Possible issue with some PIC18F2550 samples and a fix.

[dmitri]

I had a strange issue with one of two PIC18F2550 chips today. USB was working fine with one of the chips but not with another.
The only way to make this chip work was to pull pin 26 of PIC18F2550 to +5 over a 10k-47k resistor. Weird.

[dmitri]

If you happen to encounter a similar issue try this firmware:

pic18f2550usb_midi.hex

pin 26 should be connected over a 22k-47k resistor to ground instead.

[Xmattz]

Dimitri, stop talking in russian in english section of forum...

[dmitri]

This is the source:

PIC18F2550 USB.zip

[dmitri]

By now several people have had the same problem I had and all of them confirmed that this PIC18F2550 USB MIDI firmware fixes it, of course, with the required small schematic change (pin 26 should be connected over a 22k-47k resistor to the ground).

[r0bbie]

Hi Dmitri,

Finished my hardware (Synthex) yesterday, and gues what, USB problems.
I found this post this morning and added a 33k pulldown resistor to pin 26.
Problem solved, without updating firmware. should i worry?

Thanks

[dmitri]

If it works - it usually works. Just do a bit longer testing.

[NovaKaine]

Grrrrrr

[dmitri]

This file contains 5 PIC18F2550 USB MIDI firmware versions for 4/8/12/16/20MHz crystals:

PIC18F2550 USB_4_8_12_16_20MHz.zip

In theory, by using the same frequency firmware for both Atmega and PIC18F2550 it may be possible to use just one crystal instead of two.

[Ken Forgettable]

This file contains 5 PIC18F2550 USB MIDI firmware versions for 4/8/12/16/20MHz crystals:

XMEGA dev' - just with a change to CLOCK_FREQ in "HardwareProfile - pic18f2550.h" for these then?