Post by thread:[PIC]: Programming PIC16F876 via ICSP using PICSta

Dear PICListers, I am attempting to program a PIC16F876 soldered on a PCB with a PICStart+ and an adapter cable. I get serious programmer errors when attempting to program my soldered SO part. The adapter cable consists of an IC socket, some 5" wires soldered to some of the, ostensibly correct, pins and a 5 pin housing which connects to a matching header on my circuit board. The pins I have brought out to my programming header are as follows: RB7 Soldered to pin 28 of my IC socket adapter. RB6 Soldered to pin 27 of my IC socket adapter. MCLR Soldered to pin 1 of my IC socket adapter. Vss (System ground. This pin connects to both pin 8 and pin 19 of my SO package on the PCB) Soldered to pin 8 of my IC socket adapter. Vdd (System power) Soldered to pin 20 of my IC socket adapter. In other terms: ICSP Header IC Programming adapter RB7 ------------- Pin 28 RB6 ------------- Pin 27 MCLR ------------- Pin 1 Vss ------------- Pin 8 Vdd ------------- Pin 20 I don't have anything else soldered onto the PCB; Only the PIC and the programming header are on the circuit board currently. Therefore, there are no other parts to pull current away from the programmer. There are also no bias resistors present to affect any of the signal pins, programming or otherwise. The PICStart+ programmer programs PIC16F876 DIP parts when placed in the programmer itself. Additionally, I bent back pins on a DIP part corresponding to those pins I am not bringing out to my ICSP header on my PCB. Thus, these bent pins would not contact the programmer's ZIF socket contacts when the IC is placed in the programmer. I connected the two ground pins (Pin 19 and Pin 8) with a bit of wire and some soldering, as this more closely duplicated the connections I have on my PCB with the SO part. This DIP part was then placed into the programmer, and subsequently programmed successfully. I beeped out my programming adapter and my PCB. The entire system, from SO part on the PCB to the ZIF socket on the programmer, has continuity between the proper pins, point to point. I posted a 61K JPEG of my PCB, the IC programming adapter, and the modified DIP '876 that worked at http://dakott.dhs.org/mirror/ICSP_Problem01.jpg and a larger image (856K) of the same at http://dakott.dhs.org/mirror/ICSP_Problem02.jpg However, I get serious programmer errors when attempting to program my in situ SO package via my ICSP header and the IC programming adapter. The error report is as follows: -------------------- PICSTART Plus Error Log File 16-Aug-2000, 09:48:16 Device Type: 16F876 Errors during programming. Configuration bits not programmed. Program Memory Errors Address Good Bad 0000: 3000 0000 0001: 008A 0000 0002: 2818 0000 0004: 3021 0000 0005: 0084 0000 0006: 1383 000F 0007: 0800 0000 0008: 1903 0000 0009: 2817 0000 000A: 3006 0000 000B: 00F8 0000 000C: 01F7 0000 000D: 0BF7 0000 000E: 280D 0000 .... 001B: 0583 0000 001C: 309F 03FF 001D: 0084 0000 001E: 1383 0001 001F: 3007 0000 0020: 0080 0000 Any additional errors are not displayed. ---------- I had done this successfully before, with a 16F84. I prototyped the ICSP header and IC adapter before, on a breadboard. The programmer worked just fine then. Has anyone ever done anything like this before? Can you think of anything that I might be doing wrong? Can you suggest some things to try? Sincere thanks, in advance, David Kott -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

> > Dear PICListers, > > > > I am attempting to program a PIC16F876 soldered on a PCB with a > > PICStart+ and an adapter cable. I get serious programmer errors when > > attempting to program my soldered SO part. > > > Thanks to some "right on" personal replies, my problem has been mitigated. The issue appears to be that the PIC16F876's are prepped at the factory to be programmed via the LVP method. To defeat this behavior, the RB3 pin must be grounded when programming the device for the first time, if one is to use standard programming voltages of around 13 volts, or so. After the first successful programming, the RB3 pin need not be grounded to program the chips in system. My thanks to PICLister Andy for pointing this out, and solving my problem. :-) I would not have expected this behavior. Quoting from the "PIC16F8XX EEPROM Memory Programming Specification", section 2.3: "The program/verify mode is entered by holding pins RB6 and RB7 low while raising MCLR pin from VIL to VIHH (high voltage). In this mode, the state of the RB3 pin does not effect programming."(c) Clearly, however, it does depend on the state of RB3. Did they mean to say "affect" programming? Evidently, pin RB3 doesn't effect programming at all. It does quite the opposite, it precludes programming when programming using the "high voltage" programming method. -d (c) Microchip Technology Inc. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.