Cart New Account Login

HomeAbout usProductsSupportForumsBlogCustomer Service

search inside this forum
USB-ML-CF Rev C and USB-ML-CFE Rev D difference
dave s. Nov 24, 2015 at 01:04 PM (13:04 hours)
Staff: Takao Y.

  • Hi
    I have had a revision C device for production programming of MCF5208 devices and used it for many years. I recently purchased a revision D and it does not work the same. Both are funning the same flash firmware 5.54
    I should point out that I do not use all on the connections to the MCF5208 on my board as space is very limited.
    When I try with a 3rd party development board, which has all pins connected, then both programmers work perfectly so the new one is not faulty, just different.
    The only pins I use are:-
    2 BKPT
    3 GND
    4 DSCLK
    5 GND
    7 RESET
    8 DSI
    9 VCC
    10 DSO
    11 GND
    12 PST3
    13 PST2
    15 PST0
    20 GND
    23 GND
    24 CLK
    As I said previously this has worked perfectly with my old device and I have programmed 1000's of my boards.
    Any help appreciated
    Dave




    Comments

  • Greetings,

    How does it not work the same. You need to be more detailed. Does it flash program? Is the problem in debugging? Is your program not running?


    Takao Yamada

  • Hi Takao
    I am using both Multilink units for production flash programming only.

    I am using CPROGCFZ to program the devices and when the window pops up during the programming process it closes immediately when the error occurs so it is difficult to even see the error message. Eventually I got it on a PrtScr dump which I cannot attach to this message but reads:-

    Initializing. -USB-ML-CF detected - Flash Version 5.54
    Coldfire Core type detected as 52xx
    Not able to access processor. Try a software reset.
    ERROR 35 during script!

    Hope this helps.
    regards
    Dave

  • Greetings,

    You need to modify the batch script (.bat file) so that it has the parameter '?' at the beginning. Also, you should try lowering the debug shift frequency to option 10. For example:

    CPROGCFZ ? mycfg.cfg bdm_speed 10 interface=usbmultilink port=usb1

    Could you try the example above and see if you are able to get better results. If you still have issues, then you need to open the PROGCFZ software (not CPROG) and try to debug from there. Try all of the debug shift frequency options to see if you are able to talk to the chip at all.


    Takao Yamada

  • Hi Takao
    That made no difference, I do not feel that this is a timing issue.

    Did you see my comment that I do not have all of the programming lines connected and that it works if I connect all of the programming lines.

    My revision C Multilink works with the reduced pin connections.

    The only pins I use are:-
    2 BKPT
    3 GND
    4 DSCLK
    5 GND
    7 RESET
    8 DSI
    9 VCC
    10 DSO
    11 GND
    12 PST3
    13 PST2
    15 PST0
    20 GND
    23 GND
    24 CLK
    Which works on the revision C but not the revision D.

    regards
    Dave

  • Hi Takao
    Note that I do not have PST1 connected which might be considered a bit of an anomaly but it was not required on rev C. Is it possible that you handle this line differently on the two revision (pull-up, pull-down or floating)?
    regards
    Dave

  • Hi Takao
    Correction I do have PST1 connected, this was an error on my schematic.
    So I have the following connections.
    2 BKPT
    3 GND
    4 DSCLK
    5 GND
    7 RESET
    8 DSI
    9 VCC
    10 DSO
    11 GND
    12 PST3
    13 PST2
    14 PST1
    15 PST0
    20 GND
    23 GND
    24 CLK
    I have found that it seems to be some sort of ground problem. If I connect to more ground points of my board then it works. However this is not possible with my programming fixture.
    Why would the rev D be more susceptible to grounding than the Rev C?
    regards
    Dave

  • Greetings,

    After a phone call late we found that the rev D needed the jumper setting to be NOCLK. This jumper setting is found within the multilink to switch between synchronous and asynchronous devices. More information can be found in the technical summary of the multilink, which can be downloaded from the multilink's product page.


    Takao Yamada

    • The user reported that after some more testing they were still having problems connecting to their MCF5208 target despite moving the jumper to NOCLK. They found that if they transferred the TVS diode T2 from their USB-ML-CFE Rev C to the USB-ML-CFE Rev D (where it is not populated), then the Rev D worked perfectly. The RevC is still works perfectly. Their custom ribbon cable and bed of nails fixture adds about 3 to 4 inches of cable length. They would like to know our thoughts on this and if we had any possible explanation.

Add comment


   Want to comment? Please login or create a new PEMicro account.







© 2017 P&E Microcomputer Systems Inc.
Website Terms of Use and Sales Agreement