Search

Technical Discussion Group Forum

This forum is provided for user discussion. While Beacon EmbeddedWorks support staff and engineers participate, Beacon EmbeddedWorks does not guarantee the accuracy of all information within in the Technical Discussion Group (TDG).

The "Articles" forums provide brief Articles written by Beacon EmbeddedWorks engineers that address the most frequently asked technical questions.

To receive email notifications when updates are posted for a Beacon EmbeddedWorks product download, please subscribe to the TDG Forum of interest.

TDG Forum

PrevPrev Go to previous topic
NextNext Go to next topic
Last Post 06 Feb 2006 12:02 PM by  afusaro@apexar.com
LH7A404 stays resetted after blanking flash!!!
 8 Replies
Sort:
You are not authorized to post a reply.
Author Messages
afusaro@apexar.com
New Member
New Member
Posts:


--
03 Feb 2006 12:51 PM
    This problem araised after accidentally erasing the Flash memory. We thought it was a Jtag problem (becasue we could communicate with the board), but now we are sure that's not the case.

    Here's the situation:

    When I measure voltage in the reset pin of the Jtag interface (that's directly connected to the reset of the microprocessor, and to a pin in the CPLD), I get 0.2 Volts! Since this input is asserted when set to a low state, we think that the microprocessor is reseted therefore not responding to any command (jtag, uart, etc).

    We think that the problem was caused merely by the blanking of the Flash memory (a software cause), and has nothing to do with a hardware failure. The kit was working perfectly before this happened...

    I would appreciate any suggestion!!!
    kurtl@logicpd.com
    New Member
    New Member
    Posts:


    --
    03 Feb 2006 02:11 PM
    Couple questions to understand the problem better....
    1) Are you currently using the Card Engine on the SDK baseboard or in a custom product?
    2) When you say "reset pin of the JTAG interface" what pin number are you refering to on the standard 20 pin ARM JTAG interface? Pin 3 (nTRST) or Pin 15 (nSRST)?
    3) When you get the .2V measurement, is a JTAG device plugged into the JTAG interface during time of measurement?
    4) If using the SDK baseboad, how do you have the J6 jumper connected?

    Let us know and we can start from there.
    -Kurt
    afusaro@apexar.com
    New Member
    New Member
    Posts:


    --
    06 Feb 2006 07:35 AM
    Kurt, thanks for your reply. Here's the situation

    1) We're using the Card engine on the Zoom SDK baseboard.

    2) Actually we measure 0.2V in both pins (3 and 15). When I press the reset button on the kit, that voltage goes to cero.

    3) No, there's no JTag device plugged. It's just the kit, standing alone.

    4) I have the J6 pin connected according to the application note 248: J6 4-6 connected, and J6 1-3 connected. Nevertheless, If I disconnect these jumpers, I get the same voltage on these reset pins... so it seems that the
    microprocessor stays in a reset state...

    Thank you very much for your support!
    kurtl@logicpd.com
    New Member
    New Member
    Posts:


    --
    06 Feb 2006 10:33 AM
    1) First disconnect any jumpers from the J6 header on the SDK basebord.

    2) Please verify the card engine is firmly seated into the SDK baseboard. Visually verify by looking under the card engine that the J1A and J1B connectors are properly connected to the baseboard so that the card engine PCB is parrallel to the baseboard PCB. -- Lots of people miss this step so please double check.

    3) While the board is powered, look near the 144pin SODIMM connector and measure the voltage shown on the gold vias marked "59" and "61" in silkscreen near the 144pin SODIMM connector's plastic keying tab. What voltages are present?

    4) Place a 2 pin jumper accross J17 pins 23 and 24 so they are tied together, then probe J17 pin 12. What is J17 pin 12's state? high/low? Does the pin toggle? -- Hit the reset button on the kit and watch J17 pin 12. Any change?

    5) Press reset again and look at J16 pin 12, what is this pin's state?

    Keep us posted.
    afusaro@apexar.com
    New Member
    New Member
    Posts:


    --
    06 Feb 2006 11:12 AM
    Kurt, I followed your instructions:

    The board is firmly seated and all the J6 jumpers are disconnected.

    In the vias marked 59 I measure 1.77V, and in vias 61 I measure 3.34V

    I place a 2 pin jumper across J17 pins 23 and 24, and measure J17 pin 12's state: high. The pin does not toggle. When I press reset, nothing changes.

    When I measure voltage in J16 pin 12, I obtain a 2.38kHz, 3,88V(pk-pk) square wave. When I press reset, nothing happens. However when I press the wake up button the signal goes to cero.

    Does this data tell you anything? Thanks!
    kurtl@logicpd.com
    New Member
    New Member
    Posts:


    --
    06 Feb 2006 11:34 AM
    Yep, it tells me lots,
    1) Card engine is getting the correct voltages which would imply that the reset chip (U15) shouldn't be asserting MSTR_nRST.
    2) The CPU wakeup signal is toggling at 2.38Khz so the CPU should jump to run mode as soon as the MSTR_nRST line goes high.
    3) If J17 pin 12 is not toggling when J17 pin 23 is tied to pin 24, then the CPU is not asserting nCS0 which would initiate reading from flash.

    So all those details tells me the CPU is truely in reset. The only device that asserts reset is U15 by design. However, if things go awry there could be other culprits.
    1) Depending on your model number, please verify if R11 and/or R142 are present on the PCB and use and Ohm meter to verify they are intact and ~10K ohms. (R11 is found on the topside of the PCB near the SODIMM key. Between the SOT23 Q9 and the BGA U19. R142 is on the bottom side of the PCB nearest the + side of C53 which is near the TSSOP U21.)
    2) If thats not the case it is possible that the CPLD could be causing you trouble. By design MSTR_nRST is an input to the CPLD, but if the CPLD code is modified unpurpose or by accident through the onboard re-programming buffer or external JTAG device, it is possible that it is not operating correctly and causing your problems.

    -Kurt
    afusaro@apexar.com
    New Member
    New Member
    Posts:


    --
    06 Feb 2006 11:52 AM
    Kurt,

    R11 and R142 are present on the PCB. When the board is disconnected from the kit, I measure ~5Kohm in both Rs. They seem to be intact. Is this OK?

    We thought that the CPLD could be asserting the reset signal, but we were not sure about this.

    Thanks!
    kurtl@logicpd.com
    New Member
    New Member
    Posts:


    --
    06 Feb 2006 11:55 AM
    R11/R142 ~5k Ohm is fine.

    Sounds like you'll need to work through platformsupport@logicpd.com to get your board RMA'd.

    Sorry that it wasn't a simple fix on the forum!
    -Kurt
    afusaro@apexar.com
    New Member
    New Member
    Posts:


    --
    06 Feb 2006 12:02 PM
    Ok Kurt... Thanks a lot!
    You are not authorized to post a reply.