Post by thread:Homebrew PIC Programmer (MCLR) question

Hi all, I've got a question about a PIC programmer I'm building. I'm putting together a 'Tait' style PIC programmer (see attachment). I've breadboarded it for testing, and I'm getting about 0.5v coming out of the collector of Q2 and going into MCLR when the pin 5 of the 7407 is high. When it's low I get 13.5v. My questions are: will this work to program 16F84s/16F87Xs? Will it harm to the PIC? I don't want to blow out the PIC chip (I'm a very low budget hobbiest), and after that, my other concern is that the 0.5v level is above to the upper limit for a logic zero on MCLR, and may not let the programmer function properly. The Electrical Specifications state that the max for a logic low on MCLR is 0.2v. Q1 which controls the VDD input to the PIC gives me 0v when pin 9 on the 7407 is high, and 5v when it's brought low, which appears to be correct. I've tried swapping transistors (PN2907s), but still get the same voltages on Q2 regardless of which transistor is used. I noticed this when I started breadboarding the circuit. The only components in place at that time were the voltage regulators, the 7407, and Q2 and its resistors. The addition of the rest of the circuit has made no difference. I've checked the resistor values several times both by color band and with an ohm meter. I've also checked all connections (about 10 times) to make sure that everything is wired correctly. I haven't found any problems. Note that I have **not** made any tests with a PIC in place. Any ideas? Thanks in advance, Mike. ________________________________________________________________ YOU'RE PAYING TOO MUCH FOR THE INTERNET! Juno now offers FREE Internet Access! Try it today - there's no risk! For your FREE software, visit: dl.http://www.juno.com/get/tagj.

Michael P Olson wrote: ... > together a 'Tait' style > PIC programmer (see attachment). I've breadboarded it for testing, Uhh, what attachment? You have a URL for us to look at? and > I'm getting about 0.5v > coming out of the collector of Q2 and going into MCLR when the pin 5 of > the 7407 is high. When > it's low I get 13.5v. My questions are: will this work to program > 16F84s/16F87Xs? Will it harm to the PIC? No, the '87X series uses 12V on /MCLR to engage 'high voltage programming mode. The chip draws no current for programming in this mode (at least according to the tech doc). The 12V is merely a switch to engage the mode. > I don't want to blow out the PIC chip (I'm a very low budget > hobbiest), and after that, my other > concern is that the 0.5v level is above to the upper limit for a logic > zero on MCLR, and may not let the > programmer function properly. The Electrical Specifications state that > the max for a logic low on > MCLR is 0.2v. So add a pulldown resistor. 1K should do. Everyone else seems to be making the Tait programmer work without it. I doubt that they are being as careful as you are. {Quote hidden}> Q1 which controls the VDD input to the PIC gives me 0v when pin 9 on the > 7407 is high, > and 5v when it's brought low, which appears to be correct. I've tried > swapping transistors (PN2907s), > but still get the same voltages on Q2 regardless of which transistor is > used. I noticed this when > I started breadboarding the circuit. The only components in place at that > time were the voltage > regulators, the 7407, and Q2 and its resistors. The addition of the rest > of the circuit has made no > difference. I've checked the resistor values several times both by color > band and with an ohm meter. > I've also checked all connections (about 10 times) to make sure that > everything is wired correctly. > I haven't found any problems. > > Note that I have **not** made any tests with a PIC in place. > > Any ideas? Be aware that the F87x series comes from the factory with LVP (Low Voltage Programming) turned on as the default. This means that you can't use RB3 for an I/O pin, and that you have to hold it low to prevent the chip from going into programming mode. This is a GOTCHA if you don't know about it. Also, what software are you using to program this device? The 'program' command is apparently different from the norm. 18h instead of 08. I'd REALLY like to know what software you're planning on using since I'm in the process of building one of the Tait clones myself, but since the existing software came out before the F87x series, I expect that it will not work. spam_OUTRobert.RolfTakeThisOuTUAlberta.ca

Robert, Sorry for the missing attachment. I forgot to attach it when I sent the post, and when I did try to post it, it was rejected for being too big. I think Juno might have turned it into text. The URL is http://bobblick.com/bob/projects/f84pgm/index.html. This is Bob Blick's 16F84 programmer. > > So add a pulldown resistor. 1K should do. > Everyone else seems to be making the Tait programmer work without > it. I doubt that they are being as careful as you are. > There's already a 10K pulldown on that pin. I guess if I don't exceed the 600ma rating of the PN2907, 1K will be O.K. I don't have a lot of formal training , so I try to avoid 'adlibbing' when it comes to circuit design. My concern here was that an excessively high (0.5v vs. 0.2v) input would keep the PIC from recognizing that the MCLR signal has gone back to Logic Zero - i.e. it would think that the pin is still 'high'. {Quote hidden}> > Also, what software are you using to program this device? > The 'program' command is apparently different from the norm. > 18h instead of 08. > I'd REALLY like to know what software you're planning on using since > I'm > in the process of building one of the Tait clones myself, but since > the > existing software came out before the F87x series, I expect that it > will > not work. > I'll be using his DOS software to do the programming. Programming the 16F876 is still a way down the road yet. Thanks again, Mike. ________________________________________________________________ YOU'RE PAYING TOO MUCH FOR THE INTERNET! Juno now offers FREE Internet Access! Try it today - there's no risk! For your FREE software, visit: dl.http://www.juno.com/get/tagj.

For what its worth, my programmer also has a .3 to .5 volt in similar places -- but only when there is nothing connected thereto. I am sure any loading would drop that voltage to very near zero. Mine seems to work OK that way. Michael P Olson wrote: {Quote hidden}> Robert, > > Sorry for the missing attachment. I forgot to attach it when I sent the > post, > and when I did try to post it, it was rejected for being too big. I think > Juno > might have turned it into text. > > The URL is http://bobblick.com/bob/projects/f84pgm/index.html. This is > Bob Blick's 16F84 programmer. > > > > > So add a pulldown resistor. 1K should do. > > Everyone else seems to be making the Tait programmer work without > > it. I doubt that they are being as careful as you are. > > > > There's already a 10K pulldown on that pin. I guess if I don't exceed > the 600ma rating of the PN2907, 1K will be O.K. I don't have a lot of > formal training , so I try to avoid 'adlibbing' when it comes to circuit > design. > > My concern here was that an excessively high (0.5v vs. 0.2v) > input would keep the PIC from recognizing that the MCLR signal > has gone back to Logic Zero - i.e. it would think that the pin is still > 'high'. > > > > > Also, what software are you using to program this device? > > The 'program' command is apparently different from the norm. > > 18h instead of 08. > > I'd REALLY like to know what software you're planning on using since > > I'm > > in the process of building one of the Tait clones myself, but since > > the > > existing software came out before the F87x series, I expect that it > > will > > > not work. > > > I'll be using his DOS software to do the programming. > Programming the 16F876 is still a way down the road yet. > > Thanks again, > > Mike. > > ________________________________________________________________ > YOU'RE PAYING TOO MUCH FOR THE INTERNET! > Juno now offers FREE Internet Access! > Try it today - there's no risk! For your FREE software, visit: > dl.http://www.juno.com/get/tagj.

> My concern here was that an excessively high (0.5v vs. 0.2v) > input would keep the PIC from recognizing that the MCLR signal > has gone back to Logic Zero - i.e. it would think that the pin is still > 'high'. Input low voltage for the /MCLR pin is .2Vdd, so for a 5V supply that gives you 1V, so your .5V is fine. And this is a full temp range spec, so unless you are programming your parts at 0C (or -40C if you live up here), you shouldn't have any problems. (threshold drops with temperature, just like a diode's). For most others port pins it's .15Vdd (.75V for 5V)so you still have some margin. I would suggest that anyone building the various simple programmers use SCHOTTKY diodes (.3V drop instead of .6) or germanium (if you can find them. 1N34a's are scarce IME). 1N5817 is a good cheap choice. (1A 20V Schottky, but anything in that range will do fine). .....Robert.RolfKILLspam@spam@UAlberta.ca

Robert, "Oh, I see!" said the blind (visually chanllenged?) man.... Guess I should have zoomed in a little further when I was displaying the PDF. I thought it was 0.2v rather than 0.2 x Vdd. I'm used to looking at 5v ttl logic levels. I'm also used to not using my reading glasses :^) I've got a bunch of germanium diodes left over from my daughter's crystal radio science fair project from a couple of years ago. I'll give them a try. The F84s are on their way via UPS ground, so I've got a few days to experiment. Thanks, Mike. On Sun, 23 Jan 2000 13:07:06 -0700 Robert Rolf <Robert.RolfKILLspamUALBERTA.CA> writes: {Quote hidden}> > My concern here was that an excessively high (0.5v vs. 0.2v) > > input would keep the PIC from recognizing that the MCLR signal > > has gone back to Logic Zero - i.e. it would think that the pin is > still > > 'high'. > > Input low voltage for the /MCLR pin is .2Vdd, so for a > 5V supply that gives you 1V, so your .5V is fine. And this is a full > temp range spec, so unless you are programming your parts at 0C > (or -40C if you live up here), you shouldn't have any problems. > (threshold drops with temperature, just like a diode's). > > For most others port pins it's .15Vdd (.75V for 5V)so you still have > some margin. > > I would suggest that anyone building the various simple > programmers use SCHOTTKY diodes (.3V drop instead of .6) > or germanium (if you can find them. 1N34a's are scarce IME). > 1N5817 is a good cheap choice. (1A 20V Schottky, but anything > in that range will do fine). > > .....Robert.RolfKILLspam.....UAlberta.ca ________________________________________________________________ YOU'RE PAYING TOO MUCH FOR THE INTERNET! Juno now offers FREE Internet Access! Try it today - there's no risk! For your FREE software, visit: dl.http://www.juno.com/get/tagj.

Hi Paul, could you tell me where did you read the internal pull-up of MCLR. At the moment I can not recall, and the uChip manuals seem not to support directly this statement (e. g. DS30390B, p. 126). Furthermore, they state: Input Leakage Current for MCLR is less, than +/- 5uA. Regards, Imre On Mon, 24 Jan 2000, Paul B. Webster VK2BZC wrote: {Quote hidden}> Marion D. Kitchens wrote: > > > For what its worth, my programmer also has a .3 to .5 volt in similar > > places -- but only when there is nothing connected thereto. I am sure > > any loading would drop that voltage to very near zero. > > No, the very opposite in fact. Recall that MCLR has an internal > pull-UP. It "loads" the programmer circuit in the *positive* direction. > > > Mine seems to work OK that way. > > That's the bottom line. > -- > Cheers, > Paul B. > >