[sdiy] I don't feel so stupid now...

[patchell]

    I am using the PC900...but, I know it is working fine becasue Midi Thru
Data gets to down stream Midi Instruments just fine.

    Even very simple code, ie, code that makes some LEDs (8 of them) on the
front panel "count", and that is all the program does, works unreliably.  I
used a scope to check the important signals, like the Reset Line and the
oscilator, everything there is running fine.  And checking the other signals
with a scope did not reveal any obvious problem.  I began to suspect that the
AVR chip might be damaged.  So, I put it back into STK 200 board, and it runs
perfectly reliably there, so, the problem is obviously on the PC board,
somewhere.  Hopefully, I will have some time today to work on it a bit more.
I suspect the problem is in the external memory interface.  Trouble is, there
isn't a whole lot that can go wrong there.  The most obvious is that the ALE
line might be broken, or shorted, since when it doesn't work, I don't see this
signal (thin evidence, I am sure there are many ways to keep ALE from
working).

    When I get back to work, I am going to add code to the loop that counts
that will output a signal to one of the AVR peripheral pins as well as writing
to the port that lights the LEDs.  At least this way, I will be able to
determine if the AVR is just plain not working, or the if it something more
subtle like a broken ALE trace, or something else.  But, judging from what I
am seeing, it looks like the AVR just doesn't boot, which brings up the
posibility of power supply problems.  Statically, the +5 volts is just fine.
However, when the power comes on I don't know what is happening.

    The other symptom I noticed is that when it does come up and run fine, I
can pull the reset line down and release it and the AVR ceases to function.
No amount of reseting will get it to run again.  Recycling the power a few
times will.  Again, I need to emphasise that the same chip is 100% reliable in
the STK200 board.

    I am willing to bet that there is some little thing I needed to do that I
didn't.  But, so far, this is still a big mystery.

Ingo Debus wrote:

> patchell wrote:
> >     Well, at least I now know it is not my code on the AVR that is the
> > problem.  I verified that the Midi was really getting to the UART pin by
> > using the Midi Thru connector going to another instrument (My
> > Glockenspiel, in fact).  And while I was observing that the Midi was
> > just fine, I looked over and noticed that the code was running on the
> > AVR just fine as well.  But, alas, that was the last time.  I turned the
> > power off and on again, and it was back to the same old thing.  Wiggling
> > wires would make it work for a bit, so the conclusion is that there is
> > either an intermittent short or open on the PC board (these were not
> > great PC boards...these were the last boards I had the local PC board
> > shop make about 3 years ago now I think).
>
> I had a similar problem a while ago (and people from the list here
> helped me to solve it). It was even worse: the thing worked with some
> MIDI data sources and didn't with others. It turned out that I hadn't
> used the proper optocoupler. Which are you using?
>
> Ingo

--
 -Jim
------------------------------------------------
* Visit:http://www.silcom.com/~patchell/
*-----------------------------------------------
*I'm sure glad Merry Christmas comes just once a year
* -Yogi Yorgensen
------------------------------------------------