[lanley]

Hi guys

Do you have the XDF's for F-series N55?

[cmcacmca75]

Originally Posted by Anon1

I just shot you links to the DAMOS files that we have.

If anyone else would like them, please PM myself or anjuna.
ECUS:
DME8.4
DME8.6
DME8.x
ME9 N40
MEVD17.2
MEVD17.2.3
MEVD17.2.4
MEVD17.2.6
MEVD17.2.8
MEVD17.2.G
MEVD17.2.S
MEVD17.2.U
MSD80
MSD81
MSD87
MSV90

PM'd you a questioned.

[BMW63]

can someone assist with info on the XDF for N55 EWG, are there 2 boost multiplier tables? if so what is the address of the 2nd one?

[e46Tyrant]

Originally Posted by lanley

Hi guys

Do you have the XDF's for F-series N55?

They are all posted on the public github. There are like 5 rom types between the various f-series N55 models.

[e46Tyrant]

Hey guys, just wanted to keep this conversation alive. I’ve been self tuning the n55 for a few months now and have found many of the same things as Anjuna’s main post (excellent write up!) although I think there are some significant differences amongst the n55 DME’s between e, f, pwg, ewg, etc. because many of the tables are very different both in appearance and functionality.

I have an e-series n55 and a f-series pwg. I’m happy to elaborate on some of the differences I’ve found tuning between the two and do my best to contribute to this thread and answer any questions. The f series pwg is much more difficult to get boost control perfect (overboost post shift, flat boost curve, etc) but I’m pretty happy with where it’s at now. PID took some finessing and I’m happy to share my strategy. The e-series has commanded which makes things much easier, but even the stock approach with PID is easier than the f to get just right.

One issue I’ve run into is there seems to be a fuel related cap that kills load target on the f series. When hpfp tanks so does load, which makes sense. But even if hpfp stays strong (due to an upgraded hpfp) it still causes load request to drop significantly. I know it’s fuel related because if I lean out afr a bit the load request increases and vice-versa. Has anyone else ran into this issue?

Also, for those of you running a 3.5 bar sensor, is it possible to actually target higher boost than 22-23 psi? I check the mhd box or use Anjunas modified settings, but Load just seems to max out and you can’t increase boost target any further. The n55’s don’t have the custom 3 cell scaling logic of the n54 so I’m confused how it works. I have to assume it’s possible because mhd has the “boost target ex” logging parameter which is to log boost target above the stock tmap. Is there a table I’m missing that allows for higher boost targets when using the 3.5 bar sensor?

[JPuehl]

Originally Posted by e46Tyrant

Hey guys, just wanted to keep this conversation alive. I’ve been self tuning the n55 for a few months now and have found many of the same things as Anjuna’s main post (excellent write up!) although I think there are some significant differences amongst the n55 DME’s between e, f, pwg, ewg, etc. because many of the tables are very different both in appearance and functionality.

I have an e-series 135 n55 and a f30 pwg. I’m happy to elaborate on some of the differences I’ve found tuning between the two and do my best to contribute to this thread and answer any questions. The f series pwg is much more difficult to get boost control perfect (overboost post shift, flat boost curve, etc) but I’m pretty happy with where it’s at now. PID took some finessing and I’m happy to share my strategy. The e-series has commanded which makes things much easier, but even the stock approach with PID is easier than the f to get just right.

One issue I’ve run into is there seems to be a fuel related cap that kills load target on the f series. When hpfp tanks so does load, which makes sense. But even if hpfp stays strong (due to an upgraded hpfp) it still causes load request to drop significantly. I know it’s fuel related because if I lean out afr a bit the load request increases and vice-versa. Has anyone else ran into this issue?

Also, for those of you running a 3.5 bar sensor, is it possible to actually target higher boost than 22-23 psi? I check the mhd box or use Anjunas modified settings, but Load just seems to max out and you can’t increase boost target any further. The n55’s don’t have the custom 3 cell scaling logic of the n54 so I’m confused how it works. I have to assume it’s possible because mhd has the “boost target ex” logging parameter which is to log boost target above the stock tmap. Is there a table I’m missing that allows for higher boost targets when using the 3.5 bar sensor?

You're post is perfect timing as 3 weekends ago I disconnected my JB4 and started self tuning and am currently at the point of working on getting boost close to target then will work on PID. I have an early '12 135i N55. I've 0'd the PID table and am tuning the WGDC before PID table to get boost where it should be, as well as a couple other table like pressure ratio tables and the load ceiling table. I'd be interested in hearing the strategy you used to get you where you are today.

[e46Tyrant]

Originally Posted by JPuehl

You're post is perfect timing as 3 weekends ago I disconnected my JB4 and started self tuning and am currently at the point of working on getting boost close to target then will work on PID. I have an early '12 135i N55. I've 0'd the PID table and am tuning the WGDC before PID table to get boost where it should be, as well as a couple other table like pressure ratio tables and the load ceiling table. I'd be interested in hearing the strategy you used to get you where you are today.

I didn't find any benefit to zero'ing out PIDs to dial in boost.
Assuming it's a stock turbo car, easiest way to get things sorted is to set load ceiling high and use wgdc to control boost target. After about 4500 rpm you really can't go wrong since the stock turbo is so small, but be careful with duty cycle at lower rpm because you can exceed the stock tmap fairly easily and it's hard on the engine to make that much cylinder pressure and low end torque.

Are you using commanded or standard wgdc tables?

[JPuehl]

Originally Posted by e46Tyrant

I didn't find any benefit to zero'ing out PIDs to dial in boost.
Assuming it's a stock turbo car, easiest way to get things sorted is to set load ceiling high and use wgdc to control boost target. After about 4500 rpm you really can't go wrong since the stock turbo is so small, but be careful with duty cycle at lower rpm because you can exceed the stock tmap fairly easily and it's hard on the engine to make that much cylinder pressure and low end torque.

Are you using commanded or standard wgdc tables?

I have a PS2 so not using OEM anything. Commanded wgdc table doesn't work when using PID and I don't believe PID works with commanded wgdc, so I'm focusing on the standard wgdc. I'm in the "early" stages of tuning and zeroing the PID tables is helping get the wgdc table, and other tables on par, then will reset PID tables back to OEM and go from there. PID is just supposed to correct errors and I found that while trying to dial in boost, PID was giving me a false sense of accomplishment (lol) as I was hitting target but when I zeroed PID I realized just how much PID was "working" to bring wgdc up to meet target.

Are you stock turbo and how did you dial in boost, "set load ceiling high and use wgdc to control boost target"? How much did you need to adjust the PID tables?

[e46Tyrant]

Originally Posted by JPuehl

I have a PS2 so not using OEM anything. Commanded wgdc table doesn't work when using PID and I don't believe PID works with commanded wgdc, so I'm focusing on the standard wgdc. I'm in the "early" stages of tuning and zeroing the PID tables is helping get the wgdc table, and other tables on par, then will reset PID tables back to OEM and go from there. PID is just supposed to correct errors and I found that while trying to dial in boost, PID was giving me a false sense of accomplishment (lol) as I was hitting target but when I zeroed PID I realized just how much PID was "working" to bring wgdc up to meet target.

Are you stock turbo and how did you dial in boost, "set load ceiling high and use wgdc to control boost target"? How much did you need to adjust the PID tables?

PS2 is a bit different. Setting load arbitrarily high isn't the best strategy for upgraded turbos IMO. PID needed some finessing. My success with the PS2 PID tuning came from restricting PID activity around boost target but still allowing for lots of PID as boost deviates from target. This allows for smoother/flatter boost curves but still quick corrections to overshoots and sharp transitions out of spool.

If you post a log I can see where you are at and provide more specific advice.

[e46Tyrant]

Has anybody else noticed the relationship to rail pressure and load request? It's annoying as hell that a dip in rail pressure seems to result in load request following suit. I understand the logic behind why, but I'd rather tune around it and just know that my rail pressure is at its limit.

Is there a table that allows for load request to stay flat despite a dip in rail pressure?

[houtan]

Originally Posted by e46Tyrant

PS2 is a bit different. Setting load arbitrarily high isn't the best strategy for upgraded turbos IMO. PID needed some finessing. My success with the PS2 PID tuning came from restricting PID activity around boost target but still allowing for lots of PID as boost deviates from target. This allows for smoother/flatter boost curves but still quick corrections to overshoots and sharp transitions out of spool.

If you post a log I can see where you are at and provide more specific advice.

Any logs you can share on PS2 PID based tunes?

[ntomlin335i]

Hello, I recently purchased a used jb4 and was wondering if the connector is supposed to be missing prongs 3,8,9,16,19,20. i attached a photo for reference.

[anjuna]

Originally Posted by ntomlin335i

Hello, I recently purchased a used jb4 and was wondering if the connector is supposed to be missing prongs 3,8,9,16,19,20. i attached a photo for reference.

This is NOT a JB4 thread.

[ntomlin335i]

Originally Posted by anjuna

This is NOT a JB4 thread.

sorry

[335i533]

Anjuna-

I just picked up a 13 F30 with N55 manual and would love to start tweaking my own bin, its been a while since I did some tuning!

[Tunafish]

Is there an updated link to the 98G0B XDF file .. The link on first post doesn't work.

[Q_illest]

Originally Posted by anjuna

I am F series as well. I would be glad to share my BIN and XDF (privately for now) with you.

For anyone reading, the privacy is NOT related to "trying to keep secrets" but is more so to elimination of releasing information to the public that is NOT 100% accurate.

The LAST thing I want to do is lead someone down the wrong path. All the information released in the first few posts of this thread in their subcategory are 95% confident. My personal XDF and BIN are always changing day-by-day in a lot of the non-released portions of what this post discusses.

Hey so i took tuning classes and im having a hard time getting a bin file and understanding tunerpro could you send it to me and help me out?

[Q_illest]

Hey im looking for the Bin file for tunerpro for the f10 535i, im new to the fourms so i couldnt figure out how to pm you but if you use instgram dming would be a lot easier @projectmayhem__, im sure you already know that by me asking u for the bin file you know that im trying to tune the car myself and i have taken tuning classes and all

[buhala21]

Originally Posted by anjuna

I am F series as well. I would be glad to share my BIN and XDF (privately for now) with you.

For anyone reading, the privacy is NOT related to "trying to keep secrets" but is more so to elimination of releasing information to the public that is NOT 100% accurate.

The LAST thing I want to do is lead someone down the wrong path. All the information released in the first few posts of this thread in their subcategory are 95% confident. My personal XDF and BIN are always changing day-by-day in a lot of the non-released portions of what this post discusses.

A few weeks ago i started to tune my car, i have gone a long way since i started. I REALLY LIKE YOUR THREAD, it was my starting point. I am looking for a better xdf file. My car is a 2011 bmw 135i DKG. stefanpetrov01@gmail.com

[wheela]

Big thumbs up on this thread. Reminds me of an online version of Merlin's tuning guide for EVO 7, 8, and 9. I only dabbled on that platform, and am new to n55, so I don't have much to contribute, but thanks so much for starting this thread, I'm excited to see if this keeps going.

[wheela]

Okay guys, here's (hopefully) a basic question - can you please help me better understand how load(s) is calculated in these DME's? I've searched google, but haven't found any equations, or specific descriptions.

My current understanding of load is based on reviewing fueling and ignition tables on Mitsubishi ECU's (specifically 3g eclipses and EVO 7, 8, and 9). Since those tables are based on % load and rpm for the axes, I've assumed that load is just airflow expressed as a percentage of maximum airflow that the motor will consume at that rpm based on natural aspiration. So my naturally aspirated 3G had load up to 100% and the EVO's had load up to 170, or 220, or whatever (I can't remember), because the boost from the turbo could push airflow past the 100% point. So then for given RPM and %load (airflow) the ECU used the tables to determine what fueling and timing to use for that combination of %load (airflow) and rpm.

In the boost section of this thread, it's mentioned that some of the % load is also supplied by fueling, vanos, and ignition timing. What I'm not understanding, is how can fueling and ignition timing contribute to load, when fueling and ignition values are looked up from a table based on load (and rpm)? Is there some different kind of load? I can see how these would contribute to the torque value the DME calculates, but I'm not understanding how this fits my current understanding of what load is? Perhaps I'm not properly understanding what %load is in the first place?

I get that our DME's target a % load based on pedal inputs, and then make adjustments to throttle plate, vanos, valvetronic, etc. to try and hit the load target, and then continually adjust based on the actual load being achieved. But in MHD, it looks like there are at least two different loads that are measured against the load request %: "Load act. (%)" and "Load actual RAM". What are these two different loads?

Am I at least partially on the right track, or completely lost lol?

[anjuna]

I'll do my best. I have some experience with the EVO style but more with CarBerry/Subies.

Load is calculated by a metric fuck ton of variables, but it is mostly a "torque to load" conversion. This means the engine calculates the amount of torque it is producing (or, in overrun, receiving) by using a ton of variables (I know you are asking for all of these, but there are really only a few worth mentioning, because some of them get as complex as "turbo backpressure" and airflow through various manifolds - basically models). Most of them make sense and there is no need to change them. Those include friction loss, mechanical components, etc. The ones you can change are the ignition timing, VANOS, Valvetronic, airflow, air temp, barometric, blah blah blah.

Long to the short - It determines the torque it must output (or it is receiving) and generates an arbitrary value called "load" and you can see that here, in this stock table.



However, some people alter these tables, because they may find that by changing VANOS, valvetronic, boost, etc, they will run into an instance where the DME load req and DME calculated load are wayyyyy off. The hard part to understand is this is kind of a bi-directional table. The DME is taking alllll the commanded values for alllll those variables I had previously mentioned and reverse looking up through this table. Then, it also takes allllllll those ACTUAL variables and uses this table as well. Lastly, but also probably most importantly, most people tend to adjust the last few rows in this table to prohibit "torque limiting" on the transmission. I do not suggest this.





The reason that "load" is used for these tables is due to how the DME handles powaaaaaaa.

The accelerator generates a "position" which is translated into a desired load value. This is what is controlled in the "Load Ceiling" table. If you're 100% on throttle, you get the rightmost value in that column. If you have pre-ignition, you move to the left in that table.

As such, that's sort of why you want a load value. If you're telling the engine "I'd like this much 'load'" then the DME says "OK lets try to do that. These are the values we should use. Let's see how it goes"

You could essentially replace "Load" with "Throttle Percentage" as long as you scale that to your max load - For instance, my max load is 200%. This makes it easy, since I know that basically 100% load is half throttle. It's SORT of that easy, but there's some other fuckery in there (valvetronic and valve body). But to keep it simple, that's probably easier.

I think the second part of your comment is answered in the first part of this one. There are "two" loads, but they are only requested vs actual.

This is important to understand - if you have overboost (or any experience of too much air) then you will see a higher ACTUAL load. Requested load will stay where your foot puts it.

Originally Posted by wheela

Okay guys, here's (hopefully) a basic question - can you please help me better understand how load(s) is calculated in these DME's? I've searched google, but haven't found any equations, or specific descriptions.

My current understanding of load is based on reviewing fueling and ignition tables on Mitsubishi ECU's (specifically 3g eclipses and EVO 7, 8, and 9). Since those tables are based on % load and rpm for the axes, I've assumed that load is just airflow expressed as a percentage of maximum airflow that the motor will consume at that rpm based on natural aspiration. So my naturally aspirated 3G had load up to 100% and the EVO's had load up to 170, or 220, or whatever (I can't remember), because the boost from the turbo could push airflow past the 100% point. So then for given RPM and %load (airflow) the ECU used the tables to determine what fueling and timing to use for that combination of %load (airflow) and rpm.

In the boost section of this thread, it's mentioned that some of the % load is also supplied by fueling, vanos, and ignition timing. What I'm not understanding, is how can fueling and ignition timing contribute to load, when fueling and ignition values are looked up from a table based on load (and rpm)? Is there some different kind of load? I can see how these would contribute to the torque value the DME calculates, but I'm not understanding how this fits my current understanding of what load is? Perhaps I'm not properly understanding what %load is in the first place?

I get that our DME's target a % load based on pedal inputs, and then make adjustments to throttle plate, vanos, valvetronic, etc. to try and hit the load target, and then continually adjust based on the actual load being achieved. But in MHD, it looks like there are at least two different loads that are measured against the load request %: "Load act. (%)" and "Load actual RAM". What are these two different loads?

Am I at least partially on the right track, or completely lost lol?