Resistor bypass on PK3

Status
Not open for further replies.

large_ghostman

Well-Known Member
Most Helpful Member
I need to use a pk3, but it refuses to ID the chip correctly 8 times out of 10. I found a micro chip EN paper on bypassing two resistors on the PK3 to fix it, but cant find the doc now.

Anyone know which 2 resistors you put the wire link on to bypass? I always thought Expected ID xxx errors were the boards faults or things like that, turns out in some cases it not, apparently alot of pk clones dont have the resistors in and work better than the pk3...
 
Hi LG ...Suggest you post your setup / circuit, no device Id can be may things..
 
Hi LG ...Suggest you post your setup / circuit, no device Id can be may things..
Mainly its a power sag even on self powered boards! Really strange but the note TW posted below fixes the problem, many of the clones have the mod.
I think this is the one you're talking about...
Thx thats exactly the one!! Mainly affects usb ports on 64 bit machines or machines with more than 4 usb ports, the more stuff on the ports the worse the problem even when you power a board. I talked to MC about it and thought i had down loaded the sheet, but somehow it couldnt have downloaded. And if i see another $4.04 add every time the page is missing lol
 
Well I've recently had a PicKit4 (well two actually, one for work and one for home ), and that's far better at powering external boards than the PK3 was - which was my main reason for upgrading.
 
Well I've recently had a PicKit4 (well two actually, one for work and one for home ), and that's far better at powering external boards than the PK3 was - which was my main reason for upgrading.
There have been a few grumbles on the MC site over the PK4, but none of them connected to programming, i think it was more bout chip compatibility. Nearly all posts remark how it is actually built for modern USB ports and how much quicker it is and more reliable.

Toss up at some point between pk4 or ICD4, if i dont just give up pics all togther, 8 bit i got covered with some good modern 8051's and the 32bit energy micro's are very hard to beat at the price.
 
Makes no sence ? my 3 x PK3 never had this problem with my 64 bit mc. (AMD)
I think its got more to do with the 12V programming on the pk3 and certain usb systems, also depends how much you got hanging off the USB, i got a blue tooth keyboard and mouse and an external drive etc, if you hook a scope up you see it sag.

Apparently if you get unexpected you get 2 main ones, this is taking into account you got no board issues etc, you get 0X00 and then you get 0x(some numbers). If you get all zeros then its down to the LVP line and the pk3 isnt talking at all and if you get numbers it can talk but not program.

I will dig the email out and tell you which pins its seen on.

The fact they did an engineering note means there is an issue, same thing with why some win7 64 bit machines would brick the pk3 and some not.
 
Just as a point of note on USB powered devices - avoid the front panel connectors on PCs!

Most cases have long thin internal wiring between the motherboard and the case-mounted sockets, which cases a serious voltage droop at higher load currents.

The connectors at the back, directly on the motherboard, seem rock-solid by comparison.

Front panel USB3 ports seem much better than USB2 ones, as the cabling for USB3 tends to be manufactured to a rather higher standard.

A lot of cheap USB cables also have ludicrously thin internal cores and simply using a different, better quality cable can sometimes eliminate voltage problems.
 
I forget the front are cables, really valid point especially on the big tower systems i use. I am also getting problems on a old laptop, but i suspect thats a driver issue.
 
Roughly a PK3 takes 60mA when idle 70mA when active so how does 10mA cause a dip in the supply, however If the mcu has valid code , it will start to run it during the program phase, so power could be greater or cause timing issues , My standard ICSP connection has a 100R in the MCLR/vpp with a 10k pullup. a ferite on the USB cable helps, do not use the long wires supplied with some clones. don't have a connection to PK3 pin 6. if the pic is 'in circuit ' power externally ..

Edit Forgot to mention power supplied to all the Vdd pins with .1uF caps to gnd
 
Last edited:
Thats a valid point, i dont know but the usb port is already loaded.

I will post the microchip links on the subject and the replies etc, maybe we can figure it out. I am told the pk4 dosnt have the same issue. Also the brick problem is back on a win 8 machine with 64 bits. I need to get a pc or laptop set up to use this junk, i dont have the time to rewrite it all in 8051.

I cant get c18 working in mplab or mplab X, its been so long since i used it i have forgotten most the basics of setting it up.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…