Post by thread:[PIC]: Calibration Data

Hi all, When I first started mucking around with PICs, I bought 6 x PIC12C509A/JW and 6 x PIC12C671/JW - which are not cheap devices. I have since found out that erasing these with a UV Eraser actually erases the OSCCAL data for the internal clock. Apparently, I should have read this data when the chips were new, recorded the number for each chip, then re-written the calibration data each time I write to that particular chip. This I did not do. :-( Is there any method of retreiving this data or are the chips useless? I assume that I can still use them with an extenal clock, but then that leaves less I/O. Regards, Sean -- http://www.piclist.com hint: To leave the PICList spam_OUTpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

Sean, They are not useless. But, it will be really painful to get the "correct" OSCCAL value (the closest to 4MHz). I have played around with 12C509, and the OSCAL value is around b'100000'. So, you can start your code like this: movlw b'100000' movwf OSCCAL That should have the clock pretty close to 4MHz. And if you need it to be as close as possible to 4MHz, write some procedures to change the OSCCAL value when you change the input pin for example. I never use 12C671 before, but I am sure they are similar. Hope that helps! ----------------------- Rudy Rudy .....rudyKILLspam@spam@rudyrudy.com http://www.rudyrudy.com ----------------------- {Quote hidden}> Hi all, > > When I first started mucking around with PICs, I bought 6 x PIC12C509A/JW > and 6 x PIC12C671/JW - which are not cheap devices. I have since found out > that erasing these with a UV Eraser actually erases the OSCCAL > data for the > internal clock. > > Apparently, I should have read this data when the chips were new, recorded > the number for each chip, then re-written the calibration data each time I > write to that particular chip. This I did not do. :-( > > Is there any method of retreiving this data or are the chips useless? I > assume that I can still use them with an extenal clock, but then > that leaves > less I/O. > > Regards, > > Sean > > -- -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestKILLspammitvma.mit.edu

>Is there any method of retreiving this data or are the chips useless? I >assume that I can still use them with an extenal clock, but then that leaves >less I/O. There is a way of recalibrating them, I think there is a Microchip App Note on how to, but I cannot be sure of this. I know there is information available somewhere as I saw it when looking at these devices a long time ago, but cannot remember exactly where I saw it. The chips are not useless without the calibration value. You can still use the internal oscillator, it is just that it will be off the nominal 4Mhz frequency by some value, which is what is normally corrected by using the OSCCAL function to bring it closer to the nominal 4Mhz. They would be fine for any use where the timing of the PIC is non-critical. -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam.....mitvma.mit.edu

The chips are not useless. They just won't run at the same speed as they did from the factory unless you put the same value back in for OSCCAL. You could do a series of "guesses" with calculations in between until you have the correct oscillator speed (or at least a speed close enough for your application). A quick check of my 12C509 data sheets shows that changing the OSCCAL value from one extream to the other moves the frequency about 250,000 Hz (at room temperature). Temperature and supply voltage can change the internal frequency even more than that of changing OSCCAL. If you really need a more accurate clock for the processor, use a crystal or resonator. If not, don't worry about OSCCAL too much. Your programs will still run. Sean Alcorn - Avion Sydney wrote: {Quote hidden}> Hi all, > > When I first started mucking around with PICs, I bought 6 x PIC12C509A/JW > and 6 x PIC12C671/JW - which are not cheap devices. I have since found out > that erasing these with a UV Eraser actually erases the OSCCAL data for the > internal clock. > > Apparently, I should have read this data when the chips were new, recorded > the number for each chip, then re-written the calibration data each time I > write to that particular chip. This I did not do. :-( > > Is there any method of retreiving this data or are the chips useless? I > assume that I can still use them with an extenal clock, but then that leaves > less I/O. > > Regards, > > Sean > > -- > http://www.piclist.com hint: To leave the PICList > EraseMEpiclist-unsubscribe-requestspam_OUTTakeThisOuTmitvma.mit.edu -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestspam_OUTmitvma.mit.edu

Hmmm. I just ordered some PIC16C771JWs to experiment with. Can someone give me a reference to how to avoid this OSCAL problem / capture the data and write it back when I want to erase and reuse these PICs? The http://www.microchips.com website returns zilch on a search for 'OSCAL'. Ted Mawson ====================================== Mark Newland wrote... The chips are not useless. They just won't run at the same speed as they did from the factory unless you put the same value back in for OSCCAL. You could do a series of "guesses" with calculations in between until you have the correct oscillator speed (or at least a speed close enough for your application). A quick check of my 12C509 data sheets shows that changing the OSCCAL value from one extream to the other moves the frequency about 250,000 Hz (at room temperature). Temperature and supply voltage can change the internal frequency even more than that of changing OSCCAL. If you really need a more accurate clock for the processor, use a crystal or resonator. If not, don't worry about OSCCAL too much. Your programs will still run. > Sean Alcorn - Avion Sydney wrote: > Hi all, > > When I first started mucking around with PICs, I bought 6 x PIC12C509A/JW > and 6 x PIC12C671/JW - which are not cheap devices. I have since found out > that erasing these with a UV Eraser actually erases the OSCCAL data for the > internal clock. > > Apparently, I should have read this data when the chips were new, recorded > the number for each chip, then re-written the calibration data each time I > write to that particular chip. This I did not do. :-( > > Is there any method of retreiving this data or are the chips useless? I > assume that I can still use them with an extenal clock, but then that leaves > less I/O. > > Regards, > > Sean > > -- > http://www.piclist.com hint: To leave the PICList > @spam@piclist-unsubscribe-requestKILLspammitvma.mit.edu -- http://www.piclist.com hint: To leave the PICList KILLspampiclist-unsubscribe-requestKILLspammitvma.mit.edu -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

on 17/1/02 2:50 AM, Ted Mawson at spamBeGoneTed.MawsonspamBeGonePORTFOLIOPM.COM wrote: > I just ordered some PIC16C771JWs to experiment with. Can someone give > me a reference to how to avoid this OSCAL problem / capture the data and > write it back when I want to erase and reuse these PICs? The > http://www.microchips.com website returns zilch on a search for 'OSCAL'. Ted, As far as I know, the OSCCAL (two Cs) is unique to the 12C, 12CE and 16C505 series chips - ie; the ones that have the internal oscillator. This value is written in to the last register in the factory, and the first instruction is to move this value into the W register. To use this value, your first instruction (for these parts) simply needs to be 'OSCCAL' - it is used to calibrate the Oscillator. I could be wrong, but I not think you have anything to worry about with 16C711JW parts as these do not have the internal oscillator and I am not aware of any other pre-programmed registers that can be erased with the UV. It will be interesting to see how Microchip will address this issue on the forthcoming 12F series. Regards, Sean -- http://www.piclist.com hint: To leave the PICList TakeThisOuTpiclist-unsubscribe-requestEraseMEspam_OUTmitvma.mit.edu

Thanks Sean, I spoke to MChip yesterday and they matched what you said, the 16C771s don't have an OSCCAL value and can be erased with no OSCCAL problems. The guy also told me that MChip are going completely to Flash programming so this will become a non-issue in time. Ted Mawson RemoveMETed.MawsonTakeThisOuTPortfolioPM.com ====================================== Sean Alcorn - Avion Sydney wrote... on 17/1/02 2:50 AM, Ted Mawson at Ted.MawsonEraseME.....PORTFOLIOPM.COM wrote: > I just ordered some PIC16C771JWs to experiment with. Can someone give > me a reference to how to avoid this OSCAL problem / capture the data and > write it back when I want to erase and reuse these PICs? The > http://www.microchips.com website returns zilch on a search for 'OSCAL'. Ted, As far as I know, the OSCCAL (two Cs) is unique to the 12C, 12CE and 16C505 series chips - ie; the ones that have the internal oscillator. This value is written in to the last register in the factory, and the first instruction is to move this value into the W register. To use this value, your first instruction (for these parts) simply needs to be 'OSCCAL' - it is used to calibrate the Oscillator. I could be wrong, but I not think you have anything to worry about with 16C711JW parts as these do not have the internal oscillator and I am not aware of any other pre-programmed registers that can be erased with the UV. It will be interesting to see how Microchip will address this issue on the forthcoming 12F series. Regards, Sean -- http://www.piclist.com hint: To leave the PICList EraseMEpiclist-unsubscribe-requestmitvma.mit.edu -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Hi Sean, try setting the calib byte to different settings and check the clock speed on your cro or freq meter. If you don't have a freq meter I think Jaycar sells a multimeter for about $100 AU that does freq and capacitance and maybe inductance too if I remember right. Do the testing in a binary fashion, ie start at 0xC0 to see if you're in the top or bottom half, then quarter towards the correct side etc. Hope that makes sense. :o) -Roman Sean Alcorn - Avion Sydney wrote: {Quote hidden}> > Hi all, > > When I first started mucking around with PICs, I bought 6 x PIC12C509A/JW > and 6 x PIC12C671/JW - which are not cheap devices. I have since found out > that erasing these with a UV Eraser actually erases the OSCCAL data for the > internal clock. > > Apparently, I should have read this data when the chips were new, recorded > the number for each chip, then re-written the calibration data each time I > write to that particular chip. This I did not do. :-( > > Is there any method of retreiving this data or are the chips useless? I > assume that I can still use them with an extenal clock, but then that leaves > less I/O. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

I have to wonder... I don't have time to look it up in the data sheet as I have 300+ PICList messages to read, plus over 2,500 error messages to deal with thanks to our fine friends and supporters at MIT. BUT... I wonder if you could use the internel OSC and get OSC or OSC/4 out from one of the pins with the right configuration... then hang a freq. counter or scope on that pin, figure out how far off you are and try different OSCCAL values until you were dead on. Investigation and experimentation left up to the reader... Dale -- "Curiosity is the very basis of education and if you tell me that curiosity killed the cat, I say only the cat died nobly." - Arnold Edinborough On Wed, 16 Jan 2002, Rudy Rudy wrote: {Quote hidden}> Sean, > > They are not useless. But, it will be really painful to get the "correct" > OSCCAL value (the closest to 4MHz). I have played around with 12C509, and > the OSCAL value is around b'100000'. So, you can start your code like this: > > movlw b'100000' > movwf OSCCAL > > That should have the clock pretty close to 4MHz. And if you need it to be > as close as possible to 4MHz, write some procedures to change the OSCCAL > value when you change the input pin for example. I never use 12C671 before, > but I am sure they are similar. > > Hope that helps! > > ----------------------- > Rudy Rudy > RemoveMErudyEraseMEEraseMErudyrudy.com > http://www.rudyrudy.com > ----------------------- > > > Hi all, > > > > When I first started mucking around with PICs, I bought 6 x PIC12C509A/JW > > and 6 x PIC12C671/JW - which are not cheap devices. I have since found out > > that erasing these with a UV Eraser actually erases the OSCCAL > > data for the > > internal clock. > > > > Apparently, I should have read this data when the chips were new, recorded > > the number for each chip, then re-written the calibration data each time I > > write to that particular chip. This I did not do. :-( > > > > Is there any method of retreiving this data or are the chips useless? I > > assume that I can still use them with an extenal clock, but then > > that leaves > > less I/O. > > > > Regards, > > > > Sean > > > > -- > > -- > http://www.piclist.com hint: To leave the PICList > RemoveMEpiclist-unsubscribe-requestspam_OUTKILLspammitvma.mit.edu > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

With the 12C509A chip, I don't think you can make the configuration bit so that you have OSC or OSC/4 out of one of the pins. What I would suggest is to write a simple square wave program using one of the pins. That way, you can calculate the # of clock cycles it should take, and match that with the frequency counter (or use the TIMER0 module, I never use it before, so can't say anything about that). square_wave bsf GPIO, 0 bcf GPIO, 0 goto square_wave You should get a not-so-square wave with frequency of OSC/16. Then, change the OSCCAL value until you get it dead on (or closest to 'dead on'). ----------------------- Rudy Rudy RemoveMErudyTakeThisOuTspamrudyrudy.com http://www.rudyrudy.com ----------------------- > {Original Message removed}

That's work too. Thought of that after I sent the last one. Dale -- "Curiosity is the very basis of education and if you tell me that curiosity killed the cat, I say only the cat died nobly." - Arnold Edinborough On Thu, 17 Jan 2002, Rudy Rudy wrote: {Quote hidden}> With the 12C509A chip, I don't think you can make the configuration bit so > that you have OSC or OSC/4 out of one of the pins. What I would suggest is > to write a simple square wave program using one of the pins. That way, you > can calculate the # of clock cycles it should take, and match that with the > frequency counter (or use the TIMER0 module, I never use it before, so can't > say anything about that). > > square_wave bsf GPIO, 0 > bcf GPIO, 0 > goto square_wave > > You should get a not-so-square wave with frequency of OSC/16. Then, change > the OSCCAL value until you get it dead on (or closest to 'dead on'). > > ----------------------- > Rudy Rudy > EraseMErudyspamspamBeGonerudyrudy.com > http://www.rudyrudy.com > ----------------------- > > > {Original Message removed}