Confusing problen with PicKit 2

[Computerman29]

Greetings, all!

I am having a problem with my PicKit 2. It is my design as in the thread: https://www.edaboard.com/threads/118478/.

It worked flawlessly once I got the mosfets properly wired. I have since used it many times with several different pic's with no problems. Now it don't see any PICs. When I run the trobleshooting program, Vdd test reads 4.5 volts in the software, like it always has. I confirmed it with both a scope and meter, but now it says it is an "unexpected voltage on the Vdd pin." At the same time is says it passed. PGC toggle, PGD toggle, Vpp, and MCLR tests all pass.

I have checked all connections and and the board for shorts, I tried a different cable, I checked the voltage on my USB port (it reads 5.0 volts), I re-programmed the firmware, and I tried it on a different PC - all to no avail.
Any ideas what the problem might be? I am stumped!!! Help, please!!!

 

[kasamiko]

Check coil..and the area involved in Vdd voltage generation...

 

[wp100]

Hi,

Too many pages to that link to search to see what hardware you have built....

You make no mention of what you are using to program the pics, in a dedicated zif socket or icsp on a board ?

Also what program are you using to control it, Mplab, Pickit2 V2.61 or command line ?

Do any of the Pickit2 other functions work like the usart tool or logic tool ?

Although you have used two different pcs, sometimes USB voltage can cause strange problems, perhaps just try the Pk2 with battery power to prove the point.

Without meaning to sound clever, I built a cheap pickit2 Lite to backup my Microchip Pk2 for just this sort of eventuality.. you realise how stranded you are when it goes bust as my first Microchip unit did - (replaced with a new one by the supplier )

 

[Computerman29]

I built the original hardware - NOT Nishal's hardware. It has wire jumpers all over, but it worked once I got the mosfets wired properly.

I am using PicKit2 2.61 software on a breadboard, and powering the pic with the PicKit 2.

I tried the UART tool in the software. All appeared to be ok, but then again, it wasn't hooked up.

Have not tried battery power. Will try that next.

I have an old Olimex JDM that I use in emergencies - like this one.

 

[wp100]

Hi,

Not sure if you mean your Pk2 is on a breadboard or just the target circuit.

The Pk2 should be soldered on a circuit or strip board and the connections to the target should be as short as possible.
Any wires over 9" long between the Pk2 and Target chip and you can get such random problems.

The crystal and caps should be as close to the 2550 as possible and again on a board, not on jumper wires.

 

[Computerman29]

PicKit 2 is built on a circuit board.

PicKit 2 has .1uf ceramic cap bypassing the power pins on the 2550 (not in the schematic).

Crystal and caps are almost touching the 2550.

The intended target is a PIC12F675 on a breadboard. Have also tried a PIC16F819 and it would see it sometimes, but it would lose the connection whenever I tried to do anything.

The icsp cable is about 9" - maybe slightly more. Shortened the cable and trying again.

Stay tuned.

 

[wp100]

Hi,

Yup, think you have found the problem, as you said it recognised the 16F819 sometimes, had a similar problem with a 16F628A, onces the icsp leads were shortened below 6" it was fine.

Also to prove the point, try direct programming in a soldered socket on a bit of strip board and plug the short icsp lead onto it as breadboard connection are notorious for causing problems with 'fine' high speed signals like icsp and xtals.

 

[Computerman29]

Shortened the icsp cable to about 5-6". Still a no-go. Will try a different USB cable. If that don't work, then I will try the stripboard.

 

[mvs sarma]

hi, Computerman29,

Can you please replace the diode 1N5819 or some such in the Vdd path after the coil.

 

[nandhu015]

have you tried replacing the firmware chip?

 

[mvs sarma]

@Computerman29,
Please have a new artwork attached. the jumpers have been minimized to the extent possible.
A newer version PK2CM29C is posted correcting minor errors.

 

[Computerman29]

Thank you very much, Sarma!

Btw, I replaced all of the diodes and the op-amp. Still no-go. Will try new MOSFETS next. I will then a different 2550 if that don't fix it. I don't think that is the problem, but it is worth a try. Stay tuned!

 

[mvs sarma]

perhaps, your 18F2550 may not be faulty. which board you made now? is it my artwork? please upload crisp photos (close up) of the the wired board both sides so that we can have a remote check for possible mistakes
I recollect my college days and exams, where once a mistake is done , any cross check , mind feels correct.
totaling etc , we used to do from reverse order to overcome that mindset. likewise perhaps some one else could find the bug of what we wire.

all the best for 2012 and hope to see CM29 doing more projects, to encourage the edaboard members

PS: CM29, can you please check placement of Q1 FET (NTD20P06 ) for polarity correctness. this is specific as you havd earlier shown it connected reverse. the Vusb just passes thro' the diode inside, and comes out with no control by Gate drive.
Later you have placed it properly and it worked. Perhaps the same error , if committed can be corrected .

 

[Computerman29]

Replaced the MOSFET's and checked the polarity very carefully. It still reads an "Unexpected voltage on Vdd" with no cable attached. Used a different cable about 3 inches long. It now works properly with 16LF819 and 16LF833A, but It still reads an "Unexpected voltage on Vdd." It did not work at all with 12F683 on either manual or autodetect.

Btw, it is my original hardware with the MOSFET's polarity corrected.

 

[mvs sarma]

Are you able to to caibrate nd later during trouble shoot, manage to adjust it to 4.8V or so?
(see the sch i sent you pickit2 cm29_sch.pdf), imagine U5P is monted reverse, the Vusb goes thro the diode inside the device and reaches the Vdd point with no control by Gate of U5p.this voltage is monitored at pin3 of micro via R26.
secondly the device U5P once wired correctly, but leaky, then also same effect . please check wrt to hius info in mind.
regarding 12F683, perhaps many people had this issue.
please check this aspect on the forum of microchip in the pickit2 section. Mean time i wonder whether you are close to any hoigh power transmitter/?
if so
please wire 47pcaps from ecach pin if icsp to ground.
this should solve any induction getting rectified within the pcb. await feedback.

 

[Computerman29]

I checked again the polarity of all the MOSFET's. They are all wired properly.
I discovered that my mutlimeter is faulty. Using a different meter, I am able to calibrate Vdd at 2.5V to 4.7V within +/-.1V.
USB voltage on my computer is 5.0V.

Still says "Unexpected voltage on Vdd."

 

[mvs sarma]

i was unable to check that message in the list of messages.you might well place the dmm and check for any trace of voltage on the Vdd line that is output of Q1 the pmosfet after the opamp.

possible by leaky D4 1N5819, i have not heard of such leaks, though. microchip site refers to replacement of D4.