Math Wizzards - BEN Factor IAT correction

[TAQuickness]

I performed an experiment today with the BEN factor. My car had been baking out side for the better part of 9 hours when I left work. I flashed the changes from this mornings log, fired her up, let her come to full operating temp, then started logging. I noticed the BEN factor at idle was 1.02. No biggie, I'll just fix it when I get home. When I got home, I started a new log while I was idling in the driveway. BEN factor 1.00.

I went back to the data tab of both logs and noticed: The first log had an average temp of 104*F and the second 97*F. Even better, this mornings log had an average IAT of 84*F.

So now I'm wondering, is there a mathmatical formulation that can be done to correct the BEN factor to a given/fixed IAT?

My VE Table is close, but I feel like I'm chasing my tail now...

[Blacky]

The BEN factor shouldn't need to be changed to compensate for changes in charge temp. (charge temp being the temperature of the air that enters the cylinder).

What needs to be done, is the PCM needs to adjust the VE table values (on the fly) based on changes in the charge temp.
The PCM does this by estimating the charge temp using the IAT and the ECT and the estimated airflow. The estimated airflow is calculated based on RPM, TPS, MAP etc.

More discussion here:
http://efilive.com/forum/viewtopic.php?t=1829&start=14

I'm not sure if you have access to the charge temperature calibrations.
If they are in the version you have, you can find them in:
Engine calibration->Fuel->Airflow
They have only been added recently to one of the internal pre-release versions. That version should be publicly available very soon.

Regards
Paul

[SinisterSS]

If your BEN is running 1.00 to 1.02 - calibrate your MAF and be done with it man! As you said, stop chasing your tail. :D

Disclaimer: The opinions expressed here are just that and are not a reflection of EFILive, its management or subsidiaries.

[TAQuickness]

If your BEN is running 1.00 to 1.02 - calibrate your MAF and be done with it man! As you said, stop chasing your tail. :D


Yeah you right... but the anal side of me is not gonna let this go :D

Besides, the car runs so awesome in SD, why fix what ain't broke?


The BEN factor shouldn't need to be changed to compensate for changes in charge temp. (charge temp being the temperature of the air that enters the cylinder)...
Regards
Paul


I agree the PCM should do that, but...

My wife and I just got done with some figuring... Using the formula to calculate density, we came up with a factor. We applied this factor to my logged average data and the numbers add up.

I will put together a calculated PID in the next few hours. Anyone care to help me test this (knowing good and well this might be a wild goose chase)?

[TAQuickness]

Paul - I just finished reading the link in your post. Very good stuff. Now I'm on the edge of my seat to see the new tables. When do you anticipate the release of 7.2.3?

[Blacky]

Paul - I just finished reading the link in your post. Very good stuff. Now I'm on the edge of my seat to see the new tables. When do you anticipate the release of 7.2.3?

It was supposed to be ready about 3 weeks ago. But testing has shown up some issues that are still being fixed. I new beta version should go out today - **if** that proves successful then we will make a public pre-release available in a few days.

Paul

[TAQuickness]

Paul - After talking my self into a giant loop - you're right. Eventually all the data collected by the SD tuning will set the values in the VE table relative to absolute density...

[joecar]

Paul - After talking my self into a giant loop - you're right. Eventually all the data collected by the SD tuning will set the values in the VE table relative to absolute density...

I know sleep on my feet (wth eyes open) all the time, but did I miss part of this...?
:shock:

[TAQuickness]

Paul - After talking my self into a giant loop - you're right. Eventually all the data collected by the SD tuning will set the values in the VE table relative to absolute density...

I know sleep on my feet (wth eyes open) all the time, but did I miss part of this...?
:shock:

It's in the link Paul posted up there ^^^

Even though I almost convinced myself that everything would work out, I had to run a test. My laptop has a dead battery right now, but I'll post the pics of my experiment when I get it charged.

Things aren't working like I think they should

[BowlingSS]

I performed an experiment today with the BEN factor. My car had been baking out side for the better part of 9 hours when I left work. I flashed the changes from this mornings log, fired her up, let her come to full operating temp, then started logging. I noticed the BEN factor at idle was 1.02. No biggie, I'll just fix it when I get home. When I got home, I started a new log while I was idling in the driveway. BEN factor 1.00.

I went back to the data tab of both logs and noticed: The first log had an average temp of 104*F and the second 97*F. Even better, this mornings log had an average IAT of 84*F.

So now I'm wondering, is there a mathmatical formulation that can be done to correct the BEN factor to a given/fixed IAT?

My VE Table is close, but I feel like I'm chasing my tail now...

I am also seeing a difference in the Ben factor with different IAT. Do you think we should only log values that are below a certain temp? What do you think that number should be? We could apply a filter to weed out the higher IAT counts..

Bill

[SSpdDmon]

Well, if spark advance has an effect on your AFR, wouldn't it make sense that higher IAT temps where timing is being pulled will have an effect on your BEN's? My IAT's don't start pulling timing until mid-90*F. Thanks to my Austin Ram Air and the Michigan weather, I rarely see temps in excess of 93*. Maybe that's why I don't notice this so much when I tune?

[Black02SS]

I notice the same effects. I have my IAT's zero'd out so there is no difference in timing.

[TAQuickness]

I think I know what it is... I just need some time to figure out how to explain it.

[TAQuickness]

The attached screen shots are 4 logs taken over the last 2 days with no tuning changes between logs. This is an apples to apples comparison.

In chronological order:

Afternoon1:
AVG IAT: 100* F - 37.7* C
AVG ECT: 194* F - 90* C
AVG BARO: 30.0 inHg - 101.6 kPa

Morning1:
AVG IAT: 84* F - 28.8* C
AVG ECT: 188* F - 86.6* C
AVG BARO: 30.1 inHg - 101.9 kPa

Afternoon2:
AVG IAT: 101* F - 38.3* C
AVG ECT: 199* F - 92.7* C
AVG BARO: 30.1 inHg - 101.9 kPa

Morning 2:
AVG IAT: 84* F - 28.8* C
AVG ECT: 192* F - 88.8* C
AVG BARO: 30.5 inHg - 103.3 kPa

In each screen shot, the first grid represents Cell Count, the second grid represents my AVG BEN factor, the third grid represent my AVG RPM, and the fourth grid represents my AVG MAP kPa. All data has been filtered to remove values below 178* F ECT, and to remove values where TPS changes more than 5% in less than 100ms.

[GMPX]

I too have been tearing my hair out with this issue.
Before I go into why I think this happens basically you just need to be aware of the problem and try to tune with consistant IAT values, sometimes that can't be done so just filter out any data above say 25 degC IAT, sorry, I know most of you talk in degF, but I am used to degC, besides, the PCM's own calculations are in DegC :)

I am currently running SD mode of course and I have found that between IAT values of 20degC and 40degC there is a big change in AFR's at low airflow (idle etc). Changing the Charge Temp tables does not seem to have the desired results of correcting this problem. :cry:
I have seen the AFR at idle almost locked on 14.63 in open loop, yeah, my BEN factor at idle is like 1.00 :) , but then at 40degC these AFR's can drift to high 15's, not good. Again, this only seems to be at idle.

What I have done is allow the PCM to go into closed loop at idle and part throttle only, no good for tuning, but once the VE is mapped in open loop, I've enabled closed loop just under those conditions and this problem of IAT AFR drift has gone. This makes me think, perhaps the PCM is only factoring in the IAT compensation in closed loop?, I really don't know.
The problem of AFR drift does not appear to effect WOT AFR's, they have been spot on on every log I've done hot or cold IAT's.

Cheers,
Ross

[SinisterSS]

...besides, the PCM's own calculations are in DegC :)


That doesn't mean it's right. :P

[bink]

GMPX,

Have you tried bumping the "Open Loop Lean Limit (Idle)" value {B3607}. Stock is 1.00 (Eq Ratio).

Cheers,
joel

[GMPX]

GMPX,

Have you tried bumping the "Open Loop Lean Limit (Idle)" value {B3607}. Stock is 1.00 (Eq Ratio).

Cheers,
joel

That would affect the commanded AFR only, the thing is the PCM is still commanding 14.63, problem is it does not realise the actual is in the 15's when in Open Loop without any O2 feedback.

Cheers,
Ross

[GMPX]

...besides, the PCM's own calculations are in DegC :)


That doesn't mean it's right. :P

You guys :lol: :lol:

I guess I better learn imperial before we all head over for SEMA, I don't want a units interpreter on the stand!!.

Another idea, Paul, can you put a 'bug' in the software to force metric only, then you guys might just see the light :wink:

Cheers,
Ross

[bink]

GMPX,

Have you tried bumping the "Open Loop Lean Limit (Idle)" value {B3607}. Stock is 1.00 (Eq Ratio).

Cheers,
joel

That would affect the commanded AFR only, problem is the PCM is still commanding 14.63, problem is it does not realise the actual is in the 15's.

Cheers,
Ross

Yes, I understand. :) It was near 100*F here over the weekend. I bumped the {B3607} value to 1.13 and it helped. 1.30 and it was pretty stable. My point being -> could a commanded idle AFR below 14.68 put it in a range that the PCM is better able to compensate??

Cheers,
joel

[bink]

Another idea, Paul, can you put a 'bug' in the software to force metric only, then you guys might just see the light :wink:

Cheers,
Ross

Metric is a helluva lot easier once you're used to it. My .02.


Cheers,
joel

[TAQuickness]

I too have been tearing my hair out with this issue.
Before I go into why I think this happens basically you just need to be aware of the problem and try to tune with consistant IAT values...
Cheers,
Ross

I've got no hair, so we're close to the same boat :)

I'm pretty sure the PCM is doing what it's supposed to, and that the problem is not related to the PCM and environmental conditions.

I think the problem is how we are applying the BEN factor to the VE table (not a bug in the software by anymeans).

This relates directly to the screen shots I posted. All of the data we are logging is average data. So to apply the avg BEN factor to the VE table correctly, we would need to apply it to the respective avg RPM and avg MAP kPa. In most cases, this means the BEN factor would need to be applied to an interpolated value in the VE table.

If you look at the screen shot's in my previous post, you will see, the BEN factor is near identical from map to map when the average RPM AND MAP kPa is near identical.

However, if you look real close, there does appear to be an anomaly to this theory.

Another thing I have noticed. Of the 4 log's, afternoon 1 is inconsistant with the the following 3 logs. This makes me think the PCM, despite being in OL SD, is still going through some sort of a learning process.

Degrees C, Degrees F, you say po-tay-toe, I say po-tah-toe. One's thing for sure, it's effing hot in Houston right now!

[bink]

I've been open loop about 2 years.
My idle AFR has always leaned out (FJO Wideband) as the intake air temp increases.
I had to rescale the maf in the appropriate lower hz. in order to get the AFR down around 14.7 at idle (narrow LSA cam).
With the idle AFR in the 14.7 range the IAT seems to have less of an effect than it did when it was in the 15-16:1 range.

FWIW.

Cheers,
joel

[TAQuickness]

My idle AFR has always leaned out (FJO Wideband) as the intake air temp increases.

Scratching my head, but as IAT goes up, shouldn't your AFR go rich, not lean?

I'll be collecting some new data today. Hopefully the results will be consistant.

[SSpdDmon]

My idle AFR has always leaned out (FJO Wideband) as the intake air temp increases.

Scratching my head, but as IAT goes up, shouldn't your AFR go rich, not lean?

I'll be collecting some new data today. Hopefully the results will be consistant.

That depends on how the temp. variable affects the equation the PCM is using to calculate fueling. If they just got close with their curve based on an expected range for operating temps, it's possible the further you move from the "sweet spot" the worse the fueling calcs. get. After seeing what my real VE table looks like versus what they set it at from the factory, I'm hesitant to believe there is much in our PCM's that is spot on. Ideally, you're thinking along the right path that as the air get's warmer, it becomes less dense and therefore requires less fueling. But, because IAT is a variable in the equation, it shouldn't matter assuming the equation is correct. But, I doubt that's the case. Too bad we couldn't tweak that in EFI Live...

[TAQuickness]

Too bad we couldn't tweak that in EFI Live... :)


I just got done with my second log after a new tune. I'm really starting to think the it takes the PCM a little while, a drive cycle or 2, to adjust after flashing.

Paul, Ross, and/or who ever else gets way into it - Does the PCM store any history of barometric pressure or IAT or other environmental condidtions?

[Black02SS]

Too bad we couldn't tweak that in EFI Live... :)


I just got done with my second log after a new tune. I'm really starting to think the it takes the PCM a little while, a drive cycle or 2, to adjust after flashing.

Paul, Ross, and/or who ever else gets way into it - Does the PCM store any history of barometric pressure or IAT or other environmental condidtions?

I know what you are saying. On my personal car, I will start off from a cold start at about 63-64psi of fuel pressure. After the car is warmed up and has been driven for an hour or so, the pressure drops back down to the 58-60 range. Depending on where you tune the car at, it will effect the AFR I have seen.

[bink]

Too bad we couldn't tweak that in EFI Live... :)


I just got done with my second log after a new tune. I'm really starting to think the it takes the PCM a little while, a drive cycle or 2, to adjust after flashing.

Paul, Ross, and/or who ever else gets way into it - Does the PCM store any history of barometric pressure or IAT or other environmental condidtions?

Hmmm..... the idle airflow is stored and is temp referenced. I know nothing about the code/calibration though. Maybe it has something to do with idle airflow and not he fueling. Has anyone scanned pulsewidth?
Just a thought.

Cheers,
joel

[TAQuickness]

I believe I had the injector pulse width in the 4 logs the screen shots came from. I'm transistioning to a new laptop at the moment so it will be a while before I can take a look at the logs...

[bink]

I believe I had the injector pulse width in the 4 logs the screen shots came from. I'm transistioning to a new laptop at the moment so it will be a while before I can take a look at the logs...

gojo has said many times Idle AirFlow Parked {B4308} , if not corrected with modified TBs, will often skew the airflow at idle. We found this to be true with my car previously.

After thinking about this....... I think this is probably the case - if Idle pulsewidths are unchanged. Higher IATs result in greater opening of the AirFlow Parked {B4308} - you can check it by scanning Desired IAC steps or g/sec. Like most of these parameters there is a sweet spot relation between AirFlow Parked {B4308} and Idle Airflow {B4307}.
Nick and gojo are the main gurus in regard to this. (Yes nod Emoticon here).

[TAQuickness]

I did happen to log the IPW :wink:

Log files:
ftp://70.116.105.79/Autotuning/logs/

If you can't get them from the link, send me an email: TAQuickness@houston.rr.com

[BowlingSS]

I did happen to log the IPW :wink:

Log files:
ftp://70.116.105.79/Autotuning/logs/

If you can't get them from the link, send me an email: TAQuickness@houston.rr.com

How did you get those number at 15 kpa? I do not think I have any readings at 15 and only a few at 20.

Bill

[TAQuickness]

How did you get those number at 15 kpa? I do not think I have any readings at 15 and only a few at 20.

Bill

M6 :wink:

[BowlingSS]

How did you get those number at 15 kpa? I do not think I have any readings at 15 and only a few at 20.

Bill

M6 :wink:
OK...
Bill

[Dirk Diggler]

How did you get those number at 15 kpa? I do not think I have any readings at 15 and only a few at 20.

Bill

M6 :wink:


The only real transmission

[joecar]

The only real transmission

Debateable. :wink: :wink: :wink:

[TAQuickness]

Debateable. :wink: :wink: :wink:


Not really :lol:


Back to the topic at hand....

After racking my brain with the IPW's in the logs, the only thing I can come up with is the PCM is learning, calibrating, or something on the first drive after a flash.