holy driver

The problem is exactly as Braineack said. It's not MS's inaccuracy, but rather the temperature sensor heatsoaking. The temp sensor itself is reporting the temperature to be much higher than it actually is.

JasonC SBB

FWIW the AEM has a table of "additional fuel vs. coolant temp" and "how many seconds to apply this add'l fuel, vs coolant temp", which is used to address hot re-starts and address IAT heatsoak.

The basic IAT vs fuel curve should be used only to get the basic speed-density system's air density correction correct.

ctxspy

iTrader: (3)

I know i'm going against the grain here but i'm with JasonC... Anyone know of a way to find out how much fuel the MS is actually pulling based on IAT? (kinda like a gammae or a gwarmup gauge)

anarchyx34

It's not entirely a heat-soak issue. It's 100F out. I started my car after it sat all day, meaning the IAT was reading the actual 100F. It still idled lean. Gair was reading 91. I set the coolant based air-density whatever crap to 100 across the board. It's still pulling fuel. How the hell do you turn it off?!

ctxspy

iTrader: (3)

lol... welcome to the club!

Gair is going to tell you the %fuel based on air temperature.. Technically, the MS is doing it based on resistance of the IAT.. This is what you calibrate when you run easytherm and generate a new firmware to download.

If you want to completely disable IAT correction, you could probably put a resistor in place of the IAT sensor with whatever value the gmiat has at 68*F and call it a day.
(don't actually do this).

The reality is you WANT it to add and remove fuel based on temperature, otherwise you'll end up running lean in the winter (denser air vs same amount of fuel) and way richer in the summer (same fuel, less dense air).

My issue, and apparently your issue, is that the MS is overcompensating.

Sample airdenfactor.inc courtesy of braineack:
http://www.boostedmiata.com/MS/airdenfactor.inc

According to his file, at 100*F, your MS should be reporting a gAIR of 94%, not 91%..

ctxspy

iTrader: (3)

FWIW i checked my car (having not run it in 2 days, sensors should be at ambient). I saw 88*F and gAIR 96%. This matches what is in my airdenfactor file, so that''s satisfying.

However, my idle was at around 16.5:1, and was lean from 1st moment thru completely warmed up.. In colder weather (tuned at around 60*F), i'd see around 13.5:1 while warming up, leaning out to low 14s for idle...

anarchyx34

Maybe it was 94%. I don't remember really. It's annoying though.

WestfieldMX5

I noticed this morning that autotune leaned out my fuel table a lot now that temps have dropped 15°C. Started playing with the MAT correction table trying to straighten things out.
I see mention of coolant related air-density correction table in this and other threads but I can't find it in Tunerstudio. Is this the MAT correction table or something else still?

Andreas

You will find the coolant related air-density correction table at the same place as in Mega Tune, advanced - coolant related air-density correction

Andreas

Edit: Don't you speak about the same table? (with optional coolant or IAT correction)

WestfieldMX5

Mmm, I don't have that under Advanced. Probably different in my version (3.0.3u). I have no option to choose between coolant and air either.

richyvrlimited

iTrader: (1)

You have MSII so it's the MAT correction table.

What Andreas is talking about is MS1's MAT correction ( a fudge of coolant correction).

Andreas

Ahh, my misstake then. I didn't realize we were speaking about MSII. Thank you for pointing this out Rich :-)

Andreas

WestfieldMX5

thx for clearing that up guys. Mat correction table is adjusted. Will do some logging when it gets hotter again. Not this week though, rain all week long :(.

WestfieldMX5

This leaves me with one question though. How do I switch to coolant based rather than iat based correction in TS? I have problems with hot restart due to iat heat soak.

richyvrlimited

iTrader: (1)

if it's the IAT heatsoaking, surely it's that you're compensating for?

WestfieldMX5

Well not really. Although my general fueling curve is corrected for air temp now, that curve doesn't compensate for heat soak. You always need to compensate for air temp, but at startup, the compensation should be zero, not whatever value the overcorrected built in formula gives us. Suppose air temp is 25°C and my sensor thinks it's 60°C, I still want the 25°C fuel curve, not the 60°C fuel curve. I can't adjust the 60°C value in the tabel because that would give false readings when it's really 60°C.
My first thought of using coolant temp instead of air temp is no good either because coolant temp is much too slow.
It seems to me that it should be possible to clamp the air temp value at startup to a user settable maximum. For instance 30°C. If the sensor gives colder readings, use those. If the sensor gives hotter readings, ignore them and act as if air temp is 30°C.