On 27 Sep 2006 16:10:50 -0700, Robert Adsett wrote:
/cut
Keil uVision3
A man from Keil will come next week and take a look at the board. If the
CPU si dead we will try to replace only the CPU.
Thanks for your answer.
Reply by Robert Adsett●September 27, 20062006-09-27
Mad I.D. wrote:
> On Tue, 26 Sep 2006 22:22:21 +0200, Mad I.D. wrote:
> /cut
>
> Thank you Robert and Bob.
>
> When I try to access the chip via Philips flash utility I get "Can't read
> device, device error etc.
>
> When I try to access the chip via JTAG I can't erase or download the flash
> in it by I CAN DEBUG CODE ALREADY IN. I get "prefech abort" after around
> 100 instruction. (CPU fetches instruction which can't execute).
>
> So if I can't erase the flash either way it's dead althrough CPU is
> operating :-(
You might want to ask again in the yahoo group. This question comes up
fairly often. The debug failure could just be a program bug or it
could be a JTAG issue, but more information would be needed to track
that down.
The serial ISP problem sounds exactly like what you would get if you
are not pulling down the ISP pin. In that case you can download into
an unprogrammed chip but not again after the first download.
As far as programming via JTAG the obvious question would be what SW
are you using? I don't use the JTAG for programming but others do, so
if they know what SW and HW you are using they may be able to help
further.
Robert
Reply by ●September 27, 20062006-09-27
On Tue, 26 Sep 2006 22:22:21 +0200, Mad I.D. wrote:
/cut
Thank you Robert and Bob.
When I try to access the chip via Philips flash utility I get "Can't read
device, device error etc.
When I try to access the chip via JTAG I can't erase or download the flash
in it by I CAN DEBUG CODE ALREADY IN. I get "prefech abort" after around
100 instruction. (CPU fetches instruction which can't execute).
So if I can't erase the flash either way it's dead althrough CPU is
operating :-(
Ivan
Reply by BobH●September 26, 20062006-09-26
Mad I.D. wrote:
> I have one question.
>
> What happens if I send data from one pin to another directly without any
> resistor?
>
> Please answer if one is input and other output, and when both are output
> (short connection).
>
> We think that's what happend.
>
> Thanks!
>
> Ivan
It is very rare to need a resistor to connect an input to an output.
About the only time is when driving different voltage level ports like a
5 Volt output port driving a 3.3 volt input port, but if both are on the
same chip I doubt it.
Connecting an output port to an output port might kill one of the ports
but I doubt it. I have done this by mistake a number of times with no
ill effects. As I said, it might kill the specific port but the rest of
the chip should operate.
If the whole chip is unresponsive, I would look at stuff like the clock
generator setup or maybe the program got corrupted by interrupting the
flash operation. An outside chance would be ESD or static shock damage,
I have killed a few parts this way through the years.
Good Luck,
Bob
Reply by Robert Adsett●September 26, 20062006-09-26
Mad I.D. wrote:
> I have one question.
>
> What happens if I send data from one pin to another directly without any
> resistor?
You didn't like my earlier answer :) ?
To quote
"If they are both set as output, maybe. If at most one is set as an
output
then no. Even if both are set as outputs they would have to be set to
different polarities to cause damage.
It's possible that switching transients from different edge timing
could
cause a problem if both are switched at the same time but I doubt that
the
transient would be long enough."
Robert
Reply by ●September 26, 20062006-09-26
I have one question.
What happens if I send data from one pin to another directly without any
resistor?
Please answer if one is input and other output, and when both are output
(short connection).
We think that's what happend.
Thanks!
Ivan