Braineack

iTrader: (62)

the ideal curve was just too much and the car would lean out with hot restarts and in boost.

cyotani

iTrader: (8)

so fuel rail temperature compensation curve is needed?

deezums

iTrader: (3)

No, the compensation table given is plenty to tune with.

DNMakinson

iTrader: (3)

To clarify, the old code had the ideal gas law temp correction built in, and the MAT table was to be a tweak. But, and I'm not sure of the reason, the ideal gas law was way off from reality of Speed Density control. So, on the old code, the MAT correction would basically be the inverse of the IGL curve to nearly back it all out. The new code has no built in correction. The MAT table is ALL of the MAT correction, applied directly to the fueling equation. The MAT table, in the Base Map Default, is modeled by the IGL, as Brain pointed out in post 13. However, you cannot leave it that way. You will find that you must flatten it out considerably to make things work correctly. Personally, I did not find that I needed to enrich at high temp, as OP is suggesting, My MAT curve looks a lot like the 2nd table in the 1st post.

Agreed, only Canadians and those in Siberia need to worry about -40.

stefanst

iTrader: (1)

I'm also running much less compensation than the ideal curve would suggest. Measuring charge temp in the cold air tank of the IC.
But this leaves me with a bit of a problem. My tuning was mainly done driving around town. What I'm observing now is that my compensation works really well for idle and part throttle, but in cold air and full boost I'm not compensating enough (running a bit lean), and on really hot days I'm running too rich - again, only at full throttle or close to it.
My theory is that due to the expansion at the throttle we get some cooling of the charge which is less pronounced at full throttle. So at full throttle I would need compensation close to the ideal compensation, but at part throttle and idle I need less compensation.

Does anybody else have that problem, or am I imagining things?

compuw22c

Wait, the new code has no built in correction? I thought this table was to tweak the IGL, which would explain why people are running almost flat curves (that don't take into account much change in density based on temp). I assumed there was an ideal gas law running in the background and this was to add or subtract from that. Otherwise I doubt many would be able to run with an (almost) flat curve.

Furthermore, how do you guys recommend tuning it? Is it easiest to do at idle cold at night then during a hot day? Or do you need to get out on the road, hold a given load cell (rpm & map), log what AFR you get when it's cold, then go back and match it when it's warm?

18psi

iTrader: (76)

ideally you want to tune the car in the middle of the table, keep those at 100, and then add/remove the other ends

my 80-100F range is flat at 100

aceswerling

I'm glad this question came up because I'm dealing with the same issue now that the weather has gotten warm. I was seeing about a 35% swing in EGO correction between a cold startup and hot running conditions. I set the VE so it was about 15% rich when cold and about 20% lean when hot, but that still left a lot of EGO correction going on.

I tweaked the MAT table and currently have it to 10% fuel when hot, which knocks down some of the EGO correction. I also had to crank up my ASE to have 40% enrichment when hot and run for at least a minute. I'm noticing that the MAT correction is helping under non-boosted conditions but is making AFRs proportionally richer under boost.

It sure seems like there's a nonlinear behavior between cold and hot conditions, just as other people are reporting. I can't figure out what else to tweak. From process of elimination, I'd thought it might have been in the Flow Force injectors I'm running. Nigel has been great in supporting me but he says he's not hearing problem reports from other people. We're talking about swapping out my injectors, but it seems like this is just the way the Megasquirt operates based on the information in this thread.

We're saying that tweaking the MAT density table is the best option, yes?

Braineack

iTrader: (62)

Ughhhhh

compuw22c

OK, I have no idea what that guy's doing other than giving me a huge ego boost right now. 35% ego? My max is 10% and I don't even like seeing it being used at all (though I'll tolerate it because it fixes my mistake). And it definitely isn't supposed to look like that, nobody's looks like that. I know because I've been searching every thread on it for days now. I'm not saying I knot what I'm doing, but that's wrong.

Back to the topic at hand, so I'll need to wait for an 80F day, autotune long and hard (that's what she said), and then correct any error later only with the mat density table. This makes sense, except one variable. The MAT will change just by going into and out of boost, correct? Yes I have it in my end tank where it belongs (sensor) but won't that temperature not being held at a constant temperature throw off the VE table a bit? Or is that some kind of diminishing returns area where close enough becomes good enough?

Braineack

iTrader: (62)

you need to tune your fuel map before you touch that table.

sounds like your tune is a mess. Once, on a typical day, you EGO basically sits at 100% (in general, its always going to be swinging around trying to keep you to target), THEN apply MAT corrections if you notice on a very warm day that ego is averaging like 95%.


It looks like your tuner is using that table to fix his inabilities to properly tune. You shouldn't be using MAT correction for tuning cold starts or heatsoak. If you have a heatsoak problem, move your AIT sensor. If you have a cold start issue, autotune freaking warmup enrichments.

18psi

iTrader: (76)

disable ego correction until you're done tuning

compuw22c

I assume most of this was directed at Aceswerling as I've moved my AIT (and don't have an issue with hot temp readings on restart, though I do have some heatsoak at the injectors/heated fuel rail it appears that I've tuned out with some ASE which seems to be working good enough...) and my warmup enrichments are good (or again, at least good enough I don't notice, AFRs look good)

My table is decently tuned, of course every time I change something large like this, it needs tweaking... It could use some work, as just yesterday I 100%-ed my MAT table again and noticed some leanness, so I scaled most of my table (minus idle and bottom row which weren't seeing much of a problem) up by 7% which was about what ego was giving it in trouble areas (from 5-10%) which I'll now let autotune bring back down to whatever it deems good through some tuning, but I figured better to be a bit too rich than too lean while tuning it.

I was under the impression this was unnecessary as auto-tune takes into account the ego correction in it's calculations to raise or lower a cell, no? I do try to leave it relatively low (or as low as I can) usually between 7-10%. I think I have tuned with it off before and didn't notice much of a difference, and would like it to prevent me going super lean in certain cells (out of boost) while tuning. I'm not dead set against disabling it, but I would like to know what's "correct" as I've seen it mentioned before that it's fine to leave it enabled.

Joseph Conley

There is some really good info in this thread that I have found hard to find. With that said, I feel much better about how I have been tuning. EGO has been completely disabled from start. MAT correction is 100% until around 120 degrees. My idle is good, hot restarts are good, and my AFRs under load are happy. Last two tanks of 93 I got 28 mpg.

DNMakinson

iTrader: (3)

Continuing this discussion:

I tuned my MAT corrections at cruise. Now, in really cold weather, I am leaning out severely at high boost / WOT.

To combat that, I am at about 110% MAT at 30*, and will likely need more at 15*. AFR safety works really well, by the way.

This kind of makes sense as we all see lots more power in cold weather.

Anyone else seeing this?

Braineack

iTrader: (62)

sounds about what i'd expect. I always suggest keeping MAT corrections 100% and then tune it based on the weather. sounds like you need the extra fuel at those temps which makes perfect sense ( take a look at what your EGO is doing and how much it's trying to add to get an idea of how much you should add in MAT).

18psi

iTrader: (76)

I usually have them basically flatlined 180-100F, then tapering up gradually to 110%, 115%, and some cars 125% at 0F or colder.

But it really depends on the car and location. The 2 cars I'm currently tuning around the NY area needed really aggressive curves, while the cars in AZ and CA are way milder.

DNMakinson

iTrader: (3)

Thanks, guys.