BMW M5 Forum and M6 Forums banner
581 - 600 of 606 Posts
Do any of you know if the MSS52 DME is capable of doing RPM rev matching? I know that the MSS54 does have this functionality with SMG cars only. However, I think it may be limited to the SMG cars because the regular non-SMG manual transmissions don't have a transmission speed sensor. However, there is a casting for where this sensor should be in both E39M5 and E46M3 manual transmissions, just a matter of having a machine shop drill it out. As such, it is possible to add a transmission speed sensor to E39M5. My question is once that is done, can it be wired to the DME and then the DME programmed to do rev matching.
 
I had some time to update the MSS5x Binary Modification Tool, I added a few new features for you M5 guys.
Image


  • Burble and pop is now available for the M5.
  • AlphaN is now available for the M5.
  • Idle speed adjustment is available for both the M3 and M5.
  • Cat protection threshold can now be adjusted on the M5.
  • Disable EGT monitoring is now available for the M5.

Grab the latest version here: www.ecuworx.co.uk/downloads/
 
Do any of you know if the MSS52 DME is capable of doing RPM rev matching? I know that the MSS54 does have this functionality with SMG cars only. However, I think it may be limited to the SMG cars because the regular non-SMG manual transmissions don't have a transmission speed sensor. However, there is a casting for where this sensor should be in both E39M5 and E46M3 manual transmissions, just a matter of having a machine shop drill it out. As such, it is possible to add a transmission speed sensor to E39M5. My question is once that is done, can it be wired to the DME and then the DME programmed to do rev matching.
Whether or not the MSS52 specifically is capable of doing it, it's software almost certainly isn't, and even if it was, (from what I understand) it's much more complex than the DME reading a gear position sensor. (from what I understand) the SMG module actually commands the DME's torque output continuously over CANBUS to achieve the 'rev-matching' you're describing.
 
Do any of you know if the MSS52 DME is capable of doing RPM rev matching? I know that the MSS54 does have this functionality with SMG cars only. However, I think it may be limited to the SMG cars because the regular non-SMG manual transmissions don't have a transmission speed sensor. However, there is a casting for where this sensor should be in both E39M5 and E46M3 manual transmissions, just a matter of having a machine shop drill it out. As such, it is possible to add a transmission speed sensor to E39M5. My question is once that is done, can it be wired to the DME and then the DME programmed to do rev matching.
Whether or not the MSS52 specifically is capable of doing it, it's software almost certainly isn't, and even if it was, (from what I understand) it's much more complex than the DME reading a gear position sensor. (from what I understand) the SMG module actually commands the DME's torque output continuously over CANBUS to achieve the 'rev-matching' you're describing.
I think most of the code is onboard the DME itself since the rev matching doesn't really happen without tweaking the DME side of things... but it'd still be rather complex to get it to work on a manual car, and doubly so to get it to work on an MSS52.

If it's at all feasible, the path I would take is get it working right on a manual E46 M3 first, then flash the MSS52 with MSS54 software and modify all the appropriate maps to run the S62 properly.
 
Do any of you know if the MSS52 DME is capable of doing RPM rev matching?
Wait, why would you want to? This works in an SMG car because the computer also controls the clutch; even if you DID get the DME to open the throttle plates to bring the engine speed up, you would STILL need to have shifted into neutral AND re-engaged the clutch in order for to transmit the the rotational power to the transmission gears. And by the time you've accomplished that, you might as well be blipping the throttle with your right foot and just do a proper heel/toe downshift! Since you still have a manually actuated clutch and gearbox, this wouldn't really gain you any laziness. . .
 
If it's at all feasible, the path I would take is get it working right on a manual E46 M3 first, then flash the MSS52 with MSS54 software and modify all the appropriate maps to run the S62 properly.
As an aside - that's something I've long contemplated... basically, would it be a shorter path to full tunability to put in the time and effort to either run a MSS54(HP) DME in my M5/a MSS54-flashed MSS52 (I think that was a thing for a minute in the early-ish CSL clone days due to the extra H-bridge the MSS52 has, or something like that) ... or continue slowly plugging along manually creating an XDF for the MSS52 (which is what I've currently been doing).
 
I've had some trouble running MSS54 code on very early MSS52s (they have a slightly different processor and presumably the 54 code trips up some errata that was fixed in the later revisions), but otherwise non-HP software on an MSS52 would work fine. As far as I can tell, all the V8-specific and E39-specific code is present. HP non-CSL software would probably also work more or less okay if you change the EEPROMs.

Real HP would require some hardware changes (basically adding the missing components and changing the secondary MAF input's resistors to match the 52 / non-HP). CSL conversion may also be possible, though the DME might not actually be fast enough to run the MAP sensor code on an 8-cylinder engine well.

That said I don't think it's a shorter path to "full tunability" since really you'd need to be able to reference the MSS52 values to carry them over to the MSS54.

I have some XDFs, though they're far from complete.
 
Hey here's a question, can the tachometer warm-up lights be reprogrammed to work as shift lights? Just illuminate them once that threshold RPM has been reached? Not terribly useful I guess, but would be kinda fun!
You'd have to write your own code, but it's not impossible.
Keeping in mind that that if you're running it attached to a real MSS52 + IKE (cluster) CAN network you'll need your device to sit in-between the cluster and the bus, since the DME would also be sending the same ARBID message to the cluster.

That being said, there have indeed been some instances of people installing MCUs/SBCs running code to get the oil temperature and generate the proper message and send it to the bus, for purposes of having functional warm-up lights in cars with DMEs that to not natively produce the message.

Also, the IKE lights kind of react slower than you might be imagining, sadly... and there's not as many addressable LEDs as the M3 cluster.

IMO - (or really, what I'd like to do someday) it'd be much cooler (and maybe easier) to either command the LEDs on the warmup-light daughterboard directly from a Teensy (built-in CAN!) or the like, or even add your own new LEDs to the cluster and command those, and disable/remove the existing ones. LED technology has come a long way in the last 20 years, after all.....
 
I've had some trouble running MSS54 code on very early MSS52s (they have a slightly different processor and presumably the 54 code trips up some errata that was fixed in the later revisions), but otherwise non-HP software on an MSS52 would work fine. As far as I can tell, all the V8-specific and E39-specific code is present. HP non-CSL software would probably also work more or less okay if you change the EEPROMs.

Real HP would require some hardware changes (basically adding the missing components and changing the secondary MAF input's resistors to match the 52 / non-HP). CSL conversion may also be possible, though the DME might not actually be fast enough to run the MAP sensor code on an 8-cylinder engine well.

That said I don't think it's a shorter path to "full tunability" since really you'd need to be able to reference the MSS52 values to carry them over to the MSS54.

I have some XDFs, though they're far from complete.

Carrying over the values. Totally forgot. You're right.

Shortest path would then be... well, a standalone (eek), swapping MSS54HP + S54 (and hey, a turbo while I'm at it, since this is hypothetical anyway) .. or just buy an E46 M3... or lease an Accord (wouldn't that be the dream)

I've dug up a number of the more .. eclectic scalars/tables during my conquest for native wideband support, and also in the pursuit of finishing BMWs work on the M5 tune. IMO the stock M5 tune leaves quite a bit of performance and drivability on the table, and just generally looks and feels like it was rushed and left unfinished, and then just patched when needed for emissions or something. I don't think it ever got the care and attention the MSS54/HP platform did/does (internally or otherwise). That coupled with my disdain for the native torque management, throttle pedal input handling, and a few of my car's modifications (14lb flywheel, electric fan swap, IAT sensor relocation, etc) there's some niche drivability things that I've tweaked that have made a huge quality of life improvement, in addition to (perceived) performance/fuel economy/etc

Also, since I'm an M5 owner, of course I pine after the more hacker-friendly community the E46 seems to enjoy. M5 owners seem to mostly be old guys who like to keep 'er stock and just wax it up nice and take the lady out for a weekend drive. (I kid, I kid... )
 
Huge thanks to everyone that contributed to this thread. I went from zero to reproducing all the graphs in this thread in no time at all thanks to all the details posted.


I have a few questions for those who have a good handle on this stuff.

1. At offset 0x7FB8, my program version is listed as 32313133323130303136303144343039 in triplicate, which means my rev is '3231' according to everything I've read. Has anyone else seen this rev number before? I haven't seen it mentioned. FWIW the offsets I've checked have the same data as 1301, 1501, 1601 and 1701.

2. Does anyone know the location(s) of the injector scalar(s)?

3. I plan to install a Dinan 3.45 Diff. Do I have to change the ratio @ 0x1BA8? I would think it would have to change to CF for 3.45*60 but since the original number is for 3.23 (C2) instead of 3.15, what would I change it to?

4. Is there any advantage to running the euro vanos maps?


5. P0lar posted a full throttle fuel injection table at 0xAFE but in my partial that offset only produces an 18x3 map. Is the WOT fuel graph somewhere else or is this just the raw y and Z axis?

Image
 
Huge thanks to everyone that contributed to this thread. I went from zero to reproducing all the graphs in this thread in no time at all thanks to all the details posted.


I have a few questions for those who have a good handle on this stuff.

1. At offset 0x7FB8, my program version is listed as 32313133323130303136303144343039 in triplicate, which means my rev is '3231' according to everything I've read. Has anyone else seen this rev number before? I haven't seen it mentioned. FWIW the offsets I've checked have the same data as 1301, 1501, 1601 and 1701.

2. Does anyone know the location(s) of the injector scalar(s)?

3. I plan to install a Dinan 3.45 Diff. Do I have to change the ratio @ 0x1BA8? I would think it would have to change to CF for 3.45*60 but since the original number is for 3.23 (C2) instead of 3.15, what would I change it to?

4. Is there any advantage to running the euro vanos maps?


5. P0lar posted a full throttle fuel injection table at 0xAFE but in my partial that offset only produces an 18x3 map. Is the WOT fuel graph somewhere else or is this just the raw y and Z axis?

Image
I’m not an expert by any stretch - just an interested amateur, but as no one else has replied yet, I’ll answer as best I can.

1. The hex string converts to 211321001601D409 so you’re actually on version 1601 calibration D409 (US E39 M5)

2. I can’t help you with this specifically, but I can point you to the very good XDF that member kmalinich has kindly saved on Github.
Also on github, Saildot4k has listed even more comprehensive XDF’s for MSS54’s and a lot of their locations cross over.

3. It has been discussed that the diff ratio doesn’t effect anything (except in an E46 SMG) so the 3.23 ratio was left in from preproduction.
Therefore if you were going to change it you would change it to 3.45 (CF)

4. I’m not a tuner, so I will apologise in advance if this is incorrect, but as Vanos effects VE you would need to change fuel and ignition as well.
I would suggest it may be worth just flashing a complete Ver.1601 Euro partial if you want to feel the difference in the tunes.

5. In kmalinich’s XDF it’s a 3x18 table at 0x0B2A which seems correct to me.

HTH
 
Rosie,

Thank you! That is a huge help.

1. I tooled around with basic conversions but didn't decode the string. That makes sense.
2. I found this searching his name and wow.. That is well beyond what I thought was available. It looks like what I would be after is fuel injection factor, steady state but then I would have to find something for idle. On other platforms there's just a simple k factor/constant that every single fuel table references. Maybe this platform doesn't have that.
4. I would be modifying fueling anyway but I think a good indicator that it would need to be done specifically for the VANOS table swap would be if the injection factor tables were drastically different.
5. In kmalinich’s XDF, the Fuel, injection, factor, full load is at 0xAFE, which matches up. It looks like the short axis is relative filling.

I've got quite a bit more reading to do, but that's a good thing.
 
Be careful with that thing. Remember I'm just some dude with an M5.
Oh, I hear ya. I'm not going to go in there and shotgun a bunch of changes and rev it to the moon. Since my trims are good, my only near term goals are to dial in the current WOT fueling as required (wiring up a WB soon) which will establish a baseline pre-headers and then dial in the fueling after headers.

Either way I appreciate the work that went into that. Thanks.
 
i know this is old thread but maybe someone still answers, so im building a maxxecu s62 turbo m5, and i want to know what are the voltages/amperage/operation overall for the throttle body actuator? i see that it uses 2.5mm wires on both +-, and according to newtis it runs on 2000hz sample rate?, im new to this and would love to get some ideas on what extra controllers might be needed if its too high voltage or too high amperage for the ecu to run DBW/ethrottle*
 
i know this is old thread but maybe someone still answers, so im building a maxxecu s62 turbo m5, and i want to know what are the voltages/amperage/operation overall for the throttle body actuator? i see that it uses 2.5mm wires on both +-, and according to newtis it runs on 2000hz sample rate?, im new to this and would love to get some ideas on what extra controllers might be needed if its too high voltage or too high amperage for the ecu to run DBW/ethrottle*
Can't remember the amperage, but I run mine with two half-bridges (one to open, one to close) from my Motec M190 using 16 gauge M27759/43 wire and control it with a PID algorithm.

--Peter
 
581 - 600 of 606 Posts