View previous topic :: View next topic |
Author |
Message |
Emmo
Joined: 02 Feb 2017 Posts: 14
|
|
Posted: Wed Feb 08, 2017 11:33 am |
|
|
okay, got new chips with other version number (1624PT8) and other revision id today. They are not working, too. I'm getting crazy. If there is somebody who has got a PIC18F25K50-I/SO at home, please try out the code above.
I'm still confused why the pic is sending a stall command and the pic in DIP package not. Here you can see that:
I'm trying now to contact CCS Support. Hope they can help. |
|
|
Emmo
Joined: 02 Feb 2017 Posts: 14
|
|
Posted: Tue Feb 21, 2017 5:34 am |
|
|
finally I've got solved the problem!!!
It was really simple. Updated Compiler/IDE from 5.042 to 5.068 and now it is running very well. I've compared old with new hex file and the newer one is about 30 words longer. A lot have changed in the code.
Thanks for all your help! |
|
|
temtronic
Joined: 01 Jul 2010 Posts: 9247 Location: Greensville,Ontario
|
|
Posted: Tue Feb 21, 2017 6:22 am |
|
|
It'd be interesting to see what those 30 words of hex are and figure out WHAT the difference is between SMT and DIP packaged devices. I'd 'assume' the 'die' ( the actual PIC) was the same so that the package itself shouldn't matter..
It's nice your code/pkg is woking but really WHY ??
I know life is too short but.......
Jay |
|
|
Ttelmah
Joined: 11 Mar 2010 Posts: 19553
|
|
Posted: Wed Feb 22, 2017 4:48 am |
|
|
I think you will find it was the auto clock tuning. I remember there being a problem with this (had to turn it on manually), but thought it was resolved before then. However there is a reference at 5.046:
5.046 Support for auto clock tuning is added for chips that have this feature
Without this USB would be unreliable off the internal clock. Suggests that the DIP parts may have a more accurate clock than the SMD versions on the examples you have.... |
|
|
|