Pages: 1 [2]
Author Topic: MED9 diagnostic boost limit  (Read 1062 times)
Blazius
Hero Member
*****

Karma: +23/-11
Offline Offline

Posts: 575



« Reply #15 on: August 13, 2019, 08:28:04 AM »

It´s no voodoo-method. I found only one value in ps_w calculation which can be changed without being picked up on another place in this calculation. This goes as well for the other calculations
 
I can post logs, there are absolutley no asm-changes, as I´m to stupid for IDA.

But I´m actually unsure if I want to post the solution

Well then post it my friend, I am sure this forum  gave you a lot of info or help, specially also because you seem to make some claims then you might aswell back them up IMO. If it works great, if it doesnt its still an attempt.
Logged
prj
Hero Member
*****

Karma: +294/-31
Offline Offline

Posts: 3567


« Reply #16 on: August 14, 2019, 10:34:33 AM »

Yes it is limited to 2550, an won´t show more in diagnosis. But its possible to fix load to boost calculation without 5120.
And its possible for all sensor-sizes. But I´ve only tested it on these 3 cars yet.

I actually run 350kpa from Bmw F-Series (cost only 25€) on my own 1,8T.
Fueling is completley fine, even in mountains.

lol, we got a badass in here.

On ME7 pvdk scaling is *10 so is plsol. The parameter physically can not represent a bigger pressure. A single byte variable can go to 255, and that is 2550. There is no "diagnostic limit" in ME7. Same goes for ps_w, at 2550 it's 65535, you're done.
You can make it *20 and rescale stuff so it works correctly.

Or you can fudge the factor/offset of the sensors, to have atmospheric roughly in the same place, but increase the multiplier up top. Then underscale the MAF. Of course there is a LOT of shit that breaks this way, but if you don't go over 15-20% it will work. Except all your logs will show garbage. Same can be applied just fine on MED9. Works with a 3000mbar sensor, any more and you start having other problems.

I think I posted how to do that in like 2013 or 2014 on here.
I'll kindly just remind you, that I was one of the few other people here who made the 5120 mod in the first place, and unlike you I know BGSRM assembly, the fresh air integrator logic, variables and scaling in my sleep. I did that because the "method" you are using is shit and 5120 mod is as factory as you can make it.
« Last Edit: August 14, 2019, 10:40:39 AM by prj » Logged
nbdiy
Full Member
***

Karma: +1/-1
Offline Offline

Posts: 54


« Reply #17 on: August 15, 2019, 02:28:04 AM »

I contradict in not a Single word you say.

You can only use a ps_w up to dez 255, this is right. But this 255 is only a number.

You can change everthing matching to the curve of your new LDF, to your new 255.
And you can also change load to boost calculation accordingly.

You won´t even need to change anything in fueling.
Logged
prj
Hero Member
*****

Karma: +294/-31
Offline Offline

Posts: 3567


« Reply #18 on: August 15, 2019, 04:04:08 AM »

I contradict in not a Single word you say.

You can only use a ps_w up to dez 255, this is right. But this 255 is only a number.

You can change everthing matching to the curve of your new LDF, to your new 255.
And you can also change load to boost calculation accordingly.

You won´t even need to change anything in fueling.

Except:
* Your torque calculations go to shit.
* If you screw with KFURL your throttle control and boost target go to shit.
* If you screw with MAF scaling load goes to shit making ignition all wrong, even if you can scale fueling back up with KFLF and not do anything.

And there are other side effects, such as incorrectly calculated torque reserve, idle behavior and so on if things are sufficiently out.

You are presenting this shit like you're some genius with your "solution" when it's just decalibrating things and bodging it to work. Guess what, this has been done for 20 years.
The correct solution is on this very forum - doubling the range of ps_w and others without decalibrating anything.
« Last Edit: August 15, 2019, 04:06:56 AM by prj » Logged
nyet
Administrator
Hero Member
*****

Karma: +429/-53
Offline Offline

Posts: 9715


WWW
« Reply #19 on: August 15, 2019, 07:58:04 AM »

You can only use a ps_w up to dez 255, this is right. But this 255 is only a number.

You can change everthing matching to the curve of your new LDF, to your new 255.
And you can also change load to boost calculation accordingly.

You won´t even need to change anything in fueling.

You literally just described the 5120 fix.
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
prj
Hero Member
*****

Karma: +294/-31
Offline Offline

Posts: 3567


« Reply #20 on: August 15, 2019, 08:16:12 AM »

You literally just described the 5120 fix.

He did not. 5120 halves the resolution of said variables and the multiplications/divisons take care of not affecting anything else outside of the pressure variables.
His approach is just underscaling things.
Logged
nyet
Administrator
Hero Member
*****

Karma: +429/-53
Offline Offline

Posts: 9715


WWW
« Reply #21 on: August 15, 2019, 08:39:19 AM »

He did not. 5120 halves the resolution of said variables and the multiplications/divisons take care of not affecting anything else outside of the pressure variables.
His approach is just underscaling things.

I'm specifically only referring to his "255 is just a number" observation.
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
nbdiy
Full Member
***

Karma: +1/-1
Offline Offline

Posts: 54


« Reply #22 on: August 15, 2019, 12:19:17 PM »

Except:
* Your torque calculations go to shit.
* If you screw with KFURL your throttle control and boost target go to shit.
* If you screw with MAF scaling load goes to shit making ignition all wrong, even if you can scale fueling back up with KFLF and

- No, I can even leave torque modell stock.
- No, throttle working perfect, it drives like its from factory
- and No, I didn´t underscale anything. But on Ecus with 8bit load axis you can´t do real timing over 191% load (but I think you know this).
- Yes, boost request on one of these cars did a light weird behavior, in lower regions and I didn´t took time to check. Its actually nevertheless running well.
The main important things for me were: driveability, fueling, and boost control, and they do all what they should

Yes KFURL is one side of this, but there is needed more, I first thought this would be impossible too as I was looking for a solution. I´ve seen some of your posts but not all.
I came to this myself about 4 months ago, this was nothing I´ve read here. Until now it works. But I´m of course still looking for side effects.
Logged
nyet
Administrator
Hero Member
*****

Karma: +429/-53
Offline Offline

Posts: 9715


WWW
« Reply #23 on: August 15, 2019, 12:52:40 PM »

I didn´t underscale anything.

By definition if you are avoiding the 255 limit you are underscaling w/o the 5120 hack, and you are doing it via KFURL

16bit load axis is another (fairly) simple fix.
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
prj
Hero Member
*****

Karma: +294/-31
Offline Offline

Posts: 3567


« Reply #24 on: August 16, 2019, 01:25:45 AM »

- No, I can even leave torque modell stock.
- No, throttle working perfect, it drives like its from factory
- and No, I didn´t underscale anything. But on Ecus with 8bit load axis you can´t do real timing over 191% load (but I think you know this).
- Yes, boost request on one of these cars did a light weird behavior, in lower regions and I didn´t took time to check. Its actually nevertheless running well.
The main important things for me were: driveability, fueling, and boost control, and they do all what they should

Yes KFURL is one side of this, but there is needed more, I first thought this would be impossible too as I was looking for a solution. I´ve seen some of your posts but not all.
I came to this myself about 4 months ago, this was nothing I´ve read here. Until now it works. But I´m of course still looking for side effects.

Yeah of course you're scaling KFURL, then boost request is wrong and throttle control is wrong.
No, it does not "drive like factory", you don't have any clue what you are talking about.
Besides load and pressure in ME7 there is also flow, and this is all screwed up now.
Logged
nbdiy
Full Member
***

Karma: +1/-1
Offline Offline

Posts: 54


« Reply #25 on: August 20, 2019, 06:10:43 AM »

For me is underscaling using a bigger maf without adjusting MAF-Curve.

I use ori MAF-Curve. And I don´t need to adjust anything in fueling neither.

And when I say throttle is working perfect, then it does.
Logged
nyet
Administrator
Hero Member
*****

Karma: +429/-53
Offline Offline

Posts: 9715


WWW
« Reply #26 on: August 20, 2019, 08:01:13 AM »

For me is underscaling using a bigger maf without adjusting MAF-Curve.
Doesn't really matter what you think. Underscaling load via KFURL is still underscaling.

Quote
And when I say throttle is working perfect, then it does.

What you "say" is clearly not authoritative.
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
Pages: 1 [2]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.021 seconds with 18 queries. (Pretty URLs adds 0s, 0q)