• State TI Thinks Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 1 answer
  • Subscribers 81 subscribers
  • Views 4270 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Newer MSP-FET barely works

Wayne S
Intellectual 365 points
Other Parts Discussed in Thread: MSP-FET, MSP430F5659, MSP430FR5849, MSP430F5528, MSP430F5310, MSP430L092, MSP430FR5969

I purchased the newer black MSP-FET tool and use with IAR tools. It is constantly asking to update the FET firmware and complains about setting the voltage, whether I power the device or the FET powers the device. The older MSP-FETU430IF does not have this problem. This is the message I get:

Using license: Standalone license - IAR Embedded Workbench for Texas Instruments MSP430 6.30
Firmware version 3.5.1.1
Interface dll version 0.0.65471.65535
Could not set device Vcc (Vcc) , voltage=0x9c4
Could not set device Vcc : (Close) , vccOff=0x1
Fatal error: Failed to set device voltage.   Session aborted!

After this if I try again sometimes it will work, sometimes not. It does not matter whether I am programming my system (msp430f5659) or programming the same micro in the MSP-TS430PZ100USB development board. Or even if I use with other processors (msp430f5528, msp430f5310, msp430fr5849, ...). The older FET works everytime. The newer FET is frustratingly unreliable.

  • 0
    Guru 74080 points
    Would be interesting to see if the FET behaves the same when using it with CCS. Maybe you want to try that, just for interest. I work with the new MSP-FET for quite a long time now and never had problems.
  • 0 in reply to Dennis Eichmann
    Intellectual 365 points
    I don't have CCS, however, the driver for the FET is the same and comes from TI so I don't believe it will make a difference. Also, the MSP-FET behaves the same on a different computer. First attempt always fails, second attempt will sometimes work. If Second attempt fails, then it will never work. Unplugging the FET and restarting it, sometimes gets it to work or if I do that enough times will give me BSOD. For now, I am using the older grey FET.
  • 0 in reply to Wayne S
    Intellectual 365 points
    BTW, once I do get it to work for a debug session, it will always work for subsequent debug sessions, until I disconnect and re-connect the FET. There is something about the initial state of the FET when first powered on that seems to create the problem. What I don't understand is why, when if fails, it wants to update the FET firmware.
  • 0 in reply to Wayne S
    Intellectual 365 points

    In response to my own post, I found a work around, which to me indicates there is a bug in the FET firmware and/or PC driver. The work around is, after connecting the FET to the PC, do an initial attempt to load code while the JTAG is not connected. That will fail, finding no device, but then the FET JTAG can be connected to the system board and it will work as long as the FET is not disconnected from the PC.

    The initial state of the FET/Driver is the problem. My system runs at 2.5V, perhaps the initial state of the FET is 3.3V and collides with my system voltage, but gets set properly after the first attempt. As mentioned, the older MSP-FETU430IF does not have this problem.

    Another option I will try is to put a 100Ohm resistor in series with the FET Vcc voltage line.

    This does not, however, fix another problem, where running the debugger for a long time, or multiple times, followed by disconnecting/reconnecting the FET to the PC, can generate a BSOD.

  • 0 in reply to Wayne S
    Intellectual 365 points
    Plugged in the new FET JTAG today after rebooting from windows update and got immediate BSOD. Can someone at TI please fix their drivers?
  • 0
    TI__Intellectual 1140 points
    Hi Wayne,

    The firmware version which, is programmed on your MSP-FET is correct. But the Interface DLL version is wrong.

    In this state the MSP-FET won’t except any commands like setting the device voltage.

    Could you please check the MSP430.DLL version, which is part of your IAR installation.

    You can find it under:
    IAR installation folder\430\bin\MSP430.dll
    Do right mouse click and properties\details to see the version number.

    Thanks & Regards
    Florian
  • 0 in reply to Florian Berenbrinker
    Intellectual 365 points
    File Version 6.40.1.950
    This is with Version 6.40.1 of IAR tools
  • 0 in reply to Wayne S
    TI__Intellectual 1140 points

    HI Wayne,


    We are currently looking into a similar issue. I have one question about the LED status of the MSP-FET.


    Does the Green LED stays on or is it off?

    Is only the Red LED on?


    Thanks & Regards

    Florian

  • 0 in reply to Florian Berenbrinker
    Intellectual 365 points

    Green light is on, when it asks to upload firmware. Red light is on when first connecting to hardware. As I have described: If I try and program without connecting to hardware, which fails, then the second attempt usually works. Or if I happen to time it right and connect the JTAG while the FET is being initialized it also works.

    Now I have a new problem. I update to IAR 6.40.2/3 which updated the firmware in the FET and now the FET will not even attempt to program my device. It completely fails claiming "the hardware might drain too much current from the debugger". That is bogus, it is a very low power system. I tried setting the voltage to multiple levels around my operating voltage. I tried programming just the main board without the power board. Nothing works now. So I have rolled back to previous IAR 6.40.1 and restored this firmware and I am back to working.

    I suggest the issue with recent firmware is an attempt to set voltage before programming, good fix, but in the process is not letting the system settle, as there can be large caps on the power supplies. The FET needs to accommodate potential in-rush current on startup due to supply caps.

    Obviously TI tried to fix this issue with new firmware, but in the process rendered my FET useless.

  • 0 in reply to Wayne S
    TI__Intellectual 1140 points

    HI Wayne,

     

    If this message is shown by IAR. It seems like you drain more than 100mA out of the MSP-FET. This results in a mode that the FET will go into overcurrent.

     

    The Fet firmware is written in a way that I can load lager caps by loading them pulse wise. Can you please post your connection schematic.

     

    Thanks & Regards

    Florian

  • 0 in reply to Florian Berenbrinker
    Intellectual 365 points

    My hardware runs off of a 4-20mA loop and pulls less than 1mA at 2.5V. There is no way it would work at all if it pulled more than 2mA. As I said before, reverting the firmware to previous version works. I am able to get the FET to work, IAR 6.40.4 after computer reboot. Not sure if firmware is the same as in 6.40.2
     

  • 0 in reply to Wayne S
    Intellectual 365 points

    My previous answer, rebooting, worked for a while but once the FET decides it is pulling too much power, it stays that way until I reboot the computer. The previous firmware/driver would work if I first tried to load code while JTAG disconnected from hardware, then connect later and it would run. Now if I get stuck in the "pulling too much power" nothing will fix the problem except a computer reboot.

  • 0
    TI__Intellectual 1140 points
    HI Wayne,

    i think i have been able to recreate you issue. I can provide you an update driver to check, if my idea works for you. Please dorp me a private message with you mail address in this case.

    Thanks & Regards
    Florian
  • 0 in reply to Florian Berenbrinker
    Prodigy 10 points

    I am also facing similar issue with MSP-FET(Windows10 64bit) , when I try to debug MSP430L092 target board with JTag, first time debug went good. and when i unplug FET device and tried again it says unknown device :(.  again Unplugged and re started testing with new MSP430FR5969 experiment board's Jtag, showing same error 'unknown device'. and every time it is asking me to update firmware. is there any way to fix it?can any one help on this..

  • 0 in reply to Meena TANALA
    Intellectual 365 points

    The FET will stop working if there is too much current required to power the board. I have seen bugs in previous code I loaded that can prevent the JTAG from gaining control. To fix that I have had to try and get the device into BSL mode, then the JTAG could gain control.

    In my original issue, I had a shunt regulator and if I set the FET voltage too high, it would not work. It is also possible that the code you loaded the first time set some output pins that are now causing excess current. In which case, try to get the device into BSL mode first.

    Is the JTAG supplying power to the board? I suggest connecting external power to the board and see how much current it is pulling. If High, then find the reason for that and see if you can fix it, possible disconnecting parts. The JTAG can either supply power or monitor the supply voltage (JTAG pin 2 vs pin 4). If you can supply external power, set the JTAG voltage to something lower than the supply or only use the monitor feature (pin 4 vs pin 2).

    You can also try and set the FET voltage to lowest voltage the part will run at, and mass erase the device.

**Attention** This is a public forum