MS2 Gammae calculation
 

I'm currently fighting an issue of my perfectly good VE tune being complete cock the next day. I've checked my logs and some days Gammae settles to 100-101% when fully warm. The next day it sits at 107% when fully warm. Is there any additional enrichment settings that I could have over-looked? I have MAT correction set to 100% at all temp ranges.

Attachment 193045

This is with a completely hot engine having auto-tuned it again to make the AFRs good. Prior to this I was idling at ~13:1 because Gammae is adding 7% more fuel. It means I have to auto-tune EVERY day in order for the table to come good again :loser:

My MSQ is attached but my MLV log was too big to upload.

Thanks,
Craig

on the log above it shows air temps increasing enrichments by 5% alone, something else is adding the other 2%

could be the ASE percent table.

What could it possibly be? I've been through all the settings and can't see anything that would cause the extra 2%?

mines always around 101%. doesn't really effect me.

Could it be related to the position of my IAT sensor that is in the intercooler end tank? Almost a cold-soak effect?

could be...im not sure where on the code the MS wants to add fuel. reduce the MAT-Based correction number from 100 to 50%, and see what happens. that reduces the effect of the ideal gas law code.

Cool, I'll try that and report back.
For a given ambient temp change, I'm seeing too aggressive changes in AFR. So if I'm hitting stoich at idle for 10*C, the next morning ambient temps are 5*C and my stoich idle is now too rich at 13.8:1. The same happens the other way if ambient temps rise.

Just been out for a 30 min drive during which I changed the value from 100% to 50%. The idle AFR went from 14.7 to ~15.7 and gammae/Gair went from 106/104 to 104/102 during this instance. Obviously I needed to re-tune but hopefully this may be a step in the right direction if indeed the MAT correction value is now less aggressive.

Subscribed.

Craig, you're doing all my hard work for me :P

Haha, I'm trying to tune it as efficiently as possibly. I don't want to move on to AE/spark until I'm 100% with my VE.
I also did some searching on the topic (off forum) and it seems lots of guys run ~ 50% too.

Got any links?

now you should tune your MAT curve to fine tune how you want enrichments and how the software forces them.

I did notice at 47*F AIT, I'm at 103% this morning, so the code is probably starting to add fuel around that temp. I know it pulls 1% at 87*F

I found a forum with Rover V8's and also this MS2 Punto's ;)

Post 2 RX7.


MSExtra It seems experimentation is the only way to sort it.

I mean it says what it does:

This value allows you to scale down the built in air density (MAT) correction.

100% means that the PV-nRT ideal gas law is followed.

Lower values give less aggressive correction.

0% is not allowed as it ignores the physics completely.

(0% is interpreted as 100% for backwards compatibility) This value and the table are used at the same time.


and like I said, if I sit there and let me car idle and heat soak. at 87*F I can watch it add 1%, by 95*F it's added like 5%. I cant quite remember exactly. Tune your MAT correction curve to combat as needed.


With that said the curve I'm using is:

40°F 5%
50°F 0%
78°F 1%
83°F 2%
101°F 8%
121°F 16%


MAT correction value of 50%

Here is a good read. The thing I'm having trouble with is consistency.
What would be the optimum method of tuning VE right? I'm sick of having to datalog and tweak every day. I want to enjoy the drive, not monitor it. I understand that it's probably not that crucial to have the IAT sensor seeing the absolute true air temp entering the cylinders so long as there is consistency. Consistency means that it can be tuned out with the IAT correction curve.

1. Autotune VE for a few hours in one big hit with IAT correction off and a MAT-Based correction figure of 50%.
2. Day by day adapt the IAT curve as temps fluctuate and try to achieve your previous well tuned map?
3. ?
4. Profit!

It all seems a little ham fisted and not a hugely accurate basis of correction with only 6 temp bins to plot a correction curve off. I wish the IAT correction applied by MS was actually accurate in the first place and then would only need minimal correction.

pretty much what i did. i just noticed i would get really lean at idle in warmer afternoons, but be perfect in morning. I watched the code (with GAMMAE gauge on) start removing fuel after 87*F, so I tuned the MAT correction table to keep the AFR more consistent and keep the gammae correction back down to 100%.

anyone wonder why these hacks are required?

aside from the possibility of an IAT sensor heatsoak issue, the "ideal gas law" based corrections SHOULD be sufficent. Clearly, that's not the case as they've made it possible to counteract the adjustment in MS-1, and made it even easier to do so in MS-2.

as a matter of principle, I'd like to understand why such correction needs to exist. In MS-1 i verified that the IAT correction curves do in fact follow the ideal gas law as far as percentage reduction is concerned, so it's not like someone made an (obvious) coding error.

FWIW i used Braineack's adjustments he posted for the MS-1 with fairly good results. I believe it was 4% VE adjustment ever 32 or so degrees F counteracting the IAT adjustment.

I'm now encountering this issue with the MS-2. The map's been tuned in reasonably well, and my 20 minute drive to work today did a pretty good job fine-tuning at X degrees temp.

This afternoon i noticed V.E.A.L. was enriching pretty evenly across the map, so I gave it a shot and set the air temp adjustment % to 90%, discarded all the recent changes, and re-enabled V.E.A.L.

After another 10 minutes of driving i still had changes (on 'hard' setting), but they were evenly distributed between enrich and enlean.

Anyone really dial it in where tunerstudio's not trying to change the tune based on ambient temperature??

all i know is i never touch my tune. havent since march.

I think these hacks are required because 99% of us don't fit the GM sensor in the actual intake manifold (because of heat soak). The code alone probably works amazingly well in that situation. Most opt for an easier install, that still allows consistency and the ability to hopefully tune out the aggression of the changes.

While air is not an ideal gas, the biggest reason, I think, is because we do not have an ideal temperature sensor available. The equations would probably work fine if you had a direct reading of the mean effective temperature of the air at the intake port. Instead, you have a sensor a bit of distance from the intake port, so it can't see downstream temperature changes. It's threaded into a pipe that is probably not the same temperature as the air, so it's affected by conducted heat. Depending on how different the temperature of the pipe is from the air temperature, the air temperature may not even be uniform.

When you've got that many factors to deal with, a physicist might try to model all of them and add extra equations to account for how they change the temperature. An engineer adds a fudge factor and says, "Good enough."