You might find out why I do what I do.

 

Oh, I'm pretty sure I will find out!

But that's kind of the point. I've now been through countless variations, and can see now that this area is as much of an art as a science. With first-hand feedback on the advantages and drawbacks of each approach, I can hopefully get a balance that works for me...

Rich

 

Here's a good example of my ECU choosing not to use the Lower Boundary Spark Map during WOT upshifts...




I'm using all the GST Boost Pill v2.5 timing maps...

- High Octane Spark Map
- Low Octane Spark Map
- Spark Advance Upper Boundary Map
- Spark Advance Lower Boundary Map

...but the ECU is just blasting through the gearshifts using the regular spark map. My Ralliart ROM's a bit of freak, it seems...

If anyone has any bright ideas, I'm open to suggestions.


Rich

 

It looks like your RPM pull was 3900-5800 (?), where are the load cells. Kinda hard to know which parts of the maps your in without 2 reference points....

 

Here is some actual data!...

The second column is the timing. Granted the GST "Lower Bound Timing Map" entries aren't much lower than the timing seen here in the gearshifts... they are lower.

Like, at 4500rpm/240load, the LBT map timing advance is 6. ECU went with 9, from the High Octane Spark Map.

Weird...

Rich

 

The entire log matches HOT, doesn't seem to use LBT at all anymore for shifts.
Bryan, is this something relative to V2.5?

 

Are you talking your own vehicle's logs, or my excerpt?

Because my car's ECU has been seen switching to LBT for gearshifts. Sometimes. Other times, it decides to just bash away using the main High Octane Spark Map.

I have logs showing this behaviour from the original Mitsu ROM (with nothing more than "mode23" enabled).

I have one log showing it approaching one upshift via LBT, and the next one not.



I have yet to hear of anyone else reporting this sort of thing... so right now it's just a lone voice in Australia going wtf?

Rich

edit: So to clarify my rather obscure explanation, my car did this well before GST BaseMap 2.5 - even on stock map. It likes using LBT for some part-throttle upshifts, but doesn't much like using LBT for WOT upshifts. I can see a disassembler in my future.......

 

Ya I saw that with your car earlier. Must be a 3rd variable, like TPS (I've been trying to figure out if boost is also affecting it, which will 'kinda' match TPS; i.e. if you are light on the throttle you stay out of boost).

 

My boost curve is now set, and working just the way I want it! Check out all its majesty...



This was deliberately intended as a slightly more conservative boost curve than the original GST one. Still, a wee bit better than stock, eh.


That uses the following error correction setup...




Both the WGDC and BTEL tables are set up to accurately reflect the required boost. Because I've allowed a slight upward correction factor, any excessive drop after the initial peak can be recovered.

I have noticed different boost levels at night when its cool, versus hot daytime. The way the load targets work on the RA, I'm pleased to say it wants to boost slightly higher when it's cooler - much better than the other way round (as seemed to be the case on some earlier Evo ROMs, from what I've read).

On a colder night, my boost curve is limited by the load targets. On a hot day, it's limited by the WGDC - they work together really well like that... but it took some serious effort to dial in just right.



Now it's time to move on to fuelling. Just like boost-setting phase, I need safe, sensible targets!




I'd appreciate any constructive input on the following target AFRs. Please note:

• I'm talking real, actual AFRs. They aren't ROM map pretend-AFRs.
• The fuel I use is 98 RON (Australia-speak - 93 octane US-speak).
• Boost levels as above - 21-22psi, tailing off to 14-15psi.
• I have Evo X FMIC and all its piping.
• These targets need to be "most conservative", ie. hot day on a track. AFR tuning will be done with lean-spool disabled.

From trackside testing (Innovate LC-1 wbo2 logged in EvoScan), my car does seem to like 11.6:1 from peak torque to 5500/6000, then tapering gradually to 11.2:1. Zero knocksum reported. On richer mixtures like the ones the GST map delivers, the knocksum column gets busy.


Anyway, here's a first draft of target "most conservative, ideal" AFRs...




I'd be delighted to receive comments from those people who have already tamed the beast of Ralliart AFRs... bonus if you've done it with lean-spool off.

I do intend to enable a less aggressive lean-spool mapping once I'm done with my base/ideal/most-conservative AFRs.


I have already spent some time dialling in the easiest stuff, namely mixtures for WOT 3rd/4th gear pulls. That was straightforward, even for a newbie like me. Worked a treat. But as for all the transitional areas, the off-throttle/high-rpm areas... I simply don't yet have the experience to blithely set targets and tune to them - not off the top of my head. There's no way I'm hitting the road with that stuff changed without first getting a good understanding of what's what.

So I reckon the "target AFR" map above is a good place to start!

Cheers,

Rich

 

Okay, so the target AFR table disappeared from the last post. Never mind - was going to put up a revised one anyway! Here 'tis...



This is what I'm now using in my XLS number-cruncher. I merged a "MapTracer" tool with a "ROM AFR Translator" tool - both from EvoM members - and fixed up a few things to make 'em work a bit better with my EvoScan log imports.

In the process, I discovered my LC-1 AFR column is way out of sync... delayed by over half a second. So I built in an easy offset adjustment for that too. Making slow but steady progress!

Rich

 

Richard - you should open a tuning shop! You have already figured out more than 95% of tuners re RA/Evos

 

Sure! Great idea! Here are my top 5 business name choices...


Copy & Paste Performance

n00b Industries

Clueless Race & Rally

Jesus Built My Base Map

The Naked Flasher


Okay, that last one might not pass registration, and would be hard for people to Google at work.

Rich

 

Dont know if your interested, but if you adjust your boost limit it wont trail off quite so steeply.
Also, check this out, gave me a "new perspective" that has proven to be pretty accurate so far:
https://www.evolutionm.net/forums/ev...ng-theory.html

 

Boost limit has nothing to do with the curve. It's exactly what it's called, a limit, which once exceeded, will shut the party down.

 

Yeah, my boost curve falls off - but that's because I'm telling it to, via the BWGDC and BTEL tables.

It's a small t/c, and won't set the higher rpm ranges on fire. Well, actually, that's exactly what it'll do if you overstretch it!!

So what's the point of wringing its neck? My load at 5500rpm is about 215... and at 7000rpm it's about 175. If I want 225 load @ 7000rpm, I'll go for a different t/c!

Maybe later... For now, it's nice and punchy up to neary 6000rpm. Great for on-road driving. Not exactly perfect for long-track or the drags, but there you go.

Rich