• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 30 subscribers
  • Views 1880 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TM4C123 Launchpad without crystal does not JTAG/debug

snellr
Intellectual 715 points
Other Parts Discussed in Thread: TM4C123GH6PM

If the 16 MHz crystal Y2 is removed from a Launchpad, the same on-board ICDI can no longer connect to the on-board U1 mcu using the standard CCS debug functionality.  It actually connected the first couple of times I hit the debug button, but on the third and all following attempts the connection attempt fails (CORTEX_M4_0: Error connecting to the target).  Because it fails at connection time instead of program load, I don' t think it is a program issue.  Even when I touch a jumper GND-to-pin40 (OSC0) as per Issue 3 on http://e2e.ti.com/support/microcontrollers/tiva_arm/f/908/t/374640.aspx , it still fails to connect.

Is this spelled out in the datasheet for the TM4C1236GHPM somewhere?  All I see is that the mcu has an internal PIOSC at 16MHz which should be sufficient for basic operation (not USB) .  Why doesn't the JTAG work without the external crystal? Or of course I could be overlooking something.

Can TI staff replicate this? Thanks.

  • 0
    Intellectual 715 points

    I observed the same non-connectivity when I debugged-in to a Launchpad (no crystal) with an XDS100.

    If my observations are confirmed, a custom board made without a crystal would not be programmable.  But the jury is out w.r.t. if my observations are substantiated.

  • 0 in reply to snellr
    TI__Guru**** 244380 points

    Hello Snellr

    On the TM4C123 LaunchPad, I had removed the external crystal and GND the OSC0 Pin and it connected the to the MCU. Otherwise we would have mentioned this as a limitation of TM4C12x devices in the data sheet.

    Regards

    Amit

  • 0
    Guru 47900 points

    Sorry about that - we feel your pain. 

    Would you (temporarily) replace that 16MHz xtal with another xtal of different value?  Suggest this as we employ a similar TM4C with 25MHz external xtal - and often it, "fails to connect."  (is it possible that this MCU "expects" such 16MHz xtal - and hiccups when it's absent?)

    We know that certain ARM MCUs happily run w/out any external xtal - we're told that they often "start" from the internal oscillator.  Yet - perhaps there's a difference between "connecting to JTAG/SWD" and running - when under internal osc. alone.

    Might you try a substitute (different) frequency xtal & report?  We're not fans of CCS - employ paid IAR/J-Link & have built/shipped over 5K boards w/past LX4F & few hundred w/TM4C.  TM4C (to our experience) seems unique in registering this "failed JTAG/SWD connection" issue.

  • 0 in reply to cb1
    Intellectual 715 points

    Thanks for the quick replies, Amit and cb.  Having replaced and removed the crystal again and having established a secure ground to pin 40 this time (rather than just a jumper touch-contact before), I've got a Launchpad system which is now debugging successfully and repeat-ably without the Y2 crystal.  

    Unfortunately, the note about grounding OSC0 in Amit's forum that I referenced initially has been noticed very late in our design process.  The datasheet "pin 40 NC: acceptable practice" turns out to be not so acceptable. (p1352 of tm4c123gh6pm)

  • 0 in reply to snellr
    TI__Guru**** 244380 points

    Hello Snellr

    I agree. It must not be an acceptable practice. Is must be GND for both the coiumns

    Regards

    Amit