• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 32 subscribers
  • Views 481 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.

TM4C1294NCPDTI

Mike Nudelman
Intellectual 295 points

Other Parts Discussed in Thread: LM3S811

Hi!

I am having trouble with TM4C1294NCPDTI chip trying to connect to it with the debugger which is the LM3S811 eval board by TI.

Now, to anticipate some questions: the debugger works with the same processor on the other board I created before.

On this board the core was copy-pasted in the new schematic.

The CPU has clock frequency on the crystal pins.

The debugger gives message "failed to initialize the target".

All signals seems to be fine.

The debugger does not produce the hardrware reset either.

I am puzzled. I keep going through the two schematics and they are identical.

need help.

Thanks

Mike.

  • 0
    TI__Guru**** 199296 points
    Mike,

    It may be caused by many factors.

    1. Power rails on the board: 3.3V, 1.2V
    2. The length of the Wires for JTAG signal from LM3S811 board to TM4C129 board
    3. The noise level on your board
    4. etc.

    If the JTAG signals are > 2in. (51mm), or routed near an area where they could pick up noise, please use 10-k pull-up or 1-k pull-down to TCK.

    Regards,
    QJ
  • 0 in reply to QJ Wang
    Intellectual 295 points
    1. The 3.3V is 3.3V, the noise with 20MHz band cannot even be seen on 20mV/div AC coupled with short GND. Same goes for VDDC (1.2V), the noise is below what I can see on 20mV/div. With full bandwidth (1GHz) it is about 7mVp-p on either rail and I get the same when I simply short the probe with short GND.
    2. The lengths are: TDI - 1.3", TDO - 1.5", TCK - 0.5" source-terminated 22 Ohm (use this everywhere, same topology in all boards), TMP - 0.6". All signals have 10K pullups. Again, this is the same arrangement I use everywhere with this processor.
    3. As the board is similar to others and has very much the same core and the same power supplies, there is no reason for it to be any noisier which (I gave the noise numbers above) is very low. The board topology is 4lyr, top-bottom are routing, 2nd GND and 3rd PWR.
    4. Etc. This is I would like someone with better suggestions than "etc" to elaborate.
  • 0
    Intellectual 295 points
    I think I might have an idea. I followed the datasheet advice for the RBIAS resistor preferred practice with Ethernet not used as "NC". However my oscillator which is inherited from the previous board is 25MHz and this might try to enable Ethernet and there is no bias.
    So I will try to connect the resistor and see if it is going to connect. If it works, I think I will use a different crystal, like 16MHz.
  • 0 in reply to Mike Nudelman
    Guru 92251 points

    Mike Nudelman said:
    I think I might have an idea. I followed the datasheet advice for the RBIAS resistor preferred practice with Ethernet not used as "NC".

    You might have been affected by the following errata:

  • 0 in reply to Mike Nudelman
    Intellectual 295 points
    Yep that was it. So much for best practice of leaving the Rbias unconnected without very clearly spelling "unless your crystal is 25MHz".
  • 0 in reply to Mike Nudelman
    Guru 47900 points

    Glad that you succeeded - we (long ago) used that "3S811 board" w/success as ICDI.   I recall that one - of the 20 pin JTAG header pins - had to be grounded - as it held the 3S811 MCU in reset.

    Present this - here/now - as (yet another) public service for (others) who may employ this (bit long - and now departed) MCU eval board...