Post by thread:[PIC] ICSP question

when programming the 16F876 using ICSP, should the RB3/PGM low voltage programming pin be grouned? i have programmed 3 of these chips, and forgot to ground this pin, and it appears that when i make it an output, it doesn't seem to work. have i permanently damaged the RB3 pin? -- Scott Larson 3201 S. State. St. box 2561 Chicago, Il, 60616 619-400-9775 http://goldscott.com

scott larson <goldscott <at> gmail.com> writes: > > when programming the 16F876 using ICSP, should the RB3/PGM low voltage > programming pin be grouned? > i have programmed 3 of these chips, and forgot to ground this pin, and it > appears that when i make it an output, it doesn't seem to work. > have i permanently damaged the RB3 pin? > > -- > Scott Larson > 3201 S. State. St. box 2561 > Chicago, Il, 60616 > 619-400-9775 > http://goldscott.com Scott, If you are programming at high voltage (traditional mehtod), RB3 has no effect, it can be left floating. Make sure to select high-voltage programming in config, this might be causing RB3 problem. In CCS PCW use: #FUSES NOLVP Sergey Dryga http://beaglerobotics.com

> when programming the 16F876 using ICSP, should the RB3/PGM low voltage > programming pin be grouned? > i have programmed 3 of these chips, and forgot to ground this pin, and it > appears that when i make it an output, it doesn't seem to work. > have i permanently damaged the RB3 pin? I have frequently programmed the 16F877 (which has the same datasheet, I believe) using ICSP while leaving RB3 floating, and have had no problems with this configuration. I did make sure LVP is off in the configuration, though. Greetings, Maarten Hofman.

ahh, yes, the default option on my programmer is LVP ON. after turning it off, everything works fine. thank you very much. On 11/10/05, Maarten Hofman <spam_OUTcashimorTakeThisOuTgmail.com> wrote: {Quote hidden}> > > when programming the 16F876 using ICSP, should the RB3/PGM low voltage > > programming pin be grouned? > > i have programmed 3 of these chips, and forgot to ground this pin, and > it > > appears that when i make it an output, it doesn't seem to work. > > have i permanently damaged the RB3 pin? > > I have frequently programmed the 16F877 (which has the same datasheet, > I believe) using ICSP while leaving RB3 floating, and have had no > problems with this configuration. I did make sure LVP is off in the > configuration, though. > > Greetings, > Maarten Hofman. > >

> when programming the 16F876 using ICSP, should the RB3/PGM low voltage > programming pin be grouned? For LVP is should obviously be high. For HVP some chips / date codes / phases of the mooon seem to require that it is held low. In both cases you can use a resistor to avoid problems. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

> I have frequently programmed the 16F877 (which has the same datasheet, > I believe) using ICSP while leaving RB3 floating, and have had no > problems with this configuration. I did make sure LVP is off in the > configuration, though. I have frequently cossed the street without looking left and right. I am still alive, so it must be perfectly safe. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

Maarten Hofman wrote: > I have frequently programmed the 16F877 (which has the same datasheet, > I believe) using ICSP while leaving RB3 floating, and have had no > problems with this configuration. I did make sure LVP is off in the > configuration, though. If you mean in the hexfile your want to transfer to the PIC: I would think this is irrelevant. To me it seems sensible only when the LVP was already cleared in the target config bits _before_ transferrring the hex file. Regards, Rob. -- Rob Hamerling, Vianen, NL phone +31-347-322822 homepage: http://www.robh.nl/

On 11/12/05, Wouter van Ooijen <.....wouterKILLspam@spam@voti.nl> wrote: > > I have frequently programmed the 16F877 (which has the same datasheet, > > I believe) using ICSP while leaving RB3 floating, and have had no > > problems with this configuration. I did make sure LVP is off in the > > configuration, though. > > I have frequently cossed the street without looking left and right. I am > still alive, so it must be perfectly safe. Because you're not living in the US. Then crossing the street must be precedeed by looking to the left, to the right and up. cheers, Vasile

> If you mean in the hexfile your want to transfer to the PIC: I would > think this is irrelevant. To me it seems sensible only when > the LVP was already cleared in the target config bits _before_ > transferrring the hex file. I have seen troubles with at least some PIC type/revision, I think with an early F877, even when the .hex states HVP and the PIC is configured for HVP. My guess is that is has to do with the fact that erasing the chip sets the config to LVP. Note that a fresh (blank) chip is configured for LVP. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

scott larson wrote: > when programming the 16F876 using ICSP, should the RB3/PGM low voltage > programming pin be grouned? You're not supposed to have to, and I think you can actually get away with it on that chip. It's still a good idea though. > i have programmed 3 of these chips, and forgot to ground this pin, and > it appears that when i make it an output, it doesn't seem to work. > have i permanently damaged the RB3 pin? By leaving it floating during programming? Very unlikely. There is probably something else going on. ****************************************************************** Embed Inc, Littleton Massachusetts, (978) 742-9014. #1 PIC consultant in 2004 program year. http://www.embedinc.com/products

> > I have frequently programmed the 16F877 (which has the same datasheet, > > I believe) using ICSP while leaving RB3 floating, and have had no > > problems with this configuration. I did make sure LVP is off in the > > configuration, though. > > I have frequently cossed the street without looking left and right. I am > still alive, so it must be perfectly safe. It is funny how you added the "so it must be perfectly safe" which I did not write (nor meant to imply). I merely wrote what I experienced so far. If I intended any implication, it was that the 16F877 did not get damaged in any way. This might not be a "perfect" statement, but from my knowledge of microcontroller design, it seems highly unlikely it would. Greetings, Maarten Hofman.

> It is funny how you added the "so it must be perfectly safe" which I > did not write (nor meant to imply). I merely wrote what I experienced > so far. If I intended any implication, it was that the 16F877 did not > get damaged in any way. This might not be a "perfect" statement, but > from my knowledge of microcontroller design, it seems highly unlikely > it would. Maybe I 'overreacted' a little, but the point is that observations where a safety measure (tying LVP low) were necesarry should take precedence over observations where it was not necessary. Ufortunatley the latter are often more numerous, so they get mentioned more often. This applies to LVP pull-down, but also to using TTL levels for RS-232 interfacing, omitting decoupler capacitors, direct bit-manipulation of port registers, etc etc. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

Wouter van Ooijen wrote: > I have seen troubles with at least some PIC type/revision, I think with > an early F877, even when the .hex states HVP and the PIC is configured > for HVP. My guess is that is has to do with the fact that erasing the > chip sets the config to LVP. Note that a fresh (blank) chip is > configured for LVP. I agree with Wouter. I've also seen problems doing HVP with the PGM pin floating, even though it's not supposed to matter according to the documentation. I sortof remember I bumped into this on a 16F628 a few years ago. I'm pretty sure Jim Robertson also reported something like this. ****************************************************************** Embed Inc, Littleton Massachusetts, (978) 742-9014. #1 PIC consultant in 2004 program year. http://www.embedinc.com/products

Olin Lathrop wrote: > Wouter van Ooijen wrote: > >> I have seen troubles with at least some PIC type/revision, I think with >> an early F877, even when the .hex states HVP and the PIC is configured >> for HVP. My guess is that is has to do with the fact that erasing the >> chip sets the config to LVP. Note that a fresh (blank) chip is >> configured for LVP. > > > I agree with Wouter. I've also seen problems doing HVP with the PGM pin > floating, even though it's not supposed to matter according to the > documentation. I sortof remember I bumped into this on a 16F628 a few > years > ago. I'm pretty sure Jim Robertson also reported something like this. > > > ****************************************************************** > Embed Inc, Littleton Massachusetts, (978) 742-9014. #1 PIC > consultant in 2004 program year. http://www.embedinc.com/products I am on my third design with F88 and LF88. To allow flawless, reliable programming on my client's DIY K128 in ICSP mode, the PGM pin has to be connected through a 10K (or smaller) resistor to GND. Driving a base pin of a transistor through a 3.3K worked just as well. Programming failure is ZERO. But I had to re-arrange the first design to fix the PGM business. No matter what MicroChip says, it is NOT reliable enough for primetime. --Bob -- Note: To protect our network, attachments must be sent to attachKILLspamengineer.cotse.net . 1-520-777-7606 USA/Canada http://beam.to/azengineer