Hello all,

I seem to have my RS232 problem fixed but now I have another problem that
seems more strange to me.  I have an LT1790 1.25V Voltage reference
connected to the Vref+ of the PIC A/D converter and powered from RD3
(PIC16F877A) in my temperature logging circuit.  I can adjust the sampling
interval from the PC over the RS232 link and when I set it to 1 or 2 seconds
a strange problem arises.

The LT1790 along with the LM60 temp sensor are switched on with the PIC pins
when they are needed to save power.  With the sample interval set to 1 or 2
seconds the voltage reference doesn't power on properly every other sample.
Looking with a scope with the interval set to 1s I see the power to the
reference IC rise to the full value every second.  I can't see any
difference from one pulse to the next.  However the output of the reference
is correct one time and then barely rises the next.  It consistently
alternates between accurate output and barely rising above ground.  I want
to stress that this doesn't appear to be random - in the few hundred samples
I have taken and looked at it either works fine or alternates 1 correct 1
wrong.

When I set the sample interval to 4 seconds it appears to work fine.  And
just now I set it to 2 seconds and it worked OK for the few samples that I
watched.  But set it to 1 second and it consistently shows the odd behavior.

Any ideas?

Thanks,
Nick

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.