MAT Correction
 

It seems like I'm consistently running lean (or at least my ego corrections are running high) when it's warm out. The VE analyzer just wants to keep adding more fuel the warmed it gets, and tries to pull it back out the next cool day. It seems like I need to mess with the MAT correction table, but the Megamanual has next to no info on how it works:
http://www.msextra.com/doc/ms2extra/...ion.htm#matcor

I'm trying to find some info to get started with, but haven't been able to find a good example so far. If I'm running lean around 120 degrees should I be using a positive or negative number in the correction table? Any idea what actual effect a number like 10% has on the pulse width values?

Do a google search of "what is mat correction for". I did it today my self as i was not sure how to set it but there are a few links that go to MSextra.com. They will explain how it works.

change mat correction value to 50 initially.

that will work pretty well for other than really hot starts at idle. then you gotta add (positive) values to the MAT correction table. but be careful not to add too much- i added a bunch to get my idle spot on, it ended up making all other conditions/WOT too rich, ugh

It's the internal ideal gas law code. I've already had it up with the developers about it.

thy scoff at the idea that this could happen because it violates the law.

I love MS, but it bugs me that the devs are so close-minded about stuff.

Similar with the PID code, Jason does PID for a living, yet because Ken can get it to work on his car there's nothing wrong with it.

I wish I was bright enough to be able to code.

Get up with that Mario guy over on MSextra. He's putting out modified versions of the code. He sounds pretty open minded about the code ideas.

Honestly, the PID works great for me. I've had no issues tuining boost, idle and EGO.

I'm sure if Jason wanted to contribute, would write code and sign it over to Ken royality free, they'd be more than willing to take a serious look at it. I can understand that Ken and James want to use their own code, it's their baby. They did take a bunch of suggestions and implement changes done by this Mario fellow, I'm pretty sure they discussed royalities offline.

And to be fair to Ken, he agrees that while the ideal gas law cannot be disproven, there are other factors in play (not just heatsoak) that aren't being factored in.


The best way to handle the ideal gas law code at the moment is to alter the MAT correctoins table in a way that reduces the coded corrections. Setting the MAT corrections value to 50% cuts the enrichments by half, and you can fine tune the rest with the table.

Way ahead of you...

sorry for bringing back such an old thread but im having a similar issue. get my tune dialed in perfectly... outside temp changes and my afrs turn into crap.

im running ms1v3.0 with extra029y4 on a 94 BP+T using miata cas and toyota cops.

are you talking about cutting the values in the offset side of the table in half or is there an actual MAT corrections value box that needs to be changed? im not really a computer guy and im having a hard time wrapping my head around this.

it cuts the coded ideal gas law corrections, that wreek havok, in half.

Made up numbers: So instead of a MAT of 100°F causing the MS to pull 5% of fuel, it only pulls 2.5% of fuel--and I still find that too aggresive imho.

What did you end up settling on? I have MAJOR problems with hot restarts after the car comes off track.

how close is your AIT sensor to the rad?

It's about 1-2" off the throttle body attached to some plastic piping.


e: it does come out of the driver's side of the pipe so it can catch radiant heat from the radiator. Not sure if it matters because the whole engine bay is completely heat soaked after a session.

oh it matters. The best way to defeating this is moving the AIT outside the engine bay, just post IC.

tuning your MAT corrections curve will solve it completely, but it's a band-aid solution that can cause other issues.

I've been able to introduce a few changes where needed in code... so what's wrong and what needs fixing???

in MSI the MAT corrections were able to decay out after a certain RPM. So you could use them to combat heatsoak related starting issues.

Before I moved my AIT sensor to my IC endtank, I could see the AIT jump from 70°F right after to stopping the car and climb to 90, even 100°F, within a few minutes.

Since the ideal gas law code is VERY aggresive, the car runs lean as crap when you start it back up...sometimes barely able to catch/idle.

You can solve this by tuning the MAT corrections table, where you pretty much end up negating the ideal gas law temp corrections.

The only solution I was able to work out with Ken and James was for them to allow me to disable and AIT corrections during ASE. I didn't like to solution. The problem is, you really need to start getting real airflow across the sensor in order for the temperatures to drop back down, this means more than idling. Typically ASE is only a few seconds, so it's not enough time.

I moved my AIT sensor to my IC and it's been much better to protect it against heatsoak, but you still see a delta after the car has been shut off, just not as much.

So, if I use the MAT corrections to add back in fuel when the temperature climbs past 85°F (I think that's where the first 1% is removed), and since I can't have the MAT corrections table decay as the RPMs increase, if I enter boost and actually see temperatures well over 100°F, then my fueling gets screwed up--I'm fighting the ideal code, when in this case would probably would well.

This is especially apparent on the dyno when the intake temps are greater than normal.

Ken and James pretty much refused my idea of adding in a RPM point where the MAT corrections were disabled, they were open to the idea of a MAP trigger point, but it was never implemented. I made the case that the MSI already has this, but they never wanted to add the same feature to MSII or MSIII. Reverant went ahead and already custom coded this cut-off point for his MSIIs, and it was actually one of the reasons he didn't want to jump on the MSIII bandwagon until the code was available for open source, to fix this specific issue.

i still dont understand how i can change the mat corrections to 50%

Then you have bigger problems.

But I probably wouldn't start by opening up the software and looking at the dropdown where the MAT Corrections parameters reside. I certainally then wouldn't see one labled "MAT Correction value." And I wouldn't open that dialouge window and see that I can input a value from 50-100% to change the scaling of the ideal gas air density corrections.

What about doing in hardware what you can't do in software? Use a space MS output to trigger a circuit to switch from an ambient air sensor to an in-intake sensor?

ive been looking in tunerstudio advanced>coolant related air density,all there is is the table. no corrections%, i went through the settings and changed it to

air density=corrected
corrections based on= iat corrected
but i dont see mat corrections %

ive been looking through it but cant seem to find what you describe. if you could be more specific where to find it... im retarded but i am trying to learn.

I don't have a copy of TS loaded on the PC that I'm typing this from, so I can't check right now, however I don't believe that the MS1 has this feature.

fastivab6tg25mr, I might try changing from "Corrected" to "Normal correction" and see if that helps.

Otherwise, the only solution I can think of for an MS1 would be to switch "CLT-related density", use IAT, over-ride the RPM taper, and enter a map which is inverse of the normal correction map at half-scale.

Tried that.

I had two AIT sensors at one point. The second under the driver side fender, with a little computer fan blowing on it. I'd use a relay with some RPM + TIME? (I can't rember) to trigger from one to the other.

doesn't really help, still gets hot under the hood...the issue is the aggressive code that pulls way too much fuel at any temp abouve 85°F

This is what I'm talking about. Air Density table switch to AIT, with the start/end correction RPM points set to like 2500 and 3000 respectively.

same function as the MAT corrections table in MSII and MSII.

thanks so much. i guess i just confused myself looking for something that wasnt there.

I was just clarifying that in MS1, there's no way to "open that dialouge window and see that I can input a value from 50-100% to change the scaling of the ideal gas air density corrections."

Us old-schoolers don't have none of this hoity-toity percentage scaling. :D

ohhh. yeah well that part is trivial. I want the decay.

not old school... just poor...lol... i cant wait for some extra cash to build an MS3X. i want sequential fuel and spark.

finally got around to putting in my 1000s from five0 so im setting up my corrections and i can't seem to get the CLT gauge to chane to the MAT even after i went into the .ini and changed it.

do i have to reload the firmware or something?

:td: MSI

but its all that i have :'(

does anyone have an idea of what im doing wrong? do i have to reload the firmware in order for the clt corrections to turn into mat corrections? i was trying to use the lil green ball as a reference like the box says but it is still following clt values instead of the mat values i set it to.

and it's set to AIT under advanced? the green line should be following whatever that is set to, regardless of what the gauge displaying in that window shows.

its set to IAT but its following CLT thats why im so confused. any ideas?

MS1 user here. I was able to remedy the hot starting lean condition using the coolant related air density table adjustments table and it's working pretty good now. Thanks for that help.

Question- When constantly in boost at the track on a very hot day in Miami, my AFRs were getting leaner in boost, presumably due to high AITs. Is the MS1 pulling fuel due to high AITs just like on hot starts? Can I compensate for or prevent this?

well on the MSI you have a decay to turn off the CLT-Based corrections after X rpm, probably set around 2500RPM, so the MS starts pulling fuel again based on the MAT. you may want to increase that value a little.

Increase which value, the decay RPM or how it pulls fuel related to the MAT? Is that the curve editor thing I read about?

FWIW, my car was pretty dialed in with autotune to run 12.2 AFRs at 140kpa and 11.5 at 170kpa between 4000-7000RPM.

And I was reading about 13.4 AFR at 6-7psi in the same RPM range on the hot track with AIT's above 105*F. Too much difference when AIT's go up.

You have to open up the project ini file in a text editor and change one of the settings to switch the gauge and table to display MAT rather than CLT. For whatever reason, it doesn't automatically switch when you choose MAT vs CLT and there is no drop-down to switch it.

http://www.msextra.com/forums/viewto...=25788#p186868

Just tried the above, found the curve editor in C:\Program Files\EFIAnalytics\TunerStudioMS_beta\config\ecuDe f\MSExtra format hr_10.ini manged to save it but the iat correction table is still showing coolant corrected. What am i doing wrong? Surely it should be iat corrected?
https://www.miataturbo.net/attachmen...ine=1338068849
https://www.miataturbo.net/attachmen...ine=1338068849
Thanks in advance for any help.

Sorry, I should have clarified: You need to change it in your project ini file, not the firmware ini file. Look for the mainController.ini file in the "projectCfg" folder.

You guys on here are awesome!!!!!!!!!!!! Managed to find the correct folder & edit the .ini file. It is now displaying mat & not clt:bowrofl:
All i need to do now is work out what offsets to use & all will be great again in my eunos world.
thanks guys.
Rob

I have a DIYPNP and ran across this thread and decided to check my settings. Im usind TS 1.006 MSextra 3.2.1 code. It has a mat non linear correction table and all the entries were at zero so I figured it was doing nothing. I then went and changed the mat correction value to 50% as recomended and my AFR readings all went rich. So I guess the correction value is in effect regardless of any values in the non linear correction table?

As an aside, my mat went down 7 degrees at highway speeds with the headlights up. I have a cone filter at the end of the stock MAF location with an aluminum heat shield. Dont know if this is news to anyone but it cant hurt to share.

I found changing the correction value wasn't a good idea for MAT corrections when the car was actually running. On cold days it went lean, switched it back to 100 value and all was good. Sadly we have no way to taper off the non-linear MAT correction table with RPM like MS1 can.....that would solve our issues I think for hot starts.

Some of the latest firmware changes allows you to disable MAT corrections during ASE among other tweaks, which should help, but I haven't got to test them out yet.

Im not having an issue with hot restart actually. I just started messing with the MS recently and have been doing lots of reading and figured why not try a suggestion. Honestly, it made a big difference that was not just noticeable in the AFR gauge but in AE as well. I havent gone back to see what the AFR is on cold start up so I cant say how much of a change it will make there. Of course cold start up is a relative term at this time of year around here since 80* is close to an overnight low. Also, dont know if my MAF's MAT sensor is properly calibrated. I used DIY autotunes resistance values so I would say theyd be as goos as any.

I added the "disable MAT corrections during ASE" in the MS2 mod that has this functionality. I guess adding a "non-linear MAT correction table with RPM" would be doable and so if you can ellaborate further on how this would work, I'll take a look at it.

G

You gotta set that value and forget it. If you change the correction value to 50%, you're cutting the ideal gas law code in half. So whatever is currently in your MAT corrections table essentially doubles.

I would like to taper off MAT corrections both by RPM and maybe by Boost*, myself. But I'd also like to keep the MAT correciton value at 50% as I still think it's very aggressive -- even at 50%.

But what you really need to tune the MAT corrections Curve by negating any Gair addition.

for example, with a correction value of 50%, this is the table I use:

https://www.miataturbo.net/attachmen...1&d=1338386973

This completely negates the ideal gas law AIT enrichments and keeps Gair at 100% throughout the temperature range. This way, since I can't decay out the enrichments, it hot starts arent an issue and neither is my AFRs in boost. I guess honestly it's the best way to go about it.

*boost decay is probably not needed if the Gair stays at 100%, I had issues on the dyno once where my MAT corrections were keeping Gair above 100% and I was dumping in too much fuel in boost. I don't seem to have this issue anymore.


I'm the one that convinces James to add that feature. The problem is ASE only lasts for maybe 10 seconds on my car? And that's not enough time to get the temps back down at idle. My AIT sensor is located just aft of the IC...however my BOV, which leaks at idle, is located just pre-TB. That's my main source of air at idle, so air does not flow over-top of the ait sensor at idle...so when it does get a bit heated up, I have to wait until I'm actually driving before it drops back down.

Even in the current position, I can see a good 20-30°F delta, on a hot day, within 5-10mins of sitting with the engine off.

But one thing that's a bit shocking is, last summer I was doing a good deal of testing, using a second AIT sensor in various locations under the hood, is that the AITs are pretty much whatever my "heatsoaked" sensor was saying. I had an input switch a relay to my second AIT sensor at any rpm below 1200RPM, it was clever, and I had strapped a small 1" CPU fan to the element to ensure airflow. I tried it directly behind hte rad, under the headlight, in the cowl, and under the driver fender...all the temps would read within 10°F of the sensor in my IC pipes.

This lead me to belief that the ideal gas law code, while technically/logical/mathematically LAW, is flawed when it comes to how it's implemented in the MS code...because the air temps rising from 70°F to 85°F do NOT and should NOT require such an agressive amount of fuel (~6%) to be pulled as the code wants you to pull.

But like I said, I just use the above table and really have no issues anymore, at least I have bigger fish to fry and I'm tired of bring it up with Ken and James.

Issue I had with the 50 percent correction value was.... tuned WOT for good AFRs on a 70 degree day (ish), then a few weeks later it was down in the 40s, and it ran lean consistently. Wasn't adding enough fuel. I set value back to 100 and it was pretty good, retuned and haven't observed any issues with temp fluctuations while in *driving conditions*. Idle is a different story. I almost wonder if IAT reads high in heatsoaked conditions due to small amount of air flow inside of pipe and lots of heat inside the metals all around the sensor, and that hot metal is having a larger impact on the sensor than it should. Anyone run IAT with the FM/all silicone piping? I'd be interested to see it performed in that.

I don't recall seeing anything crazy like 6 percent correction for a 15 degree F change in temps, but it's been awhile since I've drove the car.

I got to see the "Non linear MAT table correction decay with rpm" in action on a Reverant unit a week ago and it seemed to work great, never had any hot start issues while doing a bunch of idle troubleshooting...sitting inside a garage with hood open (it was getting hot). He uses 100 correction value as well.

Agreed that the ASE thing is a good step but really doesn't solve the issues that I've seen. My idle didn't "get better" until I started moving/increased load/rpm on the engine. and I dont have a leaky bov or anything at idle. My AIT/IAT whatever is on coldside IC piping, in front of radiator (over the radiator piping).

PS I PMed GSlender about implementing the RPM based decay, referencing some other threads on here and on MSextra.com- I'll be stoked if it actually makes it into his latest firmware. I've considered going to Reverant DIYPNP just to get the RPM based decay stuff...

Yup, I've never had a problem or complaints with my solution. AFRs stay the same in boost/vacuum whether it is winter or summer, and heatsoaking is no longer an issue on restarts.

sounds like you didnt have things tuned correctly. like i said, if you left the MAT corrections table untouched, but changed the MAT value it 50%, you'd have issues like the above.

probably a little better, but did oyu even read my schpeal about the dual AIT sensors?

Then you've never looked at Gair (the air temp based enrichments)

using a correction value of 100% is fine, just double the MAT corrections table I posted above. Get it?

I still worry about the ideal gas law code pulling too much fuel in boost with just this strategy implemented. I'd like to hear Reverant's take on it, but let's assume you have it decay out at 3000RPM, and you get into boost on a 90°F and get your intake temps up to 150°F. Once you decay off the corrections, the ideal gas law code will pull fuel an dyou'll be lean in boost...exactly where you DONT want to be lean.


I see the same thing. One that that helps extremely well is Seq. fueling...since you can still idle smooth at like 15-16:1

can you comment on my post about getting into boost with high AIT temps?

Braineack, have you tried setting the MAT value to lower than 50% ?

It is just restricted by the ini and the ecu code will take lower values down to zero% effectively turning the ideal gas law off?

G

My problem with the 50 percent correction value was it was not aggressive enough when temps dropped and Gair was going positive..and end result AFRs were lean. If I recall @50 percent, Gair was 101 and @100 percent, Gair was 103. The 103 value put my AFRs back where they should have been more or less. Forcing Gair to always be 100 thru the non-linear table would have made it be even leaner, right? I have not personally had issues on the high AIT side of things and 100 correction value being too aggressive and making things run lean. Maybe I have just not had high enough AITs...hard to say. This is all in boost/cruise conditions.

Bottom line I haven't seen issues with the 100 percent correction value for boost and cruise conditions...but that is just my experience, I'm not questioning yours. I also forced Gair to be a constant 100 like you showed above, but I saw it go lean/rich with temp changes, depending on if it was hotter/colder...as in I needed those Gair corrections for AFRs to be consistent.

My only issue is it going mega lean on hot starts in the Florida summer. Is it due to inaccurate MAT readings (your testing would indicate this is not the case), or due to code being overly aggressive.... IDK, but I think the RPM based decay would solve my issues (how Reverant approached it).

GSlender- what he's doing with the non-linear table vs being able to change correction value *should* be the exact same thing in the end.. which for ME....made hot starts not an issue, but threw off AFRs for everything outside of idle when temp corrections were needed (hot vs cold day). I basically would like a way to alter Gair at idle/below XYZ RPM independant of cruise/WOT.

I still don't think you quite understand the correlation between the MAT Correction Value and the MAT Correction Table.

What you are suggesting above it that your VE table was tuned incorrectly as it acutally needs MAT corrections in order to get the correct AFR. This is bad and will always lead to trouble when the temperature changes. You should tune your VE table when no enrichments are active, then you can fine tune the enrichment based on temperature condition and NOT the fuel table.

For clarification... I had tuned the VE table weeks prior in moderate temps of 60-70 degrees, MAT corr value 50. Minimal MAT corrections applied, probably 1 percent at most in cruise at those conditions? I don't think the VE table was the issue. Anyways, then temps dropped down into 30-40s and that's when I ran into the issues and had to increase the Gair corrections for the now lower temps I was seeing...it was not being aggressive enough. Never touched the VE table.

Could MAT based timing retard be playing into boost/high MAT fueling as well too? Or does the code somehow adjust for any timing it's pulling as a result of high MATs.

Either way I don't want to get sidetracked on this, I just would like to see the RPM based MAT correction decay as discussed above in MS2.

but what was Gair when the temps dropped?

I run the same VE table with 0% ait corrections from JAN-DEC...

if the VE table didnt change, but the AFRs gotta shitty when cold, it was due to a poorly tuned mat corrections table. Changing the value back to 100% probably just correlated better to what you had/have plugged into your corrections table...but you could have the opposite problem when the intake temps get higher come summer.

THIS sounds like the "high AIT - lean in boost" problem I am experiencing. Why don't I just set the enrichment to decay at 7200rpm and never need to worry about going lean and blowing up the engine in boost? 95 degree track days with 140 degree asphalt temps and then going into boost on top of it means lots of fuel being pulled when you need it most.

IIRC around 101 Gair @50% MAT value, and then it went up to 103-104 @100% MAT Value. Might have been 103 to 106, but either way, after the switch, AFRs were back to what they were, pre 30-40 degree temp drop.

This is all with zeroed out non-linear MAT table. I had messed around with that in the past, but it wasn't a factor in all of this testing.

Very interesting that you run same VE table year around with no air temp corrections to fueling... that has just not been my (limited) experience at all for when outside air temp changes significantly.

Wish I had a running car at the moment to mess around with this stuff, but I'm stuck just trying to remember back several months or more.


You have MS1? If so then yeah, you should be able to do that so that table is being used all the time to get your AFRs where you want consistently in boost, at whatever temp.

I think I've found a bug in the air correction calculations that might explain why it isn't working when it should.... stay tuned!!

Staying tuned