DIY tuning

Stokes

Sergeant
Jan 26, 2018
267
166
0
Houston, TX
Ride
2011 BMW 135i
I uploaded a log to Datazap (in case that's what you meant, if not I can upload it somewhere else). Any feedback would be appreciated. I'm also going to try to make the contact you sent me, hopefully today.


Oh Geez. I can't remember how the JB4 parameters go, they are so weird to me. Trims start dropping and so does FP_H, so my first though (might be way wrong) is that you are asking for less fuel in the upper end. My experience has been the hpfp gets taxed the most mid range, not at the top end of gear. Maybe someone else more intimate with JB4 logging can help. Also, I've found the settings on the JB4 side make a huge difference as they are inline hijacking the signals to fuel rail and MAF.
 

JPuehl

Specialist
Jul 24, 2019
84
107
0
Greater Seattle area (a.k.a. Zombie Land)
Ride
2011 135i 1M clone N55e DCT
Oh Geez. I can't remember how the JB4 parameters go, they are so weird to me. Trims start dropping and so does FP_H, so my first though (might be way wrong) is that you are asking for less fuel in the upper end. My experience has been the hpfp gets taxed the most mid range, not at the top end of gear. Maybe someone else more intimate with JB4 logging can help. Also, I've found the settings on the JB4 side make a huge difference as they are inline hijacking the signals to fuel rail and MAF.
Houtan mentioned the same about the HPFP NOT getting taxed on the upper end. The Fuel_OL value, which is the only "fuel" related user adjustable value, is supposed to be a bias applied to what the DME is seeing for the actual pressure, the higher the user value, the lower that is "reported" back to the DME in an effort to make the DME raise pressure. A "trims" value of 25 is supposed to mean no added trims help to the HPFP, above 25 is more "help" and below is less, so it seems if my HPFP values drop <11 then the trims should be increasing even if the HPFP can't meet demand, not decreasing as seen in the log. Also my AFRs from about 4400 rpm on are ~12, which might be too low, so it seems demand is high. I could try raising the AFR in areas and see how it runs (just could use some guidance as to what I should be shooting for, besides not causing misfires lol).

Now that I think about it, with HPFP not getting taxed on the upper end yet an obvious downward sloop of trims can be seen in the log, perhaps the trims are causing the pressure to drop <11 and causing the misfire? Right around a trims value of ~26 is where the HPFP value drops below 11, even though timing and AFRs are OK.
 

Stokes

Sergeant
Jan 26, 2018
267
166
0
Houston, TX
Ride
2011 BMW 135i
I missed where you were getting misfires. What fuel are you running? AFRs look fine enough for 93. I found Ethanol has a different target, usually leaner. Some like 12.5.

If I would drop under 12.1 on E30 I'd get that misfire feel or if rail pressure dropped below 1700 psi.

Looks like 6 degree of timing pull on #4 from 5800 on. That's not good. Oh, and earlier around 4k. Might want to verify coil/plug first before chasing a HPFP issue. Man, if that is 93 fuel, that might be too much timing for the 20 psi mid range. I'd play it safe and get correction out of there as a first step.

This is why I didn't like the BEF tunes. They were super aggressive it seemed and everyone with the JB4 were fine with corrections all over the place and "as long as your avg ignition value is low, you can crank up the boost!" I don't follow that belief.
 

JPuehl

Specialist
Jul 24, 2019
84
107
0
Greater Seattle area (a.k.a. Zombie Land)
Ride
2011 135i 1M clone N55e DCT
For my fuel I'm running 92 pump with enough Boostane to make a "calculated" 98 octane. I'd love it if my timing was consistent across all cylinders but 1 in 5 logs has matching timing. This log it will be #4 correcting, next log it will be #2 and #6 then next log... I have my doubts about the accuracy of the JB4 logs. How can 1 cylinder have -4* correction and the cylinder next to it have +8*? That doesn't make sense. And I've made adjustments to the AFRs and timing int he BEF so it's not exactly as it was when I downloaded it from the N54tech site.
 

RealM135i

New Member
Jul 29, 2020
4
0
0
Guys

On the:
Compressor/ turbine characteristic table
Wastegate position model
Wastegate feed forward tables
And every other table under the boost section

Is the MAF axis values the MAF req WGDC logging value or just the normal MAF logging value?
 

houtan

Lieutenant
Nov 2, 2017
611
293
0
Ride
135i N55 DCT; PS2
Guys

On the:
Compressor/ turbine characteristic table
Wastegate position model
Wastegate feed forward tables
And every other table under the boost section

Is the MAF axis values the MAF req WGDC logging value or just the normal MAF logging value?
i believe the value is MAF req WGDC g/s in the MHD log.
 

JPuehl

Specialist
Jul 24, 2019
84
107
0
Greater Seattle area (a.k.a. Zombie Land)
Ride
2011 135i 1M clone N55e DCT
@houtan @Stokes @all4bspinnin

I'm sure you guys will be happy to know that I'm breaking free from the JB4 and am working on self-tuning my car. I've read most if not all of this thread and read Anjuna's N55 tuning thread over on E90post and am slowly working on my tune. I only have the public-domain XDF so I probably don't have all the cool tables that others have, but I'll go as far as I can with what I have. I'm working on using PID control because I live in the Seattle area and temps/atmospheric conditions vary throughout the year and I believe that PID control can help smooth the boost.

Having said that I've started making changes to my BIN and have a few questions that I didn't find an answer to, so any input would be helpful.

1. What table or process is responsible for the boost target? The targets in my log are close but a bit over what I want and if get close to what I want I believe that PID could raise boost to meet the target and I don't want that, so I'm trying to figure out how to lower target.

2. I've read mention of the WG pos Modell table but no specific info about what it's used for or what changes may be needed, if any. Anyone have any ideas?

3. Does the Boost Ceiling table, current value of 2 bar, prevent or try to prevent boost over target? 2 bar is 29 psi and right now I'd like to target 19 psi and was wondering if I changed the value to 1.38 bar, 20 psi, if that would help prevent over boost?

4. On the WGDC Spool table, what is the Y axis? This value doesn't line up with either the "Boost set limit" or "Boost setpoint factor" in the log.

Thanks for any and all help given in my quest to exceed the performance of the JB4!
 
Last edited:
  • Like
Reactions: houtan

Stokes

Sergeant
Jan 26, 2018
267
166
0
Houston, TX
Ride
2011 BMW 135i
@JPuehl so many questions. The public xdf has incorrect calculations. You can use the calcs from anjuna's post. I've been messing with trying to figure out pid for a long time and it's been super frustrating.

My latest aha moment was the P-factor-part table adds a weighting of sorts to the D-factor table (slows P down). I went back to factory P and D factor values and then increased the part table where boost would overshoot. The p-f-part table gives you a lot of fine tuning ability. Most will just cut D by 70%-90% instead of returning for the larger turbo. I found that a 2.5k pull is smooth but a 4k hammer down results in bouncing boost (just needed to reduce dampening weight [P-f-part] at the higher maf /setpoint).

OH, the P-factor-part table is labeled <edit> WGDC (Spool) </edit> in the public table Def. I'm think someone saw it as weighting for when you are spooling up to target. When I saw the diff name in the S55 bin I copied the values over just to see what it did and that's when I started figuring out the weighting. Now that I think of it, arjuna may have called this out, but it didn't make sense at the time to me.
 
Last edited:

JPuehl

Specialist
Jul 24, 2019
84
107
0
Greater Seattle area (a.k.a. Zombie Land)
Ride
2011 135i 1M clone N55e DCT
@JPuehl OH, the P-factor-part table is labeled P-factor-spool in the public table Def. I'm think someone saw it as weighting for when you are spooling up to target. When I saw the diff name in the S55 bin I copied the values over just to see what it did and that's when I started figuring out the weighting. Now that I think of it, arjuna may have called this out, but it didn't make sense at the time to me.

Thanks for the response, it's appreciated! I noticed in my XDF, which is the public domain version, I don't have the table named P-factor part/P-factor spool table defined. I did notice Anjuna mention it but I believe he has an f-series and some of the values in his thread and not even close to mine, so I figured it was a different between E & F series. My car is an early '12 so I use the 98G0B XDF. Thanks again for the information.
 

Stokes

Sergeant
Jan 26, 2018
267
166
0
Houston, TX
Ride
2011 BMW 135i
Okay I created a pull request for the public 98G0B XDF tonight. Been promising people to do this for months.

Added in the following tables for 98G0B
- I-Factor Ceiling
- I-Factor Floor
- PID Adder ceiling
- PID correction floor
- PID correction ceiling
- Injection Angle Start

Moved a couple tables under WGDC around that makes more visual sense.

Now we need Github project Contributors
martialhd
jyamona
Jake Yamona (@jyamona)

to review and accept the updates to that XDF file. I couldn't find Martial to tag him.
 
Last edited:

Stokes

Sergeant
Jan 26, 2018
267
166
0
Houston, TX
Ride
2011 BMW 135i
I'll look and see what vanos tables are useful that I have and add them in the next go round.
 
Last edited:
  • Like
Reactions: JPuehl

derekgates

Lieutenant
Feb 23, 2018
740
375
0
NW FL
derekgates.us
Ride
2011 335is
Okay I created a pull request for the public 98G0B XDF tonight. Been promising people to do this for months.

Added in the following tables for 98B0B
- I-Factor Ceiling
- I-Factor Floor
- PID Adder ceiling
- PID correction floor
- PID correction ceiling
- Injection Angle Start

Moved a couple tables under WGDC around that makes more visual sense.

Now we need Github project Contributors
martialhd
jyamona
Jake Yamona (@jyamona)

to review and accept the updates to that XDF file. I couldn't find Martial to tag him.

Freaking awesome. I saw the PR yesterday. I love seeing contributions to public knowledge. Will be great if the PR is approved :)
 
  • Like
Reactions: JPuehl

Stokes

Sergeant
Jan 26, 2018
267
166
0
Houston, TX
Ride
2011 BMW 135i
Okay, I updated my fork with the following...


Added in the three HPFP tables for higher commanded pump pressure
  • Copied from the N54 forum and found the hex addresses in N55 98G0B
  • Allows for E50 on factory early N55 pump (if you know how).
Fixed the Vanos IN/EX (Warm) tables
  • Base and associated breakpoint tables for IN and EX
  • They both are now pointing to the correct vanos table in the DME bin
  • <<I changed the calculation to align better to what gets logged in MHD>>
  • Feel free to remove the 120 and 115.2 offsets in the calculation if the resulting values hurt your head

I believe you can pull the updated XDF from my fork until it gets merged with the main branch.
 

derekgates

Lieutenant
Feb 23, 2018
740
375
0
NW FL
derekgates.us
Ride
2011 335is
Okay, I updated my fork with the following...


Added in the three HPFP tables for higher commanded pump pressure
  • Copied from the N54 forum and found the hex addresses in N55 98G0B
  • Allows for E50 on factory early N55 pump (if you know how).
Fixed the Vanos IN/EX (Warm) tables
  • Base and associated breakpoint tables for IN and EX
  • They both are now pointing to the correct vanos table in the DME bin
  • <<I changed the calculation to align better to what gets logged in MHD>>
  • Feel free to remove the 120 and 115.2 offsets in the calculation if the resulting values hurt your head

I believe you can pull the updated XDF from my fork until it gets merged with the main branch.

Thank you for the work you are doing on this. It is very much appreciated.
 
  • Like
Reactions: JPuehl and Stokes

Skevo8

Lurker
Oct 1, 2020
13
0
0
Hi - I am using the latest 00001A841D2801.xdf (for S55) from github. I have a backup .bin generated by MHD - and i am looking at some minor adjustment to my stock map.
can anyone help with the understanding of below "codeword" type of parameters?

Codeword load max. active (stock set to 1) - xdf comment recommends to set to 0 . Why?
Codewort für Abgasklappensteuerung - stock set to 58 (found some other tuned map with set to 18) - what is this used for?
Codewort Noise Control my stock set to 0 (other tuned map was set to 78) - what is this used for?

thanks
 
Last edited:

wheela

Lieutenant
Jun 4, 2021
1,105
617
0
Twin Cities, MN
Ride
2015 e84 X1 35i Msport
I know there are a million E85 mixing ratio apps out there, so I added the 1,000,001 app. It basically does what the others do, but in a way I like it to do it, so better. :p. Sharing in case anyone finds it useful.

Thanks for sharing, this is pretty cool! I've been pretty set on sticking to pump gas for convenience, but I have to say this app gives me some cause to reconsider ethanol blends...
 
  • Like
Reactions: Stokes

Stokes

Sergeant
Jan 26, 2018
267
166
0
Houston, TX
Ride
2011 BMW 135i
Hi - I am using the latest 00001A841D2801.xdf (for S55) from github. I have a backup .bin generated by MHD - and i am looking at some minor adjustment to my stock map.
can anyone help with the understanding of below "codeword" type of parameters?

Codeword load max. active (stock set to 1) - xdf comment recommends to set to 0 . Why?
Codewort für Abgasklappensteuerung - stock set to 58 (found some other tuned map with set to 18) - what is this used for?
Codewort Noise Control my stock set to 0 (other tuned map was set to 78) - what is this used for?

thanks
No idea. I can only guess the first one is to turn off the DME checking if you are exceeding load max (main way the DME controls boost... load based). Other two seem to be exhaust tuning related. For N55, we usually set the tables that cap load (by gear) to something really high. I could see having the ability to switch that off, could be useful if you are going after max power for the dyno on drag strip. I prefer to leave the safety controls intact, just a bit looser.