• State Resolved
  • Locked Locked
  • Replies 13 replies
  • Subscribers 80 subscribers
  • Views 2877 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

how to check msp430 is not dead or corrupted?

Sergey Vladimirov
Expert 1300 points
Other Parts Discussed in Thread: MSP430F5528

Hi,

We just got a new batch of boards using MSP430F5528.  However im not able to program them using SBW. There was nothing changed in design around programming part also I see power supply is correct. But I get error in CCS: Unknown device.

Is there a way to find out that msp430 is actually functioning properly?

Thanks,

Sergey

  • 0
    Guru 46710 points

    Sergey Vladimirov said:
    Is there a way to find out that msp430 is actually functioning properly?

    Measure all the voltages on __unprogrammed__ "OK" chip pins, compare to chip in question?

    It's hard to suggest something else if you can't even flash the chip. Unsolder msp430 chip from new board, put it onto old (proven to work) board and try to flash it. If it works, then msp430 is ok but there's something wrong with PCB itself, soldering or components as such or their placement

    p.s. Sometimes stupid **t happens - like PCB pick&place operator put symmetric component (msp430) on PCB rotated by 180 degrees and nobody notice until hundreds of boards are manufactured

  • 0
    Guru 33653 points

    Sergey Vladimirov said:

    We just got a new batch of boards using MSP430F5528.  However im not able to program them using SBW. There was nothing changed in design around programming part also I see power supply is correct. But I get error in CCS: Unknown device.

    Is there a way to find out that msp430 is actually functioning properly?

    If you have USB BSL on-board possibility, maybe you can try to flash it with USB BSL.

  • 0
    Guru 58965 points

    Sergey Vladimirov said:
    We just got a new batch of boards using MSP430F5528.  However im not able to program them using SBW.

    Are all the boards in this batch not working? Do you still have some boards from the previous batches? If so, do they still work now?

    I am thinking that there could be something wrong with your current PC or CCS.

  • 0 in reply to old_cow_yellow
    Expert 1300 points

    The old boards are still working. I tried 2 out of 20 new boards and 2 are not working.

    Would really like to avoid resoldering from new to old just to check that MCU is working properly.

    And unfortunately i dont have USB connection, so cant use USB BSL.

  • 0 in reply to Sergey Vladimirov
    Guru 46710 points

    Sergey Vladimirov said:
    Would really like to avoid resoldering from new to old just to check that MCU is working properly.

    Take in account that __every__ msp430 is tested during manufacturing, so in your list of suspects it shall be the last. I would blame/check the PCB or assembly/soldering/other_components first

  • 0 in reply to Sergey Vladimirov
    Guru 58965 points

    Sergey Vladimirov said:
    The old boards are still working. I tried 2 out of 20 new boards and 2 are not working.

    Hence I would conclude that either (a) the new PCB has a problem, or (2) the components soldered to the new PCB have a problem.

    Please check the marking on the MSP chips to see if they are correct. I assume TI ships chips with the same marking to many different customer. It is highly improbable that it there is a 10% failure rate, none of the other customer did not notice and did not complain.

    On the other hand, how many others are using the same new PCB you are using? Why do you question the MSP and not the PCB?

  • 0 in reply to Ilmars
    Guru 227245 points

    Ilmars said:
    Take in account that __every__ msp430 is tested during manufacturing, so in your list of suspects it shall be the last


    Well, there have been cases where a batch of MSPs had wrong calibration values which might lock-up normally working code. But it is a rare case.

  • 0 in reply to old_cow_yellow
    Expert 1300 points

    Of course im questioning my PCB first of all. The problem is that VCC is on pins that need VCC, GND is also vonnected. And SpyBiWire is also connected correctly. So where should I look for a mistake?

  • 0 in reply to Sergey Vladimirov
    Guru 58965 points

    What is connected to the VCORE pin?

    Is VCC connected to all AVCCx pins and all DVCCx pins?

    Is GND connected to all AVSSx pins, all DVSSx pins, and the thermal pad (if any)?

    What other things (if any) are connected to the SpyBiWire pins?

    If you have a scope, check the waveform of SpyBiWire pins. What are the voltage levels at various point in time? What are the rise and fall time of the transitions?

    You could use one of your old working PCB with a working MSP as reference. But remember to erase the code in that MSP before you try to compare with the new PCB with new MSP (which has erased Flash.)

  • 0 in reply to old_cow_yellow
    Expert 1300 points

    To VCORE we got 470nf capacitor connected.

    all xVCC pins are connected to VCC.

    all xVSS pins are connected to GND. Also VSSU is grounded.

    On RST pin we got 2200pf cap and 47K resistor.

    on TEST nothing special connected. 

    I checked with the scope and there is just constant VCC voltage on the RST pin and 0V on TEST...

  • 0 in reply to Sergey Vladimirov
    Guru 58965 points

    Sergey Vladimirov said:
    ...

    On RST pin we got 2200pf cap ...

    That seems excessive. It may cause excessive rise and fall edges when RST is used as bi-directional "data" during SBW sessions.

    Sergey Vladimirov said:
    ...

    I checked with the scope and there is just constant VCC voltage on the RST pin and 0V on TEST...

    Did you observe them while your SBW tool is trying to access the chip?
    During this, RST pin is used as a bi-directional "data" signal, TEST is used as the "clock" for that "data" from the tool side. If they are just constant DC, the tool is not reaching the MSP.

    Compare this with a known working board (from the previous batch).
  • 0 in reply to old_cow_yellow
    Guru 227245 points

    old_cow_yellow said:
    On RST pin we got 2200pf cap ...

    That seems excessive[/quote]

    2200pF is just 2.2nF and the recommended maximum if you plan to use SBW.

  • 0 in reply to Ilmars
    Expert 1300 points

    Was looking for a problem for 2 days. Then decided to remove a working MCU from old board and put a suspicious one from the new board and - it didnt work.

    Then placed a working MCU on the new board and it worked.

    So seems like I got 22 dead MSP430F5528 :S

**Attention** This is a public forum