MAT/CLT Correction and MAT Density

[VLC-05L]

I was having a look at this in the tuning software for an MS2.

Do people use this feature? If so, what are typical values that people use in the table?

[VLC-05L]

The reason that I was looking at this is because I have had a feeling that the IAT correction isn't right. From day to day the AFR of the car changes more than I think that it should. The only thing that I can think it is is differences in temperature.

I had a look at the IAT Air Density Table. This is effectively the only thing causing changes in fueling in my tune at the moment related to temperature. I see that the curve is based on the Ideal Gas Law. Talking to some friends that are tuners about air temp compensation, they have said that the curve I have is way too aggressive. Do others run this curve as it comes standard?

[billr]

I use the default, as I have no way to determine what, if any, changes to it are appropriate. Without dyno time, and means to vary the ambient temp/baro/humidity while on the dyno, I would have to do a lot of guessing.

Debate on how to best handle the MAT is one of the more popular and contentious subjects, you will find a lot of threads on it; and good luck at coming to any solid conclusions!

[nathaninwa]

I've tuned the mat correction table. Took a full year to do so. I spent a spring afternoon where the temp was a constant 65* or so, auto tuned for about 3 hours. Then as the temps and season changed I kept an eye on ego correction. I worked with the mat table to correct. Mines less aggressive than the ideal gas law and that car stays with in about 5% ego correction year round

[mrx79]

Hi together,

by physics the ideal gas law is correct, but as a tuner for OEM car's i know that there are other factors then just plain physics.

These are:
1) MAT sensor position is critical
2) MAT sensor heat soak, especialy in very warm soaked conditions the signal from the sensor might be very off depending on it's position
3) based on valve overlap, the inert gas back flow into the intake makes another heat source, that the sensor might see (add to point 2) or not see depending on it's position.

Nr. 3 leads also to the fact that many OEMs have their MAT/density correction corrected via engine speed and engine load and or valve overlap.
The same applies also for the warmup correction due to similar (but slightly different) reasons.

So in my oppinion we need a more sophisticated correction function of have to deal with tollerances which we can't fix with the current dependencies.
Some engine config's might be more effected then others i guess too...

Best regards

[turbo conversion]

I've tuned the mat correction table. Took a full year to do so. I spent a spring afternoon where the temp was a constant 65* or so, auto tuned for about 3 hours. Then as the temps and season changed I kept an eye on ego correction. I worked with the mat table to correct. Mines less aggressive than the ideal gas law and that car stays with in about 5% ego correction year round

This is exactly how I tuned MAT correction (older firmware MAT only not MAT/CLT) and after a year (my constant is 70*) at any temp EGO correction is 8% or less.

David

[VLC-05L]

Thanks for the replies.

I read a fair few topics on here and it seems that those who have actually looked at the effect temperature has on their AFR have come to the conclusion that the standard curve based on the ideal gas law is too aggressive.

One topic I read was this one: http://www.msextra.com/forums/viewtopic ... 31&t=58903 - I see you replied in there billr

I don't have time to tune this properly, but I want to change it based on the above link. The original poster there gives half and quarter normalised curves. He states that quarter normalised worked best for him. For those that have tuned MAT Correction, is their curve closer to the half or quarter normalised curve? Or completely different?

[tschultz]

2017-06-23_noBARO fullDEFAULTmat NEWautotune65Fsmoothed.msq

I've tuned the mat correction table. Took a full year to do so. I spent a spring afternoon where the temp was a constant 65* or so, auto tuned for about 3 hours. Then as the temps and season changed I kept an eye on ego correction. I worked with the mat table to correct. Mines less aggressive than the ideal gas law and that car stays with in about 5% ego correction year round

This is exactly how I tuned MAT correction (older firmware MAT only not MAT/CLT) and after a year (my constant is 70*) at any temp EGO correction is 8% or less.

David

Can you share your MAT curve?

I am going through the same issue and am not sure how to attack it. Default curve just doesn't work for me with GM IAT sensor.

I tune my VE table at about 65F and then try starting the car in the morning on the way to work. In the morning (55F), EGO control adds ~12-14%. In the afternoon on the same day (85F), the entire map is lean, with the worst being below atmospheric and near idle.

I am at 5000FT in Denver, so I am wondering if the MAT curve is too aggressive for my altitude. FRustrating as I have been having this issue for the last full month.

[tpsretard2]

i have been waiting and hoping for a table, even a small 6x6 table for air temp correction for fuel and ignition based on coolant temp and one based on air temp.

IAT correction based on load vs rpm, This is the only way to do it correctly. At least for an ECU targeted primarily for Road going vehicles.

[tschultz]

Sure-- that makes sense.

The coolant based doesn't work because the you can be at full 180F coolant temp but sitting in 15F degree temperatures. Also if you adjust the mat curve to compensate for added fuel in the higher temps, then the boost region of the curve is not quite right because fueling is exaggerated due to MAT correction.


This has been a major let down for me with MS2 as I no longer have consistent fueling throughout the day. I have tuned at 65F and then my map is lean throughout in 85F+ weather and rich throughout in 45F temps. I have played with Baro correction, no correction, injector dead times, and still having the issues.

I had noticed with the curve closest to default MAT correction (although scaled down), it seemed like the correction was slightly less, although still incorrect...

I have done 10+ hours of reading on the subject and still haven't found something that works for me.
More thoughts appreciated.

[turbo conversion]

This works great for my set up, you just have to try different settings to get it correct.

With IAT from 25* to 130* ego correction never goes over 8%, I am good with that.

David

[cactus]

I think there is a variable that simply can't be included in computations and will always require some EGO corrections.

You can have the same air inlet temperature and coolant temperatures but a hotter or colder inlet tract. Most installations recommend measuring away from the manifold, but how can you account for varying preheat at different conditions? Some can be transitions but if you are doing a lot of low speed stop and go driving you can really heat up under the hood even if the air coming in is cooler. What about hotter or colder fuel at the same time? Nothing accounts for that, so you may always be fighting being impossible to say why it sometimes runs different at seemingly same conditions by only air and water temperature measurements. That seems to be within the range of some EGO corrections? I have trouble seeing how the MAT correction can even handle this condition which could exist at any ambient temperature, really.

I think having an incorrect injector dead time probably can have a much bigger effect on how the EGO correction correctly or incorrectly applies? I feel that my troubles with varying AFRs in seemingly similar conditions diminished as I fine tuned the dead time and after more work on the VE table as well as incorporating the EGO correctIons. Initially I wasn't using it.

I spent some time and iterations working on dead time, focusing tests on heavy over run conditions. If you can't get it to run leaner or richer in over run conditions, it seems to me that the dead time has to play a big part and at the lowest pulsewidth conditions, it makes the biggest difference. I used a long downhill grade where you can be in near steady load closed throttle and get to conditions where I couldn't get it to run any leaner and also so lean it had to be flowing no fuel.

Auto tune would pull the VE down off a cliff and then couldn't pull it back, being out of range. Too rich then all of a sudden too lean when the dead time was too high. I simply tried variations shooting to hit a 14.7 AFR and found I couldn't get away from one extreme until I seemed to be balancing on the edge where I wasn't sure which way I should adjust.

[nathaninwa]

here is the curve I created up there in the PNW. I run this table in all my cars and the installs I do

[5.0Thunder]

interested in this topic as well. I've seen where people suggest moving the 100% point to a common warm cruising MAT value. For me that would be about 130°F so I shifted the whole curve to where 100% was at 130°F. Seems to be nearly opposite of what was just posted above me.

[kaeman]

My correction table is very similar to the one posted by Nathaninwa.... I got it from motec instructions years ago. its what they recommended for a naturally aspirated high compression engine. I used it on my engine when I was running the hilborn injection (ITB's). It has been working well for years...

[5.0Thunder]

what about cars with boost?

[nathaninwa]

That is the curve I still use. For others it might need skewed a little based on boosted temps, but living in the PNW and intercooled my temps are pretty much always in check. You can see how I flatlined the curve as the temp gets hotter, this richens the mixture for you

Just this last weekend with 80* ambient temps, I've launch around 110* and finish the run near 90* around 200kpa

[5.0Thunder]

Mine was basically the stock curve shifted to where 100% was at 130°F. I checked a log and see that my cruise temps are around 130 and in boost it drops down to 115-120. Should I take your curve and shift it up to where 100% is near 135 or is that too much?

[jsmcortina]

Really, 100% should be at standard temperature (20C or 70F)

James

[5.0Thunder]

I guess that's the ideal case? Any reason why people tend to have better luck shifting it?