Forum Discussion

Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
14 years ago

CONF_DONE in failed to go high jtag

With the Cyclone III starter board my USB-JTAG pod fails to configure the board. The error in the Quartus message window is:

Error (209014): CONF_DONE pin failed to go high in device 1

The on board usb blaster works fine (just remove jumper JP8 to select the onboard mode).

The strange thing is it worked briefly (when I showed a colleague of course). While it worked I moved the pod around to see if the cable was the issue, but it kept working. Once I power cycled the board it stopped working in jtag mode (but the onboard circuit always works).

My brand new boards have the same problem - but only jtag mode, so I can not use any onboard blaster. With both boards the pod always detects the device ok, and gives reasonable sounding errors if I power off the board etc.

The debug Pod I have is a "USB blaster" rev B, with a ribbon cable that ends in a 10 pin header.

I am using JTAG mode so the MSEL pins are over-ridden, and I can not think of any other reasons why it should fail.

Thank you for your time and any advice you can give on this.

Steve

14 Replies

  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    --- Quote Start ---

    Thank you for your time and any advice you can give on this.

    --- Quote End ---

    Probe the JTAG pins with a scope. Make sure you use a short ground lead, eg., I'll often solder a loop of wire to a nearby capacitor ground and then loop that around the scope tip. Look at the signal integrity of the TCK pin first; look for ringing and a dip in its rising-edge (which can be caused by a transmission line reflection on that trace). Check the voltage high and low values of the signals. Compare those to the limits stated in the device handbook.

    Cheers,

    Dave
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    Thank you Dave. I looked at the signals:

    The JTAG pins look pretty good - overshoot / undershoot is about 200mV. On the starter kit the JTAG signals are buffered with an octal bus switch that is near to the header. The bus switch should be able to cleanup the small bit of ringing.

    Running the JTAG chain debugger reports that the connection is good and consistent after 1000 IDCODE tests.

    I could load the software on a different PC (use XP instead of Win7). Is there a reasonable chance that could be an issue?

    Thank you,

    Steve
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    --- Quote Start ---

    I could load the software on a different PC (use XP instead of Win7). Is there a reasonable chance that could be an issue?

    --- Quote End ---

    Loaded the software on XP - everything behaves the same as it did on Win7-32bit.

    Possibly its the pod? But it passed the jtag IDCODE tests...
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    --- Quote Start ---

    Possibly its the pod? But it passed the jtag IDCODE tests...

    --- Quote End ---

    Depending on how they generate the IDCODE, you might have a problem with TMS or TDI. The IDCODE is loaded into the shift-DR register on TAP reset, i.e., TMS high and 6 TCK clocks. In that case, TMS does not get toggled. When the IDCODE is read out, TDI is a don't care, so that is also not tested.

    How about trying to read the USERCODE? Or put the device into BYPASS mode, and then just shift a bit pattern through the JTAG chain.

    Cheers,

    Dave
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    At a different location someone got the system working with a revC pod. I have ordered one from Amazon ($25!). Hopefully it will start working for me with the new usb blaster.

    Thank you,

    Steve
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    --- Quote Start ---

    At a different location someone got the system working with a revC pod. I have ordered one from Amazon ($25!). Hopefully it will start working for me with the new usb blaster.

    --- Quote End ---

    Can you post a link to this. The price sounds too low to be a real USB-Blaster. Its likely a knock-off, so it might work, but perhaps be slower than a real USB-Blaster. The $50 devices from Terasic are 'authentic' (approved by Altera) and work great.

    Cheers,

    Dave
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    --- Quote Start ---

    It is the SainSmart USB Blaster Download Cable For FPGA Development Board

    http://www.amazon.com/gp/product/b0057oc5ny/ref=oh_o01_s00_i00_details

    The one product review did say that it ran slower than the other pods...

    --- Quote End ---

    This is a USB-Blaster knock-off. It should work fine.

    The knock-offs are either built using an FTDI+CPLD or using a Cypress FX2 microcontroller.

    At $25 you won't feel bad about tearing it apart to see how it works :)

    Cheers,

    Dave
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    The knock off pod works!!

    It is slower, a bit crashy, and feels warm. But that is so much better than the results I was getting from my other pod, which is apparently broken.

    For the long term I think I will get a better pod.

    Thank you for your help!

    Steve
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    14 years ago

    --- Quote Start ---

    The knock off pod works!!

    --- Quote End ---

    Great! Now pull the cover off, what is inside? :)

    --- Quote Start ---

    It is slower, a bit crashy, and feels warm. But that is so much better than the results I was getting from my other pod, which is apparently broken.

    --- Quote End ---

    Its $25 well spent now that you know where the problem is.

    --- Quote Start ---

    For the long term I think I will get a better pod.

    --- Quote End ---

    Go for the Terasic model.

    Cheers,

    Dave