Strange PICkit 2 problem - can't program - Electronic Circuits Projects Diagrams Free

futz · 2nd June 2008, 03:37 PM

Saturday afternoon (coincidentally almost exactly when this site went down) my PICkit 2 stopped being able to connect to my target 16F88 circuit. I had done nothing out of the ordinary. I'd been programming it many times up to the moment it quit working with no problems whatsoever.

I scratched my head and thought maybe I killed the PIC. So I changed chips. No change.

I changed power supplys, both the on-breadboard and then the wall wart. No change.

I took a bare breadboard and built a bare minimum 16F88 circuit. No change. I plugged the 16F628 in there. No change.

I redownloaded the OS. More than once. No change.

Shut down and restarted the computer and replugged the programmer. No change.

I changed the USB cable. No change. Plugged in a powered USB hub and plugged into that. No change.

I took the PICkit 2 off and plugged in the Junebug. No change. Right here is when I really started to worry!

Plugged into my 18F4525 balancing robot. No change.

Tried all kinds of combinations of extra capacitors on the breadboard. No change. My power supply already has caps, so the on-breadboard ones are kind of redundant anyway.

This is freaking me out! Any ideas?

One hint: If I don't power the target and let the PICkit 2 power it, I can connect normally, but I can't program the chip. It makes like it's programming but doesn't verify correctly because it hasn't been programmed. The old program is still there and running. When I turn on target power again and try to connect, it won't.

I've tried erasing the chip. It goes through the motions but doesn't erase.

This afternoon I'm going to plug in my Inchworm+/Unicorn and see what transpires.

futz · 2nd June 2008, 04:36 PM

It appears to be some kind of power thing.

With the Inchworm+/Unicorn I still can't connect with the target powered. But if I turn off target power and let the Inchworm+/Unicorn power it (LCD is dim - borderline not enough power) it works perfect.

But I've used these HVWTech power supplies for a long time and they've always worked flawlessly, so I don't know what's going on. I swapped to another and nothing changes. I measured and it's putting out exactly 5.08 volts.

Switched back to Junebug and same thing. Whenever target is powered it can't read device ID and won't connect. Switch it off and the Junebug connects normally, but won't program because it can't supply enough power.

mvs sarma · 2nd June 2008, 04:54 PM

Futz,
which version driver you are using ? is it V2.5? can you make a try with a lower version and let us see?

link to your power supply is broken i fear. How ever, on the target board you may use 0.1uF MLC cap only across pins 5 and 14, and for a while avoid additional electrolytic caps.

Ps: the link has additional http ( twice) . Now i get it.

blueroomelectronics · 2nd June 2008, 04:56 PM

With the Inchworm+ / Unicorn combo try plugging in a wall adapter to the Inchworms power connector. USB power is way too weedy for some applications.

futz · 2nd June 2008, 04:59 PM

Quote:

Originally Posted by mvs sarma

which version driver you are using ? is it V2.5? can you make a try with a lower version and let us see?

Is this what you're looking for? PICkit 2 OS : 2.20.0 is what it downloads when I tell it to Download OS.

Quote:

link to your power supply is broken i fear.

Oops! Here it is again.

Quote:

How ever, on the target board you may use 0.1uF MLC cap only across pins 5 and 14, and for a while avoid additional electrolytic caps.

I usually don't put any caps on the breadboard at all, since the power supply is already filtered.

blueroomelectronics · 2nd June 2008, 05:06 PM

PS make sure you set the Inchworms VDD +5V power jumper to ON, it's diode blocked from your computers USB and will supply your project with over 500ma. If you leave the jumper off the BS250 can only handle 230ma.

mvs sarma · 2nd June 2008, 05:10 PM

Quote:

Originally Posted by futz

Is this what you're looking for? PICkit 2 OS : 2.20.0 is what it downloads when I tell it to Download OS.

Oops! Here it is again.

I usually don't put any caps on the breadboard at all, since the power supply is already filtered.

Nice
As you rightly said, it had worked well under the same circumstances earlier.
this made me feel whether a SW change took place in between? OK now i understand that you use V2.2

Please add a cap of say 0.1 (100nF)across pins 5 and 14 (18pin devices)
even if there are caps on the power board.

futz · 2nd June 2008, 05:13 PM

Quote:

Originally Posted by blueroomelectronics

PS make sure you set the Inchworms VDD +5V power jumper to ON, it's diode blocked from your computers USB and will supply your project with over 500ma. If you leave the jumper off the BS250 can only handle 230ma.

There's no jumper on this Inchworm+. There is a switch though - EX 5V On/Off.

With the Inchworm+/Unicorn:
Target power on - connects sometimes - programs sometimes
Target power off (Inchworm USB power) - connects - programs
Target power off (Inchworm wall wart power) - connects sometimes - won't program

The PICkit 2/Junebugs do similar strangeness. The odd thing about all this is that I made no changes to the circuit. It worked fine. Then it didn't.

I have to go to work now. Will continue messing with this probably this afternoon or evening.

nickelflippr · 2nd June 2008, 08:14 PM

Here's a long shot idea. Sometimes I have run into a problem when the task manager gets loaded up with a bunch of .exe files that are left hanging. That causes XP to run out of memory, and either clearing out the offending .exe files in task manager, or rebooting is required.

Really sounds like some sort of hardware problem tho, almost like the MCLR pin is acting up for some reason. Maybe try the hardware diagnostic tool and check the voltages, signal frequency, etc.

futz · 2nd June 2008, 10:03 PM

Quote:

Originally Posted by nickelflippr

Here's a long shot idea. Sometimes I have run into a problem when the task manager gets loaded up with a bunch of .exe files that are left hanging. That causes XP to run out of memory, and either clearing out the offending .exe files in task manager, or rebooting is required.

I don't think that's it. My shutting down and rebooting would have fixed it, and anyway this machine has 2GB of RAM so it would take a LOT of programs running in background to crap it out.

Quote:

Really sounds like some sort of hardware problem tho, almost like the MCLR pin is acting up for some reason. Maybe try the hardware diagnostic tool and check the voltages, signal frequency, etc.

Yes it does, but so far I'm at a loss to explain it. Will try some more things this afternoon to see if I can pin it down to something specific.

futz · 2nd June 2008, 10:06 PM

Quote:

Originally Posted by blueroomelectronics

With the Inchworm+ / Unicorn combo try plugging in a wall adapter to the Inchworms power connector. USB power is way too weedy for some applications.

Tried that. No change.

futz · 2nd June 2008, 10:10 PM

Here's another hint: I've noticed that the programmer will almost always connect the first time. But every attempt after that fails with PKWarn0003: Unexpected device ID: Please very that a PIC16F88 is correctly installed in the application. (Expected ID = 0x760, ID Read = 0x0)

futz · 2nd June 2008, 10:15 PM

Quote:

Originally Posted by mvs sarma

As you rightly said, it had worked well under the same circumstances earlier. this made me feel whether a SW change took place in between?

No hardware changes were made.

Quote:

Please add a cap of say 0.1 (100nF)across pins 5 and 14 (18pin devices)
even if there are caps on the power board.

Did that. No change.

blueroomelectronics · 2nd June 2008, 10:26 PM

Tough to tell without a schematic, can you post one?

nickelflippr · 2nd June 2008, 10:37 PM

You did run the Pickit 2 ->Tools->Troubleshoot right?

My settings for Programmer-> verify on write
Tools->fast programming
Target VDD Source->auto detect

I can hang the Pickit 2 by trying to program an 18f when there is a 16f in the socket, and vice versa. Usually hitting the Tools->check communication will put it back on track, barring that, closing and restarting does the trick.

You don't have the Program->manual select device ticked? (i.e. wrong device selected). That would fit the situation you describe.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar
Title	Starter	Forum	Replies	Latest
Program PICs without a PC! BB0703 (PICkit 2) Programmer-To-Go Tutorial	Funny NYPD	Electronic Books	6	15th May 2009 01:43 AM
Swordfish BASIC and PICkit 2 integration (one button compile & program)	blueroomelectronics	Micro Controllers	0	20th December 2007 06:15 AM
Strange Hardware Problem.	tkvenki	Micro Controllers	3	13th June 2007 10:28 AM
Strange problem	tkvenki	Micro Controllers	1	3rd May 2007 12:40 PM
Strange WPU problem.	2camjohn	Micro Controllers	3	10th April 2006 03:05 AM

2nd June 2008, 03:37 PM	#1
futz	Strange PICkit 2 problem - can't program Saturday afternoon (coincidentally almost exactly when this site went down) my PICkit 2 stopped being able to connect to my target 16F88 circuit. I had done nothing out of the ordinary. I'd been programming it many times up to the moment it quit working with no problems whatsoever. I scratched my head and thought maybe I killed the PIC. So I changed chips. No change. I changed power supplys, both the on-breadboard and then the wall wart. No change. I took a bare breadboard and built a bare minimum 16F88 circuit. No change. I plugged the 16F628 in there. No change. I redownloaded the OS. More than once. No change. Shut down and restarted the computer and replugged the programmer. No change. I changed the USB cable. No change. Plugged in a powered USB hub and plugged into that. No change. I took the PICkit 2 off and plugged in the Junebug. No change. Right here is when I really started to worry! Plugged into my 18F4525 balancing robot. No change. Tried all kinds of combinations of extra capacitors on the breadboard. No change. My power supply already has caps, so the on-breadboard ones are kind of redundant anyway. This is freaking me out! Any ideas? One hint: If I don't power the target and let the PICkit 2 power it, I can connect normally, but I can't program the chip. It makes like it's programming but doesn't verify correctly because it hasn't been programmed. The old program is still there and running. When I turn on target power again and try to connect, it won't. I've tried erasing the chip. It goes through the motions but doesn't erase. This afternoon I'm going to plug in my Inchworm+/Unicorn and see what transpires. __________________ ========================= Futz's Microcontrollers & Robotics ========================= Last edited by futz; 2nd June 2008 at 03:52 PM.

2nd June 2008, 04:36 PM	#2
futz	It appears to be some kind of power thing. With the Inchworm+/Unicorn I still can't connect with the target powered. But if I turn off target power and let the Inchworm+/Unicorn power it (LCD is dim - borderline not enough power) it works perfect. But I've used these HVWTech power supplies for a long time and they've always worked flawlessly, so I don't know what's going on. I swapped to another and nothing changes. I measured and it's putting out exactly 5.08 volts. Switched back to Junebug and same thing. Whenever target is powered it can't read device ID and won't connect. Switch it off and the Junebug connects normally, but won't program because it can't supply enough power. __________________ ========================= Futz's Microcontrollers & Robotics ========================= Last edited by futz; 2nd June 2008 at 04:46 PM.

2nd June 2008, 04:54 PM	#3
mvs sarma	Futz, which version driver you are using ? is it V2.5? can you make a try with a lower version and let us see? link to your power supply is broken i fear. How ever, on the target board you may use 0.1uF MLC cap only across pins 5 and 14, and for a while avoid additional electrolytic caps. Ps: the link has additional http ( twice) . Now i get it. __________________ Regards, Sarma. Last edited by mvs sarma; 2nd June 2008 at 05:06 PM. Reason: noted the bug in weblink

2nd June 2008, 04:56 PM	#4
blueroomelectronics	With the Inchworm+ / Unicorn combo try plugging in a wall adapter to the Inchworms power connector. USB power is way too weedy for some applications. __________________ Bill Smart Kits build Smart People http://www.blueroomelectronics.com/

2nd June 2008, 05:06 PM	#6
blueroomelectronics	PS make sure you set the Inchworms VDD +5V power jumper to ON, it's diode blocked from your computers USB and will supply your project with over 500ma. If you leave the jumper off the BS250 can only handle 230ma. __________________ Bill Smart Kits build Smart People http://www.blueroomelectronics.com/

2nd June 2008, 08:14 PM	#9
nickelflippr	Here's a long shot idea. Sometimes I have run into a problem when the task manager gets loaded up with a bunch of .exe files that are left hanging. That causes XP to run out of memory, and either clearing out the offending .exe files in task manager, or rebooting is required. Really sounds like some sort of hardware problem tho, almost like the MCLR pin is acting up for some reason. Maybe try the hardware diagnostic tool and check the voltages, signal frequency, etc.

2nd June 2008, 10:10 PM	#12
futz	Here's another hint: I've noticed that the programmer will almost always connect the first time. But every attempt after that fails with PKWarn0003: Unexpected device ID: Please very that a PIC16F88 is correctly installed in the application. (Expected ID = 0x760, ID Read = 0x0) __________________ ========================= Futz's Microcontrollers & Robotics =========================

2nd June 2008, 10:26 PM	#14
blueroomelectronics	Tough to tell without a schematic, can you post one? __________________ Bill Smart Kits build Smart People http://www.blueroomelectronics.com/

2nd June 2008, 10:37 PM	#15
nickelflippr	You did run the Pickit 2 ->Tools->Troubleshoot right? My settings for Programmer-> verify on write Tools->fast programming Target VDD Source->auto detect I can hang the Pickit 2 by trying to program an 18f when there is a 16f in the socket, and vice versa. Usually hitting the Tools->check communication will put it back on track, barring that, closing and restarting does the trick. You don't have the Program->manual select device ticked? (i.e. wrong device selected). That would fit the situation you describe.