X-Tau/EAE are sweet!
 

Well, sorta. I've had MS2 for a year and always used the simple acceleration enrichments, like what everybody here using MS1 uses, only with MS2 I can use both map and TPS based and then select a % of which it looks at. IE-75% map based and 25% TPS based, etc.

I've adjusted them countless times. I can get them decent, but it's never perfect. And if ambient temps change to much they're off again. We all know what I'm talking about. So today I disable them and try out this fancy EAE correction stuff.

Now acceleration enrichments are way off (because I haven't tuned anything yet, and they're 6 tables to tune now instead of 2...) but I can tell it's gonna be better. Aside from it falling on its face at anything other than light throttle changes, the response is crazy. You'd think I put a set of IRTB's and a lightweight flywheel. Like before, say I'm at 1k RPMs and I blip the throttle by the time it would normally go to 1100 RPMs, I'm at 1700 RPMs. Crazy stuff. Granted it drives like shit and for some reason it refused to crank one time, flooding out like mad, but other than that, it's great! :)

Now I've gotta read for days cause apparently this shit's hard as fuck to tune and get right, but this stuff is how OEM's tune acceleration enrichments. It's literally always modeling how much fuel is stuck to the walls ( like stuck to the plenum, sitting on the intake valve, etc), how much is being added to the walls, and how much is being sucked from the walls and all are a function of MAP, RPM, and coolant temp. So if I stab the gas, it looks to see how much fuel was stuck to the walls, how much of that is being sucked off, how much is currently being added to the walls, and how much the injectors are spitting out. Then it runs the numbers and figures out how much is actually getting to the engine. Crazy shit.

Now I just need to pray that AbeFM or someone with MS2 has done the leg work and tuned this for a miata before. :cool:

I could never get X-Tau working right. It made my rolling idle problem much worse. I don't know how applicable it would be for us anyway. It compensates for wall wetting/drying which sounds like throttle body injection. How much wetting could there be in our cars? It was a complete pain in the ass to tune, too.

I never gave EAE a chance when I switched to MS2E. The TPS based worked good enough.

I'm very interested in what you come up, though.

When I bought my car the PO had all this enabled and the car ran really well. Later on I can post up the settings he had. Granted, these were all for 450cc Low-Z injectors.

What do you mean by rolling idle? My idle was a little weird actually, like a very fast 20-30 RPM hunt. Very strange.

I've messed with the regular acceleration enrichments forever and I can never get them right, so I figured I'd give this a chance. I sifted through my recycle bin full of various people's datalogs and msq's I've looked at before and found somebody had one titled "95-miata" that had EAE settings, and I just printed them all out. They're a bit different than the ones that magically appeared in my msq when I enabled EAE, so I may try these to see how they do.

i'm currently running TPS based enrichment but i'm hoping to at least try EAE when i switch to the extra code.

Oh, arga, were you using MS2 or MS2E? It's my understanding that MS2 used X-tau where MS2E uses EAE. Both similar, but EAE is supposed to be easier to tune. If you used EAE, I'd like to see what you ran for comparison.

Update: I did some research and the more I learn, the more I learn I don't know shit. :) This takes A LOT to setup properly. To say the least, I'm winging it.

I first had to turn off all my acceleration enrichments and dial in the VE table perfect to my AFR table. Well, I had "added" some acceleration enrichments into the VE table in the low RPM, high load cells (ie-off idle riding the clutch in first gear from a stop) so I had to remove these and let MLV smooth the map out. I surprisingly "guessed" the correct values for these cells before letting MLV adjust them because the max cell change in this region was 1-2 points. :)

Now I turned on the EAE accel settings after adjusting them to the ones someone with an msq titled "95 miata" uses. Then I drove it for a while. Drives like ass so far. And it's very weird how you tune the EAE settings. EAE does all your transient fueling needs, so you have to go from steady state, transient, then stead state for several seconds so that any EAE that kicked in will fade away and you go back into true steady state without any EAE kicking in. So this means roll on throttle at some RPM to some load and hold it there, as to achieve steady state, then letting off and waiting for it to again go to steady state idle. Right now the EAE setting make it drive like ass. But after reading and seeing some people's datalogs who've tuned this stuff, I'm going for it. Some can go from say cruise and just slam the throttle open and AFR's drop straight to target with no deviation at all.

I used MS2 w/ X-tau. When I switched to Extra, EAE sounded too much like X-Tau so I've never tried to get it working.

By rolling idle I mean the idle speed fluctuates. This usually happens after a hot start. I still haven't completely tuned that out but it goes away in about 30 seconds so I haven't really been worried about it.

Ah, I gotcha. FWIW, I used to have a weird idle fluctuation problem sometimes too. It was caused by the timing table. I smoothed it out in the idle cells and it fixed it.

I'm hoping this EAE stuff works out. It sounds neat. I'm manually smoothing my VE table right now and I"ll fine tune it some more tomorrow before I try to mess with EAE any. I read several times that the VE table must be right to get EAE right.

Got my VE table all dolled up. Will run this through VE analyzer tomorrow to fine tune it, but it should be very close. We'll see. It was actually too rich in boost before. Smoothed a lot of stuff so hopefully this further improves drive ability. I know some here have crazy-whacko ve tables because they never manually smooth them.


Attachment 207990

I've got a shinny nickel for the first person who finds the injector latency for low ohm RC-Engineering 550's. :) I found something that said it's 0.17 ms for low ohm RC 750's at 15V. That's crazy fast. I've currently got mine setup for 1 ms.... Probably wayyy to high.

Maybe this here...

Injector-Rehab - Fuel injector cleaning, flow testing, and complete blueprinting service.

RC550 @ 15.0V 0.46

Now Pat - as I also do run the MS2 - I would be interested to see the EAE working. At the MSextra.com board they seem to have quite a good success with this. Maybe you should post a log there, some of the guys there know a lot about all this...

Greets

+1 interested in your success so that I may steal the config later. :naughty:

What pat has is the peak and hold Denso 550cc injectors. At 12v, 0.50 is the opening time for those.

edit, nevermind ... the RC 550 low ohm. That list has RC 750 and RC 1000 and the opening time for those paints a picture:

RC 1000 3ohm = .49ms @12v
RC 750 3ohm = .38ms @12v
RC 550 3ohm = .??ms @12v My guess is around .31ms

They're Desno? They say RC-Engineering on them.

For anyone that cares, attached are the settings that I had on MS2extra when I bought the car. I can't verify exactly which firmware was run (I believe standard 2.0.1 code, not any kind of beta). Hopefully this will help someone!

Hmm, that's the exact settings I got out of somebody's msq called "95 miata". I'm running those settings for added to walls and sucked to walls, but I was using the base settings for AWC and SOC. I just threw yours in a new msq and I'll try them later. :)