TMS320R2811 JTAG communication

Hello,

I have built a board with TMS320R2811 on it. When I power it up
oscillator works, and on RS pin I can see the watchdog reset as it is
supposed to, so I assume that the core is alive. But I can not
communicate with the DSP, through JTAG with XDS510PP+. I connected all
the JTAG pins as they are supposed to (according to ezDSP R2812
schematics). I had no problems with my previous boards with 24xx DSP's
on them. I have checked all connections and they are fine.
Has anyone experienced similar problem?

Mitja

Issues with the XDS51PP emulator seem to be quite common from the
number of times the subject appears to have come up.  Try searching the
archives of this group and also on TI's discussion groups.  You will
probably find a number of ideas and suggestions to try.

On 6 Dec 2005 23:00:44 -0800, "Korenje" <korenje@yahoo.co.uk> wrote in
comp.arch.embedded:

> Hello,
> 
> I have built a board with TMS320R2811 on it. When I power it up
> oscillator works, and on RS pin I can see the watchdog reset as it is
> supposed to, so I assume that the core is alive. But I can not
> communicate with the DSP, through JTAG with XDS510PP+. I connected all
> the JTAG pins as they are supposed to (according to ezDSP R2812
> schematics). I had no problems with my previous boards with 24xx DSP's
> on them. I have checked all connections and they are fine.
> Has anyone experienced similar problem?
> 
> Mitja

Are you powering the debugger with the separate wall wart, or are you
depending on power being supplied from the JTAG header on the board?

There is a mistake that is easy to make.  The XDS510PP+ can be powered
directly from the target board if and only if the board provides +5
volts on the JTAG header power pin.  If you have 3.3 volts on the JTAG
power pin, the debugger uses it to detect power on in the target, but
can't run from it, and needs the plug-in supply.

We made that mistake on the first spin of our first 2812 board, and it
was quite a nuisance.  We fixed it on the second spin.  Fortunately we
had a 5 volt supply on the board.

-- 
Jack Klein
Home: http://JK-Technology.Com
FAQs for
comp.lang.c http://www.eskimo.com/~scs/C-faq/top.html
comp.lang.c++ http://www.parashift.com/c++-faq-lite/
alt.comp.lang.learn.c-c++
http://www.contrib.andrew.cmu.edu/~ajo/docs/FAQ-acllc.html

Thank you both for advice.

Jack, I did route +5V to PD pin, though I admit it is probably very
common mistake. I learned that by myself on earlier 24xx boards.

Since I have check everything at least three times I am going to
assume, that the one of the DSP's JTAG pins is faulty (maybe more), and
I am going to replace the DSP.

I just had to be sure that no one else experienced similar problem,
since this is my first board with 28xx device.

Thanks again

Mitja

On 8 Dec 2005 04:49:28 -0800, "korenje" <korenje@yahoo.co.uk> wrote in
comp.dsp:

> Thank you both for advice.
> 
> Jack, I did route +5V to PD pin, though I admit it is probably very
> common mistake. I learned that by myself on earlier 24xx boards.
> 
> Since I have check everything at least three times I am going to
> assume, that the one of the DSP's JTAG pins is faulty (maybe more), and
> I am going to replace the DSP.
> 
> I just had to be sure that no one else experienced similar problem,
> since this is my first board with 28xx device.
> 
> Thanks again
> 
> Mitja

Does your board have an external watchdog circuit?  Ours does, because
it is in a safety critical device (medical).  Without one, a single
point failure such as a solder connection to the crystal stops both
the DSP and the internal watchdog.

If you do have an external watchdog, you must disable it somehow when
using the JTAG.

-- 
Jack Klein
Home: http://JK-Technology.Com
FAQs for
comp.lang.c http://www.eskimo.com/~scs/C-faq/top.html
comp.lang.c++ http://www.parashift.com/c++-faq-lite/
alt.comp.lang.learn.c-c++
http://www.contrib.andrew.cmu.edu/~ajo/docs/FAQ-acllc.html