[stewey]

Would it actually change the throttle sensitivity and torque mapping on your car? The exhaust idea sounds cool!

[Terraphantm]

On my car (which isn't an E90 or even an MSV70... though as Chris and I have discovered, the two DMEs are similar in many ways), it definitely makes the throttle more sensitive, and to me it feels like it allows the torque to build up more quickly.

[hassmaschine]

Originally Posted by stewey

Would it actually change the throttle sensitivity and torque mapping on your car? The exhaust idea sounds cool!

of course it would. Getting it to work at all is the key.

[rjahl]

Originally Posted by hassmaschine

of course it would. Getting it to work at all is the key.

Would it help to load and run my Z4 files again and look at the memory dumps?

Iv' been busy this week relearning to program. I've got a VBA project that is rebuilding a 0DAs file from a Bin. It's not complete as I need to add the checksum routines and add a more complex Block offset correction system so I can play with the OPA files. They are just too hard to do manually.

I'v found a few CRC32 checksum routines on line that look like they can be used to correct the Bin Checksums. The Intel Hex line checksum should be easy and I have yet to look at the end of file check sum. It's four bytes but I don't know the logic yet.

I's so out of date with this stuff, last time I did any programming you only had a few variable types to choose from, now it's crazy and translating between Unicode, hex and decimal is a real pain.

[Terraphantm]

Originally Posted by hassmaschine

Deleting the DTC error doesn't work for sure and I think I know why. This is really more for my own thoughts (I'll forget it if I don't write it down, lol).

I have some ideas of what is going on. There's a whole check process for the sport switch monitoring LV's. I labeled them Sport_Switch_MC_MON_1 and Sport_Switch_MC_MON_2. They trigger an error with the MC (master controller) which shuts down the injection/ignition. There's also a whole long MC process I've labeled Master_Controller_State and another one I've called Monitoring_Unit. I did see a reference to "reset state" somewhere, which I believe is what actually triggers 2D54 (once again, I don't trust German translations of technical language - 2D54 has nothing to do at all with engine RPM or speed as some translatiosn suggest).

So the reason deleting the error code doesn't work is because the MC error has already triggered and shut down the ignition/injection, the error code isn't triggered until after this is complete.

I can't really force it from shutting down the MC (because there are many other legitimate reasons for doing so), but I can probably make it pass Sport_Switch_MC_MON_1 and Sport_Switch_MC_MON_2 without triggering the mc_state errors.

I think the most likely reason I'm seeing this of course is because the E90 never got the sport switch as an option. On a Z4 it would probably be as simple as hooking it up and turning on the lv's. What is probably doing it is the vehicle type being E90, and somewhere it's checking that and causing it to shut down.

I'm going to pull some ram dumps later and see what mc_state, err_cod_mu_and err_cod_mc are. I think that will tell me what the problem is and how to fix it..

So I've been going through those routines - definitely a lot more to it than just those high/low checksums. I'm not convinced your issue has anything to do with vehicle type however. I think it's (one of) the program sections themselves that might be tripping the error. I think I got lucky on the E46 since I didn't have to modify a whole lot.

[stewey]

Originally Posted by hassmaschine

of course it would. Getting it to work at all is the key.

Ah ok, I figured you would have already optimized those with your tune. I feel like I'm missing something.

[Terraphantm]

Originally Posted by BMW..

If it's anything like my old E46 M3 the change is very noticeable. Many times have I kept tapping on and off the sport button while staying steady on the throttle. I'd always end up hitting my headrest because of how much it alters the pedal feel haha

It's kinda hard to compare to the E46 M3 since the E46 M3 had excellent throttle response even without sport mode, and then sport mode was just insanity.

On my 330i (E46), I would say it's still less responsive than the M3 (in non-sport), but still much better than stock. I swear the actual delay is reduced, which makes driving much more tolerable. N52 might feel better since valvetronic + quick response should get you close to that ITB feeling. On these DMEs, BMW also programmed it so that you can only enter sport mode when you're not giving any throttle (but you can exit whenever).

[hassmaschine]

I'm not saying it's the vehicle type specifically - I'm just saying, there's another config or setting that must be causing the error to trip. I don't think it's a checksum. I searched through all the parameters and found every checksum related item, and I'm certain we have the checksums identified.

I made a patch to test but I left my flash drive at work. I did get some RAM shots - the mc_state errors are definitely tripped. I think my patch will work.

Need to find the k-line pin for MSV70 - the E90 obviously doesn't use it, but mainly I want to be able to read RAM dumps out on my workbench. I tested PAsoft with MS45.1 and it looks like it works perfectly. taking a screenshot at a time is a huge waste of time, lol.

[Terraphantm]

Originally Posted by hassmaschine

I'm not saying it's the vehicle type specifically - I'm just saying, there's another config or setting that must be causing the error to trip. I don't think it's a checksum. I searched through all the parameters and found every checksum related item, and I'm certain we have the checksums identified.

I made a patch to test but I left my flash drive at work. I did get some RAM shots - the mc_state errors are definitely tripped. I think my patch will work.

Need to find the k-line pin for MSV70 - the E90 obviously doesn't use it, but mainly I want to be able to read RAM dumps out on my workbench. I tested PAsoft with MS45.1 and it looks like it works perfectly. taking a screenshot at a time is a huge waste of time, lol.

Just checked my MS45; the mc_state error was tripped in my failboot dump as well (set to 0x3B fwiw; normal state is 0xAD apparently).

K-line pin for MSV70 is X60002 Pin 2.

[hassmaschine]

Ram dumps are helpful. I can see that "err_cod_mc" is set to 09 - which is "ROM_LV2_ERR". There is exactly one place in the code where it gets set to 9, and it's in the same routine as the checksum test.

I've called that routine "Fail_Checksum". It sets the checksum status to 0 and sets "lv_dr_off_mu_mon" to 1. This then causes it to trigger the RAM_LVL2_ERR - although I'm not sure exactly why this is happening. Following up the chain, lv_rom_crc_ready is likely set to 0. But I have no CRC errors.

I'll just have to keep digging.

[Terraphantm]

Originally Posted by hassmaschine

Ram dumps are helpful. I can see that "err_cod_mc" is set to 09 - which is "ROM_LV2_ERR". There is exactly one place in the code where it gets set to 9, and it's in the same routine as the checksum test.

I've called that routine "Fail_Checksum". It sets the checksum status to 0 and sets "lv_dr_off_mu_mon" to 1. This then causes it to trigger the RAM_LVL2_ERR - although I'm not sure exactly why this is happening. Following up the chain, lv_rom_crc_ready is likely set to 0. But I have no CRC errors.

I'll just have to keep digging.

Well it has to be one of the changes that tripped the error, right? Did you implement the ones I mentioned in our email exchange (reproduced below)?

So to sum up:
Set lc_var_sof_swi and lc_var_sof_swi_mon to 1
Set lc_conf_city_ena and lc_conf_city_ena_mon to 1
Set c_v_sof_swi_on_tol to 0x200
Set c_v_sof_swi_on_bol to 0xf6
Set c_v_sof_swi_max to 0x333
Set c_v_sof_swi_min to 0x7b

You'll also need to set the BN2000 branches in:
sub_47a4c4
sub_47a6ac
the one you labeled "Sport_switch_monitoring"

Anything else?

It might be worth turning off the "city" settings. I don't know if that's breaking anything, but the E60 doesn't have them enabled.

But otherwise since you have those mo3 checksums corrected, none of the c/lc settings should be triggering the state. So I'm wondering if the BN branch modifications are somehow tripping it.

I guess the other possibility is you're missing a switch.

[hassmaschine]

yeah I have all of that set. It's the same settings as the Z4 actually.

I could try turning off city mode I guess. Looking through the MC and MU code, it's not immediately obvious why it's triggering the error, but I can at least see where it's being triggered.

I thought about trying to flash with out the BN2000 patches and the sport switch settings, then another with the BN2000 patches and without the sport switch settings. Maybe it's one, the other, or both.

[hassmaschine]

Just confirmed PAsoft can read out msv70 via k-line. That makes RAM dumps easy!

[Terraphantm]

Originally Posted by hassmaschine

Just confirmed PAsoft can read out msv70 via k-line. That makes RAM dumps easy!

Awesome. I was wondering whether or not that would be enabled in BN2000 mode

[dstrickland]

Shout out to Hassmachine & TRM Tuning! After tons of headaches trying to load the 330i software on my 325i, he along with a local tuner in Atlanta(TRM Tuning) I've got it successfully running with no error codes! Success!

[rjahl]

Originally Posted by dstrickland

Shout out to Hassmachine & TRM Tuning! After tons of headaches trying to load the 330i software on my 325i, he along with a local tuner in Atlanta(TRM Tuning) I've got it successfully running with no error codes! Success!

Congrats!

How does it drive?

[dstrickland]

Originally Posted by rjahl

Congrats!

How does it drive?

I've only been able to put 30 miles or so on it, only 5 of which were highway so I could open it up, needless to say, I'm smiling ear to ear. The manifold made a small difference but the software really woke it up, once the adaptations set in I'm sure I'll be even more impressed! Best decision I've made!

I'm getting ready to flash 330i EGS software over to complete the conversion!

[rjahl]

Originally Posted by dstrickland

I've only been able to put 30 miles or so on it, only 5 of which were highway so I could open it up, needless to say, I'm smiling ear to ear. The manifold made a small difference but the software really woke it up, once the adaptations set in I'm sure I'll be even more impressed! Best decision I've made!

I'm getting ready to flash 330i EGS software over to complete the conversion!

I'm interested in the EGS flash. I've looked into it but have not gone too far. Keep in mind the 330 and 325 do not share the same torque converter. I have it been able to find the difference between them. I read somewhere that the stall speed was higher in the 330 version.

Are you able to do you own tuning now or do you need the tuning shop to make any more changes to your tune?

[hassmaschine]

Originally Posted by dstrickland

Shout out to Hassmachine & TRM Tuning! After tons of headaches trying to load the 330i software on my 325i, he along with a local tuner in Atlanta(TRM Tuning) I've got it successfully running with no error codes! Success!

Hey, I knew Matt at TRM from way back. Don't know if he still works there though.

glad it's running well!

[dstrickland]

Originally Posted by rjahl

I'm interested in the EGS flash. I've looked into it but have not gone too far. Keep in mind the 330 and 325 do not share the same torque converter. I have it been able to find the difference between them. I read somewhere that the stall speed was higher in the 330 version.

Are you able to do you own tuning now or do you need the tuning shop to make any more changes to your tune?

Did the EGS flash & took it for a 5 mile stretch, crazy enough "D" mode is less spunky & more conservative while "DS" is AMAZING. Due to the torque converter concern I just flashed back to 325i which is good due to me driving in "D" 95% of the time anyway.

Not doing my own tuning, don't have the equipment. I called multiple tuners here in Atlanta & they all were trying to complicate a process @hassmachine had already figured out. The guy I found took the project on for free because it was interesting & the hard work was already done. All he had to do was pull, splice & flash.

[hassmaschine]

I should give credit to rjahl as he also worked on figuring out how to set it up (although it is trivial really).

There's a really old post on the 3 stage swap that says it's "not really a copy and paste operation", which is completely false. It IS a copy and paste, except for one pesky byte.

[rjahl]

Originally Posted by dstrickland

Did the EGS flash & took it for a 5 mile stretch, crazy enough "D" mode is less spunky & more conservative while "DS" is AMAZING. Due to the torque converter concern I just flashed back to 325i which is good due to me driving in "D" 95% of the time anyway.

Not doing my own tuning, don't have the equipment. I called multiple tuners here in Atlanta & they all were trying to complicate a process @hassmachine had already figured out. The guy I found took the project on for free because it was interesting & the hard work was already done. All he had to do was pull, splice & flash.

I hope I did not scare you away from the 330 EGS tune too quickly. I was really hoping to learn if that tune changed the RED line shift points at all. In any case there is a thread around here somewhere where people were flashing 325 EGS Software into their 330's looking for improved shift points.

These transmissions are so adaptive you would probably not notice a difference in just a few short minutes of driving.

Regarding the torque converters, I've looked for hours trying to find the difference, it seems that even ZF stopped offering the replacement toque converter on the 325. From what I can tell, rebuild 325 transmissions have the 330 torque converter and the the 325 software.

So far, I have 170,000 miles on my ZF auto. It's holding up well. Just don't let your fluid go over 70,000 miles.