From: Robert Baer on
David L. Jones wrote:
> Robert Baer wrote:
>> Frank Buss wrote:
>>> David L. Jones wrote:
>>>
>>>> Robert Baer wrote:
>>>>> You can permanently destroy a PicKit 2 programming pod in an easy
>>>>> un-documented manner.
>>>>> 1) Tie one or more pins of a PIC MCU to ground; it does not help to
>>>>> have MCU program use the pin(s) as inputs.
>>>>> 2) Program then power up the MCU with the PicKit 2; you may need
>>>>> to do power down and power up a few times.
>>>>> ZZZZzzzzzaaaaaa:P:! No Poof, No Frap, No Zap; it just gets killed.
>>>>> As far as i can tell the USP port is not damaged; my 1Gbyte stick
>>>>> still reads OK.
>>> I have no such problems with this test setup:
>>>
>>> http://www.frank-buss.de/pic18f2550/index.html
>>>
>>> The PIC is always powered from external power in my test setup. But I
>>> noticed that the PicKit pulls VDD to low, if disabled, which was not
>>> much of a problem, because of my current limited power supply, but I
>>> think this could destroy the prorgammer. How does your schematic
>>> looks like?
>>>> Err, yes they are, at $35 it's one of the cheapest official
>>>> programmers on the market for any micro.
>>> There are some other programmers within the same price range:
>>>
>>> http://search.digikey.com/scripts/DkSearch/dksus.dll?keywords=428-2021-ND
>>> http://search.digikey.com/scripts/DkSearch/dksus.dll?keywords=336-1182-ND
>>>
>>> But you are right, there are more expensive ones, but then usually
>>> with in-circuit debugging support etc.
>>>
>> I use the same programming scheme; the +5V was not activated and i
>> tried running the MCU using the PicKit 2 as the power source.
>>
>> That Cypress programmer is only slightly less expensive and i think
>> i am better off using a "known" where all software and hardware are
>> from the same company.
>
> The Cypress programmer Frank pointed out is not for the PIC, it's for
> Cypress micro's, he was just using that as an example of pricing.
>
> I really fail to see how you can damage your PICkit2 based on your
> explanation of tying PIC I/O pins to ground. I think there has to be more to
> this...
>
> Dave.
>
Well, i have done 3 revs of my application - each time adding
external circuitry, and each time have programmed AND powered the app
via the pod with no problems.
Dozens of used to this point with zero problems.
The only thing different is that i added two hex switches (4
un-protected MCU lines each) and code to support reading of them.
From: Robert Baer on
David L. Jones wrote:
> Robert Baer wrote:
>> David L. Jones wrote:
>>> Robert Baer wrote:
>>>> You can permanently destroy a PicKit 2 programming pod in an easy
>>>> un-documented manner.
>>>> 1) Tie one or more pins of a PIC MCU to ground; it does not help to
>>>> have MCU program use the pin(s) as inputs.
>>>> 2) Program then power up the MCU with the PicKit 2; you may need to
>>>> do power down and power up a few times.
>>>> ZZZZzzzzzaaaaaa:P:! No Poof, No Frap, No Zap; it just gets killed.
>>>> As far as i can tell the USP port is not damaged; my 1Gbyte stick
>>>> still reads OK.
>>> Do you know what exactly got fried?
>>> From your description it appears as though the only overload path
>>> could be if the programmed micro sets one of those grounded pins to
>>> a positive output after it's programmed. That doesn't seem like it
>>> would be enough to destroy the PICkit2 to me. You sure something
>>> else funny isn't going on?
>>>> And the damn pods are not cheap.
>>> Err, yes they are, at $35 it's one of the cheapest official
>>> programmers on the market for any micro.
>>>
>>> Dave.
>>>
>> The program definitely sets the (offending?) pins to input - the
>> state of most pins of MCUs are unknown and random until the MCU
>> program sets their state.
>
> That is not so with PICs. The I/O pins have a known state at power-up. The
> TRIS registers for all the ports default to "1" (input mode).
>
> Dave.
>
I d not know that; i assumed that PICs had the same randumb I/O
problems as other devices i have seen.
From: David L. Jones on
Robert Baer wrote:
> David L. Jones wrote:
>> Robert Baer wrote:
>>> David L. Jones wrote:
>>>> Frank Buss wrote:
>>>>> David L. Jones wrote:
>>>>>
>>>>>> Robert Baer wrote:
>>>>>>> You can permanently destroy a PicKit 2 programming pod in an
>>>>>>> easy un-documented manner.
>>>>>>> 1) Tie one or more pins of a PIC MCU to ground; it does not help
>>>>>>> to have MCU program use the pin(s) as inputs.
>>>>>>> 2) Program then power up the MCU with the PicKit 2; you may need
>>>>>>> to do power down and power up a few times.
>>>>>>> ZZZZzzzzzaaaaaa:P:! No Poof, No Frap, No Zap; it just gets
>>>>>>> killed. As far as i can tell the USP port is not damaged; my
>>>>>>> 1Gbyte stick still reads OK.
>>>>> I have no such problems with this test setup:
>>>>>
>>>>> http://www.frank-buss.de/pic18f2550/index.html
>>>>>
>>>>> The PIC is always powered from external power in my test setup.
>>>>> But I noticed that the PicKit pulls VDD to low, if disabled,
>>>>> which was not much of a problem, because of my current limited
>>>>> power supply, but I think this could destroy the prorgammer.
>>>> It can only pull VDD low with a 1K in series, so that isn't going
>>>> to destroy the programmer.
>>>> www.modtronix.com/products/prog/pickit2/pickit2%20datasheet.pdf
>>>>
>>>>>> Err, yes they are, at $35 it's one of the cheapest official
>>>>>> programmers on the market for any micro.
>>>>> There are some other programmers within the same price range:
>>>>>
>>>>> http://search.digikey.com/scripts/DkSearch/dksus.dll?keywords=428-2021-ND
>>>>> http://search.digikey.com/scripts/DkSearch/dksus.dll?keywords=336-1182-ND
>>>>>
>>>>> But you are right, there are more expensive ones, but then usually
>>>>> with in-circuit debugging support etc.
>>>> The PICkit2 has in-circuit debugging capability, stand-alone field
>>>> programming support, and can power your circuit under test with any
>>>> voltage from 2.8V to 5V. And as a bonus can be used as a 4 logic
>>>> analyser and serial protocol analyser too. Pretty good value for
>>>> money! Dave.
>>>>
>>> There was NO (1K) "protection" resistor (8 lines) from MCU to
>>> ground; at worst 2 pins were shorted.
>>
>> Huh?
>> I'm refering to Frank comment about the PICkit2 circuit and it's
>> ability to pull the VDD pin LOW.
>> According to the schematic for the PICkit2 it's got a 1K series
>> resistor in there for that.
>>
>> Dave.
>>
> There is an *actual* (and correct) schematic for the PicKit-2
> programming pod?
> Where, oh where pray tell?

Well, I typed "PICkit 2 schematic" into Google and the first hit is:
www.modtronix.com/products/prog/pickit2/pickit2%20datasheet.pdf
It's in the users guide that should have came with your PICkit2 on the CD.
Is it "correct"? I don't know that for sure, but it's the official
Microchip document so you'd like to think so.

Dave.

--
================================================
Check out my Electronics Engineering Video Blog & Podcast:
http://www.eevblog.com


From: Robert Baer on
ehsjr wrote:
> Robert Baer wrote:
>> Jan Panteltje wrote:
>>
>>> On a sunny day (Sun, 13 Sep 2009 21:07:10 -0700) it happened Robert Baer
>>> <robertbaer(a)localnet.com> wrote in
>>> <V7ednUh1WdnzXDDXnZ2dnUVZ_qSdnZ2d(a)posted.localnet>:
>>>
>>>> You can permanently destroy a PicKit 2 programming pod in an easy
>>>> un-documented manner.
>>>> 1) Tie one or more pins of a PIC MCU to ground; it does not help to
>>>> have MCU program use the pin(s) as inputs.
>>>> 2) Program then power up the MCU with the PicKit 2; you may need to
>>>> do power down and power up a few times.
>>>> ZZZZzzzzzaaaaaa:P:! No Poof, No Frap, No Zap; it just gets killed.
>>>> As far as i can tell the USP port is not damaged; my 1Gbyte stick
>>>> still reads OK.
>>>>
>>>> And the damn pods are not cheap.
>>>
>>>
>>> I am still using this one, wrote software for it to support every PIC
>>> I needed so far:
>>> http://www.covingtoninnovations.com/noppp/
>>> Blow up as many as you want, less then 1$ part count.
>>>
>> Nice. But..using unsupported PIC parts and have a fast computer - so
>> i will pass on the NOPPP.
>> That site referred to Dontronics which has 4 programmers; all of
>> them are MUCH more expensive.
>>
>
> Maybe you can build your own - don't give up on that idea too soon.
> If you don't like the NOPP, there's plenty more. Here's one example:
> http://www.rentron.com/Myke4.htm
> That's for a serial port. For USB try
> http://www.mcuhobby.com/articles.php?article_id=7
>
> At least reading about it is free. Building a programmer
> might save you some $ so it is worth considering.
>
> Ed
What little i saw at the mcuhobby site is extremely interesting and
it would be nice to make that clone.
BUT.
Other than the "fact" that it may exist, how does one (1) get a
schematic, (2) get the program for the chip tht it uses, and (3) find
out what that chip might be?
The site takes over 3 minutes to load; i gave up at tht point.
From: IanM on
E wrote:
>>> Have you tried updating it (tools ->Download Pickit2 Operating System)
>>> I have noticed it tends to lose its operating system if mishandled.
>> No, but i think that i cannot do that as i fully expect the file to be
>> over a few megs in size (am on dial-up).
>
> Megabytes on a pic?
> Its actually 27 kB zipped from microchip website
>
>
No website access required.
Look for PK2Vnnnnnn.hex in your PICKit2 directory. nnnnnn is numeric
and depends on the firmware version No. If present, download it using
the PICKit2 application.

If *that* doesn't work, go online and grab PK2CMDv1-20.zip (118KB of
command line programmer software for the PICKit2) from Microchip.
<http://ww1.microchip.com/downloads/en/DeviceDoc/PK2CMDv1-20.zip>
unpack it and start a command prompt in its directory. Plug in the
PICKit2, with no target circuit connected, direct to a USB port and run:

PK2CMD -S#

to view connected PICKit2's. If it doesn't see it, re-plug with the
button held down (forces bootloader mode - Busy led flashes) then
release the button and try again.
If it sees it, especially if it sees 'bootloader', run

PK2CMD -D PK2V023200.hex

If it succeeds, re-plug the PICKit2 and run the GUI software, check it
finds the pod then use tools->troubleshoot... and a multimeter to verify
the ICSP interface is functional.

If you are still stuck after that you have two possible options, raise a
support request with Microchip, or crack the case and reprogram the
PIC182550-I/SO via J1 the ICSP pads using another known good programmer.
If it fails, order that PIC, replace the U1 chip and program the new
chip with the firmware the same way.

--
Ian Malcolm. London, ENGLAND. (NEWSGROUP REPLY PREFERRED)
ianm[at]the[dash]malcolms[dot]freeserve[dot]co[dot]uk
[at]=@, [dash]=- & [dot]=. *Warning* HTML & >32K emails --> NUL: