Programming Infineon: Flashing Error

[Kingfish]

Keywords: Keywin, Start Transmit, Fail

Greetings –
I finally got my piece parts in to make my USB/TTL adapter so I can finally program my Infineon.
Unit that I have is: EB806XC-A-12, bought from eBikes.Ca, 6FET IRLB4040 25A (for 36-48V nominal).

Followed reference threads:

• Methods 100V 100A Programmable Regen Controller
• Keywin E-bike Lab Parameter Designer Software Manual-ver 1.2
• Headerless 18fet usb/ttl

Using the Keywin software, I kept params nominal, no crazy stuff. The controller is disconnected from everything except the dongle.

• Hit “Start Transmit” and tap the +5V wire to the dongle to complete the circuit
• Software app gets about 86% complete on the status indicator (6/7ths progress) before the app indicates transmit “fail”.

There isn’t a lot of depth on this subject; if you do a forum search with keywords “transmit fail” it returns exactly six finds, four are from
120% speed advanced timing on the controller programming

Has anyone else had this problem? If so – what did you do to resolve it?
Thanks, KF

 

[Lyen]

Hi Kingfish! Please shoot a picture of the MCU (Motor Control Unit) so I can tell if it is programmable or not. Thanks!

 

[Kingfish]

Hi Lyen
Thank you kindly for having a peek at my problem; We must be of like-minds for I was in the process of taking more pics at the time of your posting 8)

[pre]Pic 1: This is an image of my functioning P0 controller; it doesn't have the header mounted,
though we can see the underlying traces and labels better.[/pre]

Additional Notes:

• I performed the HyperTerm test on the dongle and it worked correctly with RX/TX in loop.
• The controller in Pic 2 and 3 has never been hooked up to battery power ever; it is my reference – the one I use for study, and it is unmodified, stock, except for the header. This is the unit I am attempting to flash.
• According to the Schematic for a 6FET (here) the RX and TX should come from CONN-6 and tie into pins 30 and 29 respectively – but they do not. Instead, I have traced them to pins 25 and 26.
• In addition, I have extra resistors (example: R33 in Pic 1) that are not on the schematic. How very odd.
• Speaking of odd, the test pads for X1, X2 and X3 do not go to pins 30, 29, and 28.
• In Pic 1, JZ102 is given as the label for where I would expect CONN-6. JZ102 is not listed on the schematic either.

Conclusion: I need another schematic

[pre]Pic 2: This is the backside of the Reference Controller; it's the same base unit as P0 Controller.[/pre]

[pre]Pic 3: With the stock unit we can clearly see the identifying markings better.[/pre]

Question: Does anyone have an ASV file of the default settings for a 6FET?
I have done a bit of research on these Infineons and they are likely read-protected in terms of firmware. I don't know enough about the write-buffer yet to determine if we can read current contents. I need to be able to establish flash communication first

Thanks in advance for your keen insights, KF

 

[Lyen]

Hello again! I see a few things from your pictures although you did not take a picture with clear marking on the MCU.

1. It uses the previous generation Infineon XC846 MCU
2. It is obviously modified the "unoffical way"

You may have everything wired up correct since you were able to get up to 86%. The only issue is your computer may not supply enough current to the MCU. You may try to use another desktop computer instead of a laptop or notebook type since a typical desktop supply more current to the USB port. Also try to disable the FIFO buffer.

As for the MCU data retrieval. I have already inquired directly in Chinese. The answer is "they" do not want people to hack their MCU so therefore no retrieval program will be provided as of today unless you bought out the company.

Be aware that there are multiple versions of the software. The one you use perhaps is for the new XC116 MCU. Although it also has the 6 FET option but may not be compatible with your XC846.

The ASV file is a very simple text file. You can easily edit with a notepad in Windows and save it as ASV instead of TXT extension.

Your Truly,
Lyen

 

[Kingfish]

Mr. Lyen, kind Sir ~

You nailed it!

I just happened to have several past versions of the software and I tried it with the NON-116 and if flashed perfectly (both with and without FIFO disabled).

Sadly the older software has fewer features. Bummer.

MCU data retrieval: Yep, I suspected that it would likely be password-protected. Not terribly interested in the firmware as I was simply after the previous configuration which is placed in a writable buffer – this way we could amend/adjust instead of wipe/replace. No great loss I suppose. But now that I know that this is an “older model” I must begin a hunt for something more accommodating.

Reading about that 18-FET wonder of the last year was amazing! But I think I need something a bit smaller… and one for each wheel at that. Any ideas come to mind? :wink:

Thank you again! KF

 

[Lyen]

I'm glad you are finally able to get it working.

By the way, which size and power requirement of the controllers you are looking for besides the 18 FET? 6 FET, 9 FET, 12 FET? Please let me know, I should be able to custom made to your specification. Thanks!

Best Regards,
Lyen

 

[Kingfish]

Hello my friend Lyen

By the way, which size and power requirement of the controllers you are looking for…

Oh my! A wish list
Let me address the software first.

Now that I have toyed with both versions of the software I realized that there were some discrepancies between the two, and I have affected the best English translations and literal garbage collection. No code has changed; this is just a bunch of typo fixes on the labels and drop-downs. I cannot ADD to the string lengths but I can EDIT the string contents.

• The 116 prog is more robust; has more features. Use this if your boards are EB2XX series.
• The Infineon prog is older and designed for the EB8XX series boards - like I have evidently...
• The English corrections for both are now in parity. And I corrected the pigeon "all parameter is stuffed" message. Can’t do much with “no use comm” error; string-length limited.

BTW - has anyone thought of using the RESET pin on the controller to re-flash the unit? I'm just thinking this might be easier, and it might also restore default values out of firmware.

Um, these are simple old VB6 apps that opens a COMM port and sends a concatenated string; nothing special except the format/param diffs between the two chips. It would be easier to replace these old apps with something more contemporary and friendly if I had access to the specs. Otherwise I might wait for a rainy day (which are many in the PNW, er – let’s make it a thundersnow since those are a bit more rare with our global warming) before I can crank out a hack: I have bigger fish to fry, like finishing my 2WD 8)

Wish List: Lyen, let’s have a chat!