esher Posted October 24, 2013 Report Posted October 24, 2013 (edited) Hi all, i've got strange chinese pic programmer (G540) which supports 4685 chip at last. I can load hex and burn it, but sofware can't read chip configuration bits (or fuses) from hex (i doubt if they really exist in hex). The line of fuses now looks (in hex): 0007 1F1F 8285 3FC0 3FE0 3F40 Without proper fuses i can't really start testing new board because i dont know if the chip programmed correctly or not. Any help? edit: i found screenshot which shows me configuration bits but i have 2 bytes less than on my config windows: 0600 130F 8000 00B1 C03F E03F 403F edit2: ok, i found that configuration on the screenshot is working. Edited October 25, 2013 by esher Quote
esher Posted October 28, 2013 Author Report Posted October 28, 2013 After testing with these config bits (0006 0f13 0080 3fc0 3fe0 3f40), i had come to new strange thing. The board on which i testing the 4685 chip, old 452 chip working fine, detected by MIOS Studio, MIOS uploading and screen telling me "ready". Ok. But with 4685 i receive only "FF"s instead of "f0 00 00 7e 40 00 0f 00 00 f7". there it is: Any hint what's going on? I tried to change FOSC3-FOSC0 to HS w/o PLL, and in that case i catch on midi input another hex "C0 00 random hex" Quote
TK. Posted October 28, 2013 Report Posted October 28, 2013 It could be, that your programmer doesn't clear the ID bits - in this case, the wrong baudrate will be selected. There is a workaround which is documented here: http://www.ucapps.de/howto_debug_midi.html Search for "TEST SW2" Best Regards, Thorsten. Quote
esher Posted October 31, 2013 Author Report Posted October 31, 2013 It could be, that your programmer doesn't clear the ID bits - in this case, the wrong baudrate will be selected. There is a workaround which is documented here: http://www.ucapps.de/howto_debug_midi.html Search for "TEST SW2" Best Regards, Thorsten. Thanks a lot! After tens of tries to write this hex with G540, i did very simple 3-resistor serial programmer (jdm compatible), and then wrote bootloader with PICPgm. Very simple, fast, and with right ID's. Now all chips working fine. Quote
Marxon Posted May 15, 2016 Report Posted May 15, 2016 Hi, this topic is a bit older but i'm facing the same ID problem with my G540 programmer. There is also no option in software to reconfigure the programmer so that only the first 4k block will be overwritten. On 27.10.2013 at 8:21 PM, TK. said: It could be, that your programmer doesn't clear the ID bits - in this case, the wrong baudrate will be selected. Since the baud rate is set COM, would it not be possible to upload the workaround somehow via a LTC module? Thanks for your help. Best regards Marxon Quote
Marxon Posted May 18, 2016 Report Posted May 18, 2016 (edited) Problem solved! I found a tool called PICPgm which supports my old Willem parallel port programmer (the manufacturer software doesn´t support 18F4685) and burned my PIC without any problems. Have a nice day! Best regards Marxon Edited May 18, 2016 by Marxon Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.