DIY: F8x Software Update Tips

[aboulfad]

Originally Posted by HDDMark

Aboulfad,

I was on 3.26.XXX before. Evidently from PSdZdata 58.0 onward you need E-SYS 3.27.1.

http://cartechnology.co.uk/printthread.php?tid=13675

Cool tip, we usually always use the latest of ESys but glad that you solved it and I will update the main thread!

[aboulfad]

So I brought my car for service, and they noticed that the Weather widget displays the location and the date over each other making a garbled display, it never bothered me as it disappears after a while, but they decided they wanted to update the software : i wanted to face FEM_GW in battle, now I may not have that opportunity and I know some people are begging to have their cars updated :

[terahertz]

I was reading another thread and shawn posted this which seems to say the minimum voltage is 12.6v instead of the 13v I thought it was:

*NOTE: A continuous power source of at least 12.6 Volts is needed. Failure to maintain proper voltage can lead to corruption of ECU Data. As such, ideally the car is connected to an external charger. In lieu of using an external charger, the car can instead be coded with the engine running in order to maintain proper voltage, which is acceptable for coding all ECU's EXCEPT the DME (Digital Motor Electronics) ECU or DDE (Digital Diesel Electronics) ECU. The DME/DDE are typically not coded anyway, so this ECU limitation is generally not an issue.

[aboulfad]

No, reading the BMW programming doc says 13.0V I'll link it later.

[shawnsheridan]

Originally Posted by aboulfad

No, reading the BMW programming doc says 13.0V I'll link it later.

Really?



BMW Programming, Coding, and Dignosing.pdf

[aboulfad]

Actually I forgot I had it linked in my first post, click on the words minimum 13.0 in the first post. The source is Rheingold for a 2015 M4 which is circa 2016, the document that shawnsheridan showed is from 2008. Both could be correct but I'll use the recent recommendation for our F8x platform.

[shawnsheridan]

Originally Posted by aboulfad

Actually I forgot I had it linked in my first post, click on the words minimum 13.0 in the first post. The source is Rheingold for a 2015 M4 which is circa 2016, the document that shawnsheridan showed is from 2008. Both could be correct but I'll use the recent recommendation for our F8x platform.

I don't think .4 Volts matters one way or another.

[aboulfad]

Originally Posted by shawnsheridan

...
I don't think .4 Volts matters one way or another.

: I think you said the same thing over at BF. Given i don't have your experience level in flashing BMWs, I prefer to refer DIYers to the "letter of the law" and then anyone can do what they want at their own risk if any...

[dzvero]

Question regarding flashing as I will need/planning to flash/update . Is it ok if I start with some errors : missing airbags and strut with EDC for example or it is not advisable untill I fix these issues on the car?

[aboulfad]

Originally Posted by dzvero

Question regarding flashing as I will need/planning to flash/update . Is it ok if I start with some errors : missing airbags and strut with EDC for example or it is not advisable untill I fix these issues on the car?

Good question, I don't know the answer, Mark read somewhere that you need to clear all faults before flashing, but if you think about it, all that could happen is maybe you loose those faults/errors and it becomes more difficult to diagnose after? Given you have a salvaged car, maybe fix the errors then flash, in any case what would you get by flashing first?

[dzvero]

pass smog smog doesn't care for airbags, only check engine. If I lose errors means they are not persistent , if they come back new diagnostic will be in place.

[aboulfad]

Originally Posted by shawnsheridan

I don't think .4 Volts matters one way or another.

Funny, investigating other items in the other thread http://f80.bimmerpost.com/forums/sho...&postcount=119 basically reminded me why your document says 12.6V, that's the nominal voltage of lead acid based car batteries!!!

For the F8x lithium batteries, the nominal voltage of a LifePO4 battery is 13-13.2V and hence why the more recent appropriate programming recommendation for this platform.

At 12.6V, the LifePo4 battery is close to less than 10% charge... So I believe but not 100% sure that the PSU voltage needs to be above 13.0V for the F8x lithium batteries.

[terahertz]

Originally Posted by aboulfad

Funny, investigating other items in the other thread http://f80.bimmerpost.com/forums/sho...&postcount=119 basically reminded me why your document says 12.6V, that's the nominal voltage of lead acid based car batteries!!!

For the F8x lithium batteries, the nominal voltage of a LifePO4 battery is 13-13.2V and hence why the more recent appropriate programming recommendation for this platform.

At 12.6V, the LifePo4 battery is close to less than 10% charge... So I believe but not 100% sure that the PSU voltage needs to be above 13.0V for the F8x lithium batteries.

The recommended programming voltage doesn't really have anything to do with the battery... The modules need sufficient voltage to write the flash memory. Like shawn said, it probably doesn't matter lol.

[aboulfad]

@terahertz, So you think it's for no reason that BMW changed the specific recommendation for programming?

If you followed the post in your thread, you can maybe deduce a correlation between programming voltage and battery nominal voltage.

Hypothesis: The programming voltage at the PSU can not be lower than the battery nominal voltage, if not its the battery that is supplying the current and that is undesirable.

[HDDMark]

Originally Posted by aboulfad

@shawnsheridan, some errors at the end of TAL execution, is it again the stupid ignition ? or a dependency? This is when i was flashing EGS, GHAS, VDC. car is alive though but with the usual transport mode and faults.

MCDDiagService<id=46900, job=com.bmw.psdz.jobs.uds.MCD3_GetSVKfromECU, service=RDBI_SVK - ReadDataByIdentifier SVK, description=error: timeout, link=Physical request to ECU with address 0xf>
MCDDiagService<id=46900, job=com.bmw.psdz.jobs.uds.MCD3_GetSVKfromECU, service=RDBI_SVK - ReadDataByIdentifier SVK, description=error: timeout, link=Physical request to ECU with address 0xf>
MCDDiagService<id=46900, job=com.bmw.psdz.jobs.uds.MCD3_GetSVKfromECU, service=RDBI_SVK - ReadDataByIdentifier SVK, description=error: timeout, link=Physical request to ECU with address 0xf>
[GHAS - F] There was an error during TAL execution, please check the log files.
[GHAS - F] - [Exception - GHAS - f] negative response error: 
 code: TIMEOUT P2/T2 occurred
 description: P2 timeout on Service RDBI_SVK has error in result: [Severity=2051][ErrorCode=58640][ErrorCodeDescription=resource not available][VendorCode=2142][VendorCodeDescription=P2 timeout occured]; ECU: GHAS_0F_ETHERNET
 severity: ERROR

MCDDiagService<id=80900, job=com.bmw.psdz.jobs.uds.MCD3_FinalizeECUCoding, service=WDBI_CPS - WriteDataByIdentifier Codierpruefstempel, description=error: timeout, link=Physical request to ECU with address 0x76>
MCDDiagService<id=80900, job=com.bmw.psdz.jobs.uds.MCD3_FinalizeECUCoding, service=WDBI_CPS - WriteDataByIdentifier Codierpruefstempel, description=error: timeout, link=Physical request to ECU with address 0x76>
MCDDiagService<id=80900, job=com.bmw.psdz.jobs.uds.MCD3_FinalizeECUCoding, service=WDBI_CPS - WriteDataByIdentifier Codierpruefstempel, description=error: timeout, link=Physical request to ECU with address 0x76>
MCDDiagService<id=230900, job=com.bmw.psdz.jobs.uds.MCD3_FinalizeECUCoding, service=DSC - DiagnosticSessionControl, description=error: timeout, link=Physical request to ECU with address 0x76>
[VDC1 - 76] finalizeECUCoding error
[VDC1 - 76] There was an error during TAL execution, please check the log files.
[VDC1 - 76] - [Exception - VDC1 - 76] negative response error: 
 code: TIMEOUT P2/T2 occurred
 description: P2 timeout on Service DSC has error in result: [Severity=2051][ErrorCode=58640][ErrorCodeDescription=resource not available][VendorCode=2142][VendorCodeDescription=P2 timeout occured]; ECU: VDC1_76_ETHERNET
 severity: ERROR

[] finalizeVehicleCoding started
[] finalizeVehicleCoding finished
[] finalizeTALExecution started
[] finalizeTALExecution finished
TAL execution finished
TAL-Execution finished with status: "FinishedWithError". [C207]
TAL execution finished. Duration: "6min 55s". [C206]

I had similar errors while flashing VDC and GHAS. I cleared the codes, deleted the transport mode, closed E-Sys and tried again. Second time everything went fine. I was beginning to get a little worried but since I knew you had this issue I didn't freak out. There was no indication of the ignition cutting off during the process. Very strange.

Oh well. All's good.

[aboulfad]

Originally Posted by HDDMark

I had similar errors while flashing VDC and GHAS. I cleared the codes, deleted the transport mode, closed E-Sys and tried again. Second time everything went fine. I was beginning to get a little worried but since I knew you had this issue I didn't freak out. There was no indication of the ignition cutting off during the process. Very strange.

Oh well. All's good.

Cool! Most of research and experiments I did above show its a loss of connection to the FEM_GW, but I can't explain why that happened now twice with two different users for the same ECUs... From my other experiments in the PSU thread, the ignition behaviour is very bizarre. Did your errors also happen during VO coding (cdDeploy)?

I did set up a DHCP server on the PC, and I was able succeed in communicating to the car, but never got the chance to flash again to see if it helps. (And I discovered after that ISTA/D didn't like the setup, so when I went back to APIPA, the car was stuck with the DHCP IP Addr... It took me a while to get it back to APIPA).

Btw, I know you are somewhat following what I initially did, but I believe that to complete the pic requires also flashing the fuel pump, steering and other essential ECUs. whats your next step?

[HDDMark]

Aboulfad,

I forgot to mention that I'm on a native Windows 10 laptop. So I'm not sure it's a problem with a virtual OS on Macs.

Unfortunately, I didn't save the error messages. I looked at the E-Sys log files and they are huge. I'll see if I can find the actual error messages in there.

As far as my next step, I haven't thought about it until you mentioned it. My original goal was to get improvements in the throttle mapping, suspension tuning and DSC without wiping out my coding. The fuel pump and steering ECUs might also yield some improvements. I also heard in a thread that I don't remember that the update to 58.3 might improve the function of the NGHB.

[aboulfad]

I don't run VM, it's bootcamp, which installs Windows native on a Mac, it's a dual boot system you don't have to dig the details as most likely it was the same. Good luck with the rest !

[dzvero]

Originally Posted by shawnsheridan

I don't think .4 Volts matters one way or another.

Is that what she said ,
In serious note 13 or 12.6 is same, no issues IMHO. Even 12V will be ok as long as PSU can sustain the Amps drawn and not fall below 12 ! main point is not to fall below as there is some cable resistance which can lead to lower voltage on some ECU's . There is always insurance - when they say 12 they mean 10

[aboulfad]

Originally Posted by dzvero

Is that what she said ,
In serious note 13 or 12.6 is same, no issues IMHO...

Read this carefully and then please show your experiments and results, please. Have you flashed the F8x ECUs correctly with either 10V or 12V with battery in the car? My experiments, equipment and theories could be flaud so I am all ears if someone can prove or show otherwise, not just say "it doesn't matter". I really genuinely want to get to the bottom of this.

The point is to flash safely and ensure that the voltage supplied is stable and the current is supplied by the PSU and not the battery itself. When Vpsu is way lower (10V) than Vbatt, guaranteed its your battery that is supplying the 20-40Amps and it will drain and kill a lengthy flash session... because a LifePO4 charge is like 10% at 12.6V.

And here again is the LifePO4 battery document from BMW.

[dzvero]

Again the BMW folks want to insure their self's as any other manufacturer does. 13V comes from batteries voltage as you noticed. So if your ECU on 428 can work with 12.6 why same ecu on M4 cant ? You will be really in trouble if the voltage goes below 10 for sure. Will it work when its 11 ? possibly yes. will it work if its 12V Im sure it will . Electronics in this case flashing the memory wont just fail if the voltage is 12. It might fail if its 11, I have no specific details what chips they use etc. I can ensure you it wont fail if its still 12V. The voltage will go down if the PSU+battery cant handle the current. If the PSU can handle more then 20A (someone mention this as current withdrawn when coding) it will be ok .

[aboulfad]

I follow what you say and agree that with a sufficiently suitable stable voltage that there won't be any damage to any ECU, be it M4 or 428. But my point was not that.

Another reason those recommendations exist is that those PSU dealers use are meant as "Battery support chargers", meaning that after whatever programming/other operation is done, you are not left with a drained battery.

For that not to happen, the external PSU needs to supply the majority of the current to the car. That can not happen when the PSU voltage is way lower than the battery voltage.

Now, why did BMW change their recommended programming voltage to 13.0 for F8x, and I'll verify with a 428 VIN if it was changed from 12.6 too., don't know! One key difference between those two platforms is the battery type, coincidence or is there a correlation, or to be more safe like you say?