PIC programmer acting up
Will
chips (DIP socket) for a little over a year now. Very recently, it
has started to fail very frequently. I ordered a bunch of new chips
just in case I fried some or something, and only one attempt out of
twenty or so succeeded with the new chips. Even that one did not
behave the way it should have when I plugged it into the circuit so
I'm suspicious of it. Has anyone else evey had a picstart plus fail?
Is there an easy fix? Other ideas?
-Will
Dennis Clark
What's the failure mode? Is it possible you got a bad batch of PICs?
DLC
: I have been using a PICSTART PLUS programmer to program PIC16F877
--
============================================================================
* Dennis Clark Aristocrat at heart d...@frii.com www.techtoystoday.com *
* Be well, do good work, and stay in touch -- Garrison Keillor *
============================================================================
F. DeLuca
not yet support?
"Will" <lark...@aol.com> wrote in message
news:5faf6114.02092...@posting.google.com...
M. Adam Davis
First test your power brick under load. If it isn't outputting what
it should be outputting then replace all the elcrolytic caps in it.
Test it again and you ahould be good to go.
If it tests out ok, then open up your picstart and see if the caps
look bad. Replace the electrolytics for good measure if you don't
know what voltages to measure for where.
Alternately call microchip and complain. They may send you a new
power brick.
It goes without saying that if you got newer chips then the picstart
may not program them without a firmware and mplab update, but MPLAB
should complain about it, instead of trying to program them.
-Adam
M. Adam Davis
The newest chips, the 18fxxx series, have support for the production
parts but only with the beta version of MPLAB (which doesn't work with
the older chips right now). You, of course, need to upgrade the
firmware in the picstart for these later chips.
-Adam
F. DeLuca
new 16F87XA family. In the 16F87XA family the FLASH array has been changed
and the programming algorithms have changed. The PIC 16F87XA has only been
available for a couple of months now. As far as I can tell (and I'm an FAE
for a Microchip distributor) Microchip has not released upgraded firmware
for the PIC Start Plus that supports the 16F87XA family.
Frank
"M. Adam Davis" <dav...@engin.umich.edu> wrote in message
news:3297pug9nn2iprlul...@4ax.com...
Will
> First test your power brick under load. If it isn't outputting what
> it should be outputting then replace all the elcrolytic caps in it.
> Test it again and you ahould be good to go.
>
> If it tests out ok, then open up your picstart and see if the caps
> look bad. Replace the electrolytics for good measure if you don't
> know what voltages to measure for where.
>
> Alternately call microchip and complain. They may send you a new
> power brick.
>
> It goes without saying that if you got newer chips then the picstart
> may not program them without a firmware and mplab update, but MPLAB
> should complain about it, instead of trying to program them.
>
> -Adam
I double checked both the new batch of chips and the couple of old
ones that had been failing. All are 16F877, not 16F877A. I tried
lokking at the power brick. It outputs 11.55VDC when connected to the
PICSTART which I thought might be a tad high since it is rated to 9VDC
and the programmer requires 9VDC on the input (or so the case says).
I opened it up and replaced the capacitor, but it is still at 11.5VDC.
I tried measuring it while the PIC was programming and it was still
11.5VDC. Do I need a new supply or is this within specs?
Also, to rule out any other problems, I switched to a new computer to
program and am using a new serial cable. Nothing seems to make a
difference. I did manage to get it to program once yesterday by
placing a 10KOhm resistor between RB3 and GND which an obscure section
of the help file mentions for low voltage programming on the F877, but
the second time I tried it it failed again so I'm guessing it was just
a fluke. What else should I look at?
-Will
Robert Gush
> I double checked both the new batch of chips and the couple of old
> ones that had been failing. All are 16F877, not 16F877A. I tried
> lokking at the power brick. It outputs 11.55VDC when connected to the
> PICSTART which I thought might be a tad high since it is rated to 9VDC
> and the programmer requires 9VDC on the input (or so the case says).
> I opened it up and replaced the capacitor, but it is still at 11.5VDC.
> I tried measuring it while the PIC was programming and it was still
> 11.5VDC. Do I need a new supply or is this within specs?
> Also, to rule out any other problems, I switched to a new computer to
> program and am using a new serial cable. Nothing seems to make a
> difference. I did manage to get it to program once yesterday by
> placing a 10KOhm resistor between RB3 and GND which an obscure section
> of the help file mentions for low voltage programming on the F877, but
> the second time I tried it it failed again so I'm guessing it was just
> a fluke. What else should I look at?
You could try checking that pin 1 on the pic (mclr) is raised to 13 volts
during programming. There is a little switchmode supply in the programmer
that generates this voltage and it could be faulty. Also check that the pic
VCC is applied and that Rb6 and Rb7 toggle furiously during programming.
We've actually made a header that goes to the programmer picking up the
right pins and use it for ICP. If the target has an odd PSU voltage then we
only use 4 wires and not the VCC one
Regards - Robert
Will
9VDC power supply and got a single success out of it, but then two
failures so I don't know if it was a fluke or not.
-Will