Post by thread:PIC16C54-RC/P Programming

Does anyone know of a change recently to the way 16C54-RC/P accepts programming? I used to buy these all the time and set the oscillator bit to XT and never had a problem but the last batch of 100 is causing me some problems. I use parallax programming tools and the setup bits are in the file so unless I accidently changed it to LP I'm pretty sure I programmmed and verified them correctly. After programming the 100 pieces I even recall putting in one from an earlier run and reverified it. I shipped them to the assembler and he indicated that none of them seemed to work. Had them sent back to me and they show that the device is set for LP as an oscillator type. Before I give up and say it's operator error I thought I'd ask the group if you knew of any change in the way things operate. The way I figure it ... in order of most likely to have occurred 1. I goofed and hit the 'O' key and advance the Oscillator tyoe to LP 2. My Parallax programmer let me down 3. Microchip changed something and things don't work the same as they used to. Just thought I ask before I rebuy parts. Thanks Larry ----------------------------------------------------------------------- Larry P. Thomas wa0gwa voice : 1 913 888-0282 Krell Technologies fax : 1 913 782-9359 8960 Bond pager : 1 816 989-HELP Overland Park, KS 66214-1764 e-mail : spam_OUTlpthomasTakeThisOuTwinning-edge.com www : http://www.krell.com -----------------------------------------------------------------------

> Does anyone know of a change recently to the way 16C54-RC/P accepts > programming? I used to buy these all the time and set the oscillator bit to > XT and never had a problem but the last batch of 100 is causing me some > problems. > I shipped them to the > assembler and he indicated that none of them seemed to work. Had them sent > back to me and they show that the device is set for LP as an oscillator > type. Before I give up and say it's operator error I thought I'd ask the > group if you knew of any change in the way things operate. > > The way I figure it ... in order of most likely to have occurred > > 1. I goofed and hit the 'O' key and advance the Oscillator tyoe to LP > 2. My Parallax programmer let me down > 3. Microchip changed something and things don't work the same as they used to. Hmm... are you code protecting the parts? If so, have you always done so? I don't know what a 16C5X reads out of protected parts these days; my recollection is that that the status register reads "scrambled" but that the top 8 bits are normally all zero so that wouldn't matter. If Mchip added some feature to the new silicon which uses bit 4, that could cause a protect- ed chip to LOOK like it was set for LP mode. As for failure to operate, if you've confirmed that the clock is indeed not running, then it would appear that either: [1] The parts really have gotten mis-programmed to LC mode. [2] You got a batch of bad crystals/resonators (did you buy those in a batch with the PICs?) [3] "You get what you pay for": you bought chips which were not specified as having useable oscillator circuits (an RC OTP would only be tested with RC oscillator mode; while most will work if programmed for XT or LP mode, this is not guaranteed. In fact, it's possible (though IMHO not terribly likely) that Microchip found it had a wafer full of parts which passed all the tests EXCEPT the XT oscillator test and decided to package and sell this batch as RC parts. Without further information, I can't say whether #1, #2, or #3 is the cul- prit, but you may want to consider all those options.

>[1] The parts really have gotten mis-programmed to LC mode. We have a batch a PIC16C73A-04/SO parts which are programmed to XT, but only have LP drive and therefore don't work with our resonator circuit. Burning other chips (same tube) to HS resulted in almost-XT-level drive on the outputs, and therefore work. Anybody else have similar comments? Perhaps Microchip messed up? Andy ================================================================== Andy Kunz - Montana Design - 409 S 6th St - Phillipsburg, NJ 08865 Hardware & Software for Industry & R/C Hobbies "Go fast, turn right, and keep the wet side down!" ==================================================================

Larry; Used to be able to change the osc type on OTP parts. You might recall that A OSC of RC is the erased state of a JW type part. I ran into a similar situation some time ago and called a Microchip FAE he told me that on the chip is another programmable fuse that is now set after they qulify the part for OSC type. Turns out that the good-ol days of buying cheeper RC parts and then changeing them to LP, XT, HS etc are over.... David Jeffers OSP Technologies, Inc. ps: I was a RC Cheep-Skate....

According to the Microchip documentation (Microcontroller Data Book 1996/7), even in the protected mode the low nibble of the configuration word on 16C54 reads unscarmbled. The code protect feature scrambles the output in the following way: bits 4 to 11 are all zeros and the low nibble equals to XOR of all three nibbles. Since the bits 4-11 of the config word are all ones, they XOR to all zeros and do not change the low nibble. Someone called me recently with a similar problem, he had some 30 OTP PICs programmed with a buggy code. He is trying to erase these chips with a therapy X-ray machine. The diagnostic one turned out to be too weak. He promised to call me what the results are, so do not trash them yet. Even if everything fails, you may still find an artist and have your 54's arranged in some interesting sculpture ;-). Josef Hanzal .....euroclassKILLspam@spam@pha.pvtnet.cz

> According to the Microchip documentation (Microcontroller Data Book 1996/7), > even in the protected mode the low nibble of the configuration word on 16C54 > reads unscarmbled. The code protect feature scrambles the output in the > following way: bits 4 to 11 are all zeros and the low nibble equals to XOR > of all three nibbles. Since the bits 4-11 of the config word are all ones, > they XOR to all zeros and do not change the low nibble. On one PIC seminar the Microchip guy said that the code protection scheme may be changed at will and without notice by Microchip. Maybe they've done something like that. Niki steenkmpKILLspamfirga.sun.ac.za