline bubble up in my fuel table from the 3500rpm row to the 4500rpm row. Your results may be different - but don't be surprised if your final map table turns out different to the way you expect!

Furthermore, beware wbo2 data delays! Mine is awful - 0.7 seconds out in the EvoScan logs! Logging IPW (injector pulse width) showed it up as clear as day. There's some blurb on it here...

https://www.evolutionm.net/forums/ev...spool-off.html

Rich

 

I'm now enjoying cable-free in-car logging... having got the Tactrix OpenPort 2.0 performing direct sampling of a 0-5V wideband analog output.

It's all calibrated and double-checked, and seems to be very accurate indeed. It lines up very well against a sample LogWorks3 log file I took to cross-check it.

Specifics here...

https://www.evolutionm.net/forums/ev...ml#post9106692

Here's my full OP2.0 "logcfg.txt" showing the "type=adc" configuration...


If I drop one of the two load parameters I'm logging here, I get about 13 samples a second. With the LC-1's analogue output configured to smooth/average to 12 samples a second, it's a damn fine setup.


Now... enough of the fiddling with logging, gauges, config. Time to hit the road!


Rich

 

Thanks for posting that txt file. I've been wanting to get this going, but I was to lazy to get it all set up.

 

No worries. In theory, your EvoScan "Log and Power Graph" display should work from the OP2.0 logfile. All you need to do is:

- Have the same log item headings in both EvoScan and OP2.0.
- Use the same addresses as EvoScan (obviously).
- Edit the OP2.0 logfile header line, changing "time" column heading to "LogEntrySeconds".

If I can find some way around that last step, I'll be able to load up OP2.0 files straight into EvoScan. But neither product seems to have a flexible column heading name - "time" / "LogEntrySeconds". At least the data itself is good.

Note: I've found the OpenOffice spreadsheet thing (Calc?) is better at editing .csv files than Excel. My version of Excel goes on a reformatting frenzy on .csv files, and EvoScan spits the dummy.

Rich

 

After some gloriously unsuccessful attempts at AFR map tuning, I'm finally making some good progress.

Let's hope so, anyway. I do recall saying that before - then coming badly unstuck... and binning the whole project in a hissy fit. However, I'm not smart enough to know when to quit!

One reason I've been having so much trouble is that my Ralliart ECU's interpretation of the High Octane Fuel Map is anything but "flat". I'm so jealous of all the Evo maps I see posted that look smooth, pretty and sensible. Mine is anything but!


I started with Bryan's GST Base Map 2.5... but with Lean Spool OFF.

Also, I reverted the Closed Loop LT Trim Control to stock settings. Note: It was this way up to GST Base Map 2.0. Bryan changed this for 2.5, no doubt to disable the excessive variation in mixtures that occurs when LTFT drifts off the zero mark. I wanted this back to stock to avoid tripping over the "kink" in the AFR curve as I set up my map during spool-up. I'll probably put it back once I'm done, to lock in my WOT mixtures.

Here's a look at my vehicle's measured AFR on the GST fuel map, Lean Spool OFF...




And here's the High Octane Fuel Map being used, with a couple of readings noted against map values...




As you can see, the map's AFR value is fully 1.2 points richer than reality at 3500rpm, and 0.4 points leaner than reality at 6500rpm.


Cell by cell, variations like that aren't really a problem. At each point, just plug the "actual measured", "actual map" and "desired measured" values into a simple formula, and get the "desired map" value.


For me, the real puzzle was how to get readings from all the other parts of the map - consistent readings that could be relied upon when revising AFR values.


On that score, map tracers turned out to be pretty much useless! That was a bummer, as I'd spent a lot of time messing around with them. There are two main reasons why map tracers are now off my Christmas card list...

1. If I run through X/Y cell with load steady and RPM increasing, I get a totally different reading compared to passing through the exact same cell with RPM steady and load increasing.
2. There's a lot of ECU jiggery-pokery going on in the background that has nothing to do with the map. Throttle-on accel enrichment. Throttle-off rich readings. You name it. All of that stuff messes up the map tracer data.

The only "road tuning" approach that really worked for me was to start from a nice, rich base map (eg. GST with Lean Spool OFF), and perform 4th gear WOT pulls from incremental RPM starting points. Like, 2000, 2250, 2500, 2750, right on up the range.

Based on the results, I then revised the map cells between load 100 and 160. This got most of the "spool-up" area set correctly. My target at this stage was to have the readings pass through 12.5:1 at 11psi (approx. load 150), on the way down to richer mixtures.

Once a good solid area of the map was set (say up to 4000 rpm, and up to load 160), I then moved on to load 180 and above... up to 4000rpm. I repeated the incremental pulls from rpm 2000, 2250, 2500... and ensured mixtures dropped to mid-11s before peak boost.

To illustrate, here are some example logs of work-in-progress...


WOT from 2500rpm...





WOT from 3500rpm...




WOT from 4000rpm...





Once you've got each example pull recorded, it's actually pretty straightforward to do what has to be done. And working from a rich basemap is nice and safe.


However, one interesting thing I found was that I did have to enrich the GST base map in the 4000 rpm spool-up area to get meet the desired 12.5:1 @ 150-load target... and that was with lean-spool off.

In fact, the more I chiselled away at the fuel map like this, the more it revealed itself to be anything but flat. Compared to all the other Evo "Lean Spool OFF" maps I've seen posted, it looks like a skate park!


I'll post up my final version of the map real soon. I haven't done the spool-up work above 4500 rpm yet... so that part of the map is still super-rich...

More soon!

Rich

 

Ah, why wait? Here you go... this should give you a laugh...!



To put some of that into context, here's the map itself, and a note showing the part that I haven't actually driven/logged/revised yet!...




It's hard to find a stretch of safe tarmac that allows you to get on-throttle in 3rd/4th at 4500rpm and above. Reckon this final area will have to wait until I book some track time.


For now, the highlighted bit of the map is an interesting combination of:

• Known and verified "typical WOT pull" tuning, 4500rpm and above (ie. load 220+).
• Known and verified spool-up tuning at and below 4000rpm.
• GST Base Map fuelling in the "Not Tested Yet" region, smoothed to the known, verified values.

Still, it looks completely unlike anything I'd expected... not that I've had any experience prior to this! To be honest, I would have liked my first tuning experience to be on an ECU/engine that that could handle being block-set. This has been far more challenging than I'd expected.

Still, it's very satisfying when you get it right! Each log I check has WOT spool-up cross 12.5:1 at 11psi, then drop neatly into the 11s before 20psi. It's bang-on every time. When you've had a hand in your own vehicle's tuning, it does add a whole other dimension to driving...

Rich

 

Richard, are the evoscan address's you have in post #11 the most up to date for the AUS/EDM 2009 RA's?