CoreSTM32 fried???

fcebreros · October 5, 2011

I accidentally touched some upper contacts of the core with a 9v lead so I'm pretty sure I fried the core but let's hope I'm wrong.

I have a short on the 3.3v (I think from the stm )

The status led blinks but with less intensity.

Windows recognizes the virtual midi ports when the card is connected through usb to the pc.

MiosStudio recognizes the card only when I start the sw but not manually afterwards.

No lcd's (or junk data)

Any hope??

rosch · October 5, 2011

if your not sure about whether or notthere's a short you could try at first to measure the resistance between the voltage rails and ground, when powered off.

fcebreros · October 5, 2011

if your not sure about whether or notthere's a short you could try at first to measure the resistance between the voltage rails and ground, when powered off.

Powered off between ground and 3.3v, short. (even without regulator, have searched for solder shorts but nothing..and it was working before)

Edited October 5, 2011 by fcebreros

TK. · October 5, 2011

As long as the LED is working, and if even an USB connection is possible, there is a big chance to solve this! :)

Best Regards, Thorsten.

fcebreros · October 5, 2011

Thanks for the hope!

Any clue?

Could the stm be generating the shorcut because of a sw problem? Could a new bootloader burn solve the problem?

TK. · October 5, 2011

Could the stm be generating the shorcut because of a sw problem?

Which application did you install?

A properly programmed application wouldn't cause such an issue, but if somebody really wants to challenge it, he could enable the output drivers of all pins - even those which are connected to external drivers. This could cause a "short circuit", resp. due to the output resistance a higher current consumption of about 10..20 mA per driver conflict, which could result into a voltage drop if the core module is connected to a weak powered USB port.

Could a new bootloader burn solve the problem?

Under the current "unsafe" condition I wouldn't do this.

E.g. programming the flash will lead to a higher power consumption, with the short circuit this could cause a failure.

Please explain a little bit more about what you did in the past (e.g. which application did you upload, which bootloader version is currently installed), and how your core module is powered, to get more explicit hints.

Is your core working properly in BSL hold mode? (application not started)

Best Regards, Thorsten.

fcebreros · October 5, 2011

Hi,

this is what I get from MIOS Studio2.2.2

Operating System: MIOS32

Board: MBHP_CORE_STM32

Core Family: STM32F10x

Chip ID: 0x00000000

Serial: #35FFFFFF5836393640582543

Flash Memory Size: 524288 bytes

RAM Size: 65536 bytes

MIDIbox SEQ V4.047

with jumper 27 in place i get

Operating System: MIOS32

Board: MBHP_CORE_STM32

Core Family: STM32F10x

Chip ID: 0x00000000

Serial: #35FFFFFF5836393640582543

Flash Memory Size: 524288 bytes

RAM Size: 65536 bytes

MIOS32 Bootloader

I'm powering it through J1 with 9V (1.2A source).

I have changed the 3.3 regulator for a lm1117-33 (respecting the pin order change)

No lcd's and the stm gets very hot

(my multimeter emits an intermitent/noisy beep between vs and vd 3.3V instead of a steady tone, I don't now if it means something to someone)

Best wishes

Frank

Edited October 5, 2011 by fcebreros

fcebreros · October 7, 2011

Hi,

I have checked every cm. and component of the board and find no short, still there is one. I think its the core even if the usb is working.

Any second opinion before I start the long procedure of buying and shipping a new core to my home(I live in the Andes in Cusco, no tech).

Thanks

Frank

rosch · October 8, 2011

maybe slowly rework all the solder joints (with a magnifying glass?). could be there's some bridge that's not easy to see. if there's a short there must be something like that somewhere...

good luck!!

TK. · October 9, 2011

Good news: I just remembered that I wrote an IO test for the MBHP_CORE_STM32 module some time ago which could be very helpful to debug your hardware! :)

-> http://www.ucapps.de/mios32/mbhp_core_stm32_io_test_v1_000.zip

From the README.txt



This application scans all IO pins of the MBHP_CORE_STM32 module to

check for shorts/bridges.


The test will be repeated with a period of 5 seconds


Please open the MIOS Terminal in MIOS Studio to see the test results.


Ensure that no external module/device is connected to the board while

the test is running (no JTAG Wiggler, no MIDI cables, no LCD, no SD Card, no pots, etc...)

[/code]




I checked this with my fully stuffed MBSEQ V4 (no components removed as recommended in the README) and get following ignorable errors:

[code]

[806773.381] ====================================

[806773.381] Testing GPIO Pins for shorts/bridges

[806773.381] ====================================

[806773.383] Push-Pull Toggle Test found 0 errors

[806773.383] Toggle PA3 H->L FAILED

[806773.384] Toggle PB14 H->L FAILED

[806773.385] Pull-Down/Up Toggle Test found 2 errors

[806773.386] Running-1 Test found 0 errors

[806773.387] Running-0 Test found 0 errors

[806773.387] 

[806773.387] ====================================

[806773.387] Test found 2 errors

PA3 fails because MIDI IN3 is connected to this pin

PB14 fails because a DIN shift register is connected to this pin.

What are your results?

Best Regards, Thorsten.

fcebreros · October 10, 2011

Hi,

this are my results:

[169848.821] Testing GPIO Pins for shorts/bridges

[169848.822] ====================================

[169848.822] Toggle PB6 0->1 FAILED

[169848.822] Toggle PB7 0->1 FAILED

[169848.822] Toggle PB8 0->1 FAILED

[169848.822] Toggle PB9 0->1 FAILED

[169848.822] Push-Pull Toggle Test found 4 errors

[169848.824] Toggle PC10 L->H FAILED

[169848.824] Toggle PD2 L->H FAILED

[169848.825] Toggle PB6 L->H FAILED

[169848.825] Toggle PB7 L->H FAILED

[169848.825] Toggle PB8 L->H FAILED

[169848.825] Toggle PB9 L->H FAILED

[169848.825] Pull-Down/Up Toggle Test found 6 errors

[169848.826] Toggle PB6 0->1 FAILED

[169848.826] Toggle PB7 0->1 FAILED

[169848.826] Toggle PB8 0->1 FAILED

[169848.826] Toggle PB9 0->1 FAILED

[169848.826] Running-1 Test found 4 errors

[169848.827] Toggle PB6 0->1 FAILED

[169848.827] Toggle PB7 0->1 FAILED

[169848.827] Running-0 Test found 2 errors

[169848.827] 

[169848.827] ====================================

[169848.827] Test found 16 errors

[169849.827] [MIOS32_MIDI_Receive_Handler] Timeout on port 0x20

[169849.827] [MIOS32_MIDI_Receive_Handler] Timeout on port 0x21

without J26 I get (don't know if it's normal)

[169960.956] Testing GPIO Pins for shorts/bridges

[169960.956] ====================================

[169960.958] Toggle PB6 0->1 FAILED

[169960.958] Toggle PB7 0->1 FAILED

[169960.958] Toggle PB8 0->1 FAILED

[169960.958] Toggle PB9 0->1 FAILED

[169960.958] Push-Pull Toggle Test found 4 errors

[169960.958] Toggle PC13 L->H FAILED

[169960.958] Toggle PC14 L->H FAILED

[169960.960] Toggle PC10 L->H FAILED

[169960.960] Toggle PD2 L->H FAILED

[169960.960] Toggle PB5 L->H FAILED

[169960.960] Toggle PB6 L->H FAILED

[169960.960] Toggle PB7 L->H FAILED

[169960.960] Toggle PB8 L->H FAILED

[169960.960] Toggle PB9 L->H FAILED

[169960.960] Pull-Down/Up Toggle Test found 9 errors

[169960.962] Toggle PB6 0->1 FAILED

[169960.962] Toggle PB7 0->1 FAILED

[169960.962] Toggle PB8 0->1 FAILED

[169960.962] Toggle PB9 0->1 FAILED

[169960.962] Running-1 Test found 4 errors

[169960.963] Toggle PB6 0->1 FAILED

[169960.963] Toggle PB7 0->1 FAILED

[169960.964] Running-0 Test found 2 errors

[169960.964] 

[169960.964] ====================================

[169960.964] Test found 19 errors

[169961.963] [MIOS32_MIDI_Receive_Handler] Timeout on port 0x20

[169961.963] [MIOS32_MIDI_Receive_Handler] Timeout on port 0x21

I hope there is something useful here?

Best regards

Frank

TK. · October 12, 2011

It seems that there is a serious problem with pins PB6..PB9 :-/

When you look into the schematic: http://www.ucapps.de/mbhp/mbhp_core_stm32_v2.pdf you will notice that these are neighbored pins.

Based on the described details I guess that the internal pads are damaged - probably the internal protection diodes are fried, and this causes a short which heats up the chip.

There is a little hope: it could be that the short can be omitted if pad drivers are not activated, or if they are driving logic-0

If this is the case, I could provide a special compile option for you (and teach you, how to compile the MBSEQ application by yourself ;))

In order to find this out, just measure the current in BSL hold mode (when the application isn't started).

If it's still much higher than ca. 100 mA, there is no hope :-(

Best Regards, Thorsten.

fcebreros · October 13, 2011

Hi,

thanks for the support!

But I think it's time to upgrade the core.

If it's still much higher than ca. 100 mA, there is no hope :-(

I have 450mA current draw from the source .

Best

Frank

Edited October 13, 2011 by fcebreros

Sign In

CoreSTM32 fried???

Recommended Posts

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Join the conversation