TheBackShed.com

Forum Index : Microcontroller and PC projects : PIC-KIT3 29 volts on MCLR. Why?

Author

Message

crez

Senior Member

Joined: 24/10/2012
Location: Australia
Posts: 152

Posted: 02:14am 01 Sep 2015

Copy link to clipboard

Print this post

device: OLIMEX PIC-KIT3
operating under IPE v2.25
device selected : PIC32MX470F512H
under advance options: power device from tool.

when I click connect I get two 100ms pulses 1.5 seconds apart of 29 volts on the MCLR line.

Do I have some setting incorrectly set or do I have a faulty PIC-KIT?

two dead mx470's so far...

MicroBlocks

Guru

Joined: 12/05/2012
Location: Thailand
Posts: 2209

Posted: 02:18am 01 Sep 2015

Copy link to clipboard

Print this post

Goto settings->Advanced mode.

Click Power and make sure the checkboxes under ICSP options are set right.
There is a "High voltsge on MCLR" checkbox that might be on.

I use IPE v3.05 (Maybe upgrade to that version, might be a bug in the previous ones as the mx470 are newer chips)
Edited by TZAdvantage 2015-09-02

Microblocks. Build with logic.

robert.rozee
Guru

Joined: 31/12/2012
Location: New Zealand
Posts: 2294

Posted: 02:29am 01 Sep 2015

Copy link to clipboard

Print this post

could be a good reason to not use a PICkit...

it sounds like your PICkit 3 could be mistakenly identifying the PIC32 devices as some other family that requires/supports high-voltage programming. although, from what i recall, 29 volts is just a tad high for Vpp.

before doing any further testing, i'd be soldering a 3v3 zener diode between MCLR and ground! either at the target, or on the cable coming out of your OLIMEX programmer.

cheers,
rob :-)

crez

Senior Member

Joined: 24/10/2012
Location: Australia
Posts: 152

Posted: 02:30am 01 Sep 2015

Copy link to clipboard

Print this post

Thanks TZ,
under voltage settings it has "VPP: 3.3 N/A"
under ICSP options I have ticked "power target from tool"
the other two options ('low voltage program' and 'high voltage on MCLR') are greyed out and not ticked.

I will see If i can find V3.05

crez

Senior Member

Joined: 24/10/2012
Location: Australia
Posts: 152

Posted: 02:32am 01 Sep 2015

Copy link to clipboard

Print this post

Rob, what options are there besides PIC-KIT?

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 8601

Posted: 02:34am 01 Sep 2015

Copy link to clipboard

Print this post

When you selected PIC32MX470 did the system load the PIC32MX firmware to the programmer? The pickit3 needs differnt firmware depending on the target chip. It sounds like the programmer is operating in a mode for one of the early PICs that needed a high voltage on MCLR to initiate the programming sequence. However, even for these the voltage shouldn't exceed 13.5V maximum.

Try using one of the dead chips but use external power for the PIC and then measure the voltages. You do have VCAP connected and some 0.1uF caps between VCC and GND near the processor?

Difficult to diagnose remotely but it sounds like a faulty PicKit. The genuine Microchip programmers aren't very expensive and I have found them to be very reliable

robert.rozee
Guru

Joined: 31/12/2012
Location: New Zealand
Posts: 2294

Posted: 02:40am 01 Sep 2015

Copy link to clipboard

Print this post

crez said Rob, what options are there besides PIC-KIT?

pic32prog and an Arduino NANO. see this thread:

http://www.thebackshed.com/forum/forum_posts.asp?TID=7762

cheers,
rob :-)

JTR0701
Regular Member

Joined: 10/07/2013
Location: Australia
Posts: 71

Posted: 03:17am 01 Sep 2015

Copy link to clipboard

Print this post

Sound to me the pickit 3 clone is very faulty, like one of the feedback resistors is open. I would not be looking to use it for any PIC without really troubleshooting this. A pickit should not ever produce that sort of Vpp output under any normal operating condition...

crez

Senior Member

Joined: 24/10/2012
Location: Australia
Posts: 152

Posted: 02:35am 02 Sep 2015

Copy link to clipboard

Print this post

The MCLR(yellow) and VDD(blue) pins of the programmer do this:

but the 13v peak only happens on the first CONNECT command after the USB is connected. Subsequent CONNECT commands behave as they should. These readings were taken with nothing connected. With a 4k7 load on the MCLR to GND, the peak is much shorter but still way over 3.3v

As a precaution I have added 3.3v zeners between all the lines and GND. I also issued a connect command before connecting the target, and it seems the latest programming attempt has been successful. I also have a arduino mini on it's way so I can try Rob's method.

Thanks all for your input.

David

Print this page

To reply to this topic, you need to log in.