Post by thread:Using Picstart+ for production programming

part 0 806 bytes <META content=text/html;charset=iso-8859-1 http-equiv=Content-Type> <META content='"MSHTML 4.72.3510.1400"' name=GENERATOR> </HEAD> <BODY bgColor=#ffffff> <DIV><FONT size=2>I don't have a production programmer but would like to use the picstart+ to burn chips for a short production run (15).  Does anyone have experience as to how reliable the picstart+ will burn chips.  During several months of product development I have never had a problem with chips burned with the picstart+.  If I reburn the chips several times will this help? Can anyone recommend a good production programmer for doing short production runs (100) for a reasonable amount of money?</FONT></DIV></BODY></HTML> </x-html>

part 0 2947 bytes <META content=text/html;charset=iso-8859-1 http-equiv=Content-Type><!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN"> <META content='"MSHTML 4.72.3110.7"' name=GENERATOR> </HEAD> <BODY bgColor=#ffffff> <DIV> <BLOCKQUOTE style="BORDER-LEFT: #000000 solid 2px; MARGIN-LEFT: 5px; PADDING-LEFT: 5px"> <DIV><FONT face=Arial size=2><B>Assunto: </B>Using PicStart+ for production programming<BR><BR></DIV></FONT> <DIV><FONT size=2>I don't have a production programmer but would like to use the picstart+ to burn chips for a short production run (15).  Does anyone have experience as to how reliable the picstart+ will burn chips.  During several months of product development I have never had a problem with chips burned with the picstart+.  If I reburn the chips several times will this help? Can anyone recommend a good production programmer for doing short production runs (100) for a reasonable amount of money?</FONT></DIV></BLOCKQUOTE></DIV> <DIV><FONT color=#000000 size=2>I know one company had to remove 1000 PICs from returned</FONT></DIV> <DIV><FONT color=#000000 size=2></FONT><FONT size=2>board because of programming error by Picstart+. The boards</FONT></DIV> <DIV><FONT size=2>failed on the field.</FONT></DIV> <DIV><FONT size=2>I would suggest you ProPic 2 Gold. Take a look at</FONT></DIV> <DIV><FONT size=2><A href="http://www.propic2.com">www.propic2.com</A></FONT></DIV> <DIV><FONT size=2></FONT> </DIV> <DIV><FONT color=#000000 size=2>Octavio</FONT></DIV> <DIV><FONT color=#000000 size=2>==============================================<BR>Octavio Nogueira - <A href="spam_OUTnogueiraTakeThisOuTpropic2.com">.....nogueiraKILLspam@spam@propic2.com</A> ICQ# 19841898<BR>   From the creator of ProPic,   ProPic 2  now much better<BR>     New ProPic 2 homepage:    <A href="http://www.propic2.com">http://www.propic2.com</A>             <BR>    PIC Programmer for Windows with down to earth price <BR>==============================================</FONT></DIV> <BLOCKQUOTE style="BORDER-LEFT: #000000 solid 2px; MARGIN-LEFT: 5px; PADDING-LEFT: 5px"> <DIV> </DIV></BLOCKQUOTE></BODY></HTML> </x-html>

> Jeff Jolie wrote: > > I don't have a production programmer but would like to use the > picstart+ to burn chips for a short production run (15). Does anyone > have experience as to how reliable the picstart+ will burn chips. <snip> The Picstart programs and verifies at 5 volts. So operation at the margines isn't guaranteed, or even tested for. A production programmer programs and verifies at high and low line, as published in the data sheet. If you don't do that, then voltages that the PIC is spec'ed to run at might not work. Russell Hedges

My understanding on the major difference between a professional programmer versus a PICSTART Plus is the lack of verification check at +/-10% of 5V power supply to the PIC. Correct me if I am wrong. Why can't we just program the PIC with the PICSTART Plus and then test the PIC in the final product with the supply adjusted to 4.5 and 5.5 volt?? It may be time consumming though. If the PIC works at these voltages extremes, I would assume the PIC has been programmed propertly - regardless of the programmer used. This, if true and acceptable, will greatly help those on small production run cases. Sure hope somebody knowledgeable can clarify on this issue. Thanks, Bill Bass {Quote hidden}> > I don't have a production programmer but would like to use the > picstart+ to burn chips for a short production run (15). Does anyone > have experience as to how reliable the picstart+ will burn chips. > During several months of product development I have never had a > problem with chips burned with the picstart+. If I reburn the chips > several times will this help? Can anyone recommend a good production > programmer for doing short production runs (100) for a reasonable > amount of money? > > I know one company had to remove 1000 PICs from returned > > board because of programming error by Picstart+. The boards > > failed on the field. > > I would suggest you ProPic 2 Gold. Take a look at > > http://www.propic2.com > > > > Octavio > > _________________________________________________________ DO YOU YAHOO!? Get your free @yahoo.com address at http://mail.yahoo.com

>My understanding on the major difference between a professional >programmer versus a PICSTART Plus is the lack of verification check at >+/-10% of 5V power supply to the PIC. Correct me if I am wrong. > >Why can't we just program the PIC with the PICSTART Plus and then test >the PIC in the final product with the supply adjusted to 4.5 and 5.5 >volt?? It may be time consumming though. Sorry but you are wrong. It's not +/- 10% of 5V most of them need to be tested at 2.5 V and 6V. Octavio ============================================== Octavio Nogueira - nogueiraKILLspampropic2.com ICQ# 19841898 From the creator of ProPic, ProPic 2 now much better New ProPic 2 homepage: http://www.propic2.com PIC Programmer for Windows with down to earth price ==============================================

Hello, Octavio, Thanks for taking the time to answer. Can you kindly also tell me, and others, if I can test the programmed pic at 2.5V and 6V in the final product for verification check? Also, I assume in yours Propic, you are really reading back the program at the two extreme voltages. I would think running the program directly serve the same purpose. Again, thanks and appreciate your assistance. Bill Bass > >My understanding on the major difference between a professional > >programmer versus a PICSTART Plus is the lack of verification check at > >+/-10% of 5V power supply to the PIC. Correct me if I am wrong. > > > >Why can't we just program the PIC with the PICSTART Plus and then test > >the PIC in the final product with the supply adjusted to 4.5 and 5.5 > >volt?? It may be time consumming though. > Octavio wrote: > Sorry but you are wrong. It's not +/- 10% of 5V most of them need to be > tested at 2.5 V and 6V. > _________________________________________________________ DO YOU YAHOO!? Get your free @yahoo.com address at http://mail.yahoo.com

At 17:51 1998-11-18 -0800, you wrote: >> Jeff Jolie wrote: >> >> I don't have a production programmer but would like to use the >> picstart+ to burn chips for a short production run (15). Does anyone >> have experience as to how reliable the picstart+ will burn chips. > ><snip> > >The Picstart programs and verifies at 5 volts. So operation at the >margines isn't guaranteed, or even tested for. A production programmer >programs and verifies at high and low line, as published in the data >sheet. If you don't do that, then voltages that the PIC is spec'ed to >run at might not work. > >Russell Hedges Exactly what i«ve been thinking about today. There is a 7085 in the picstart +. I wonder if possible tu just exchange that for a variable one, and put a knom on thePicstart so i can set it to 5V and program the PIC, set it to 3V and verify, then set it to 6V and verify (or something like that) I don«t know how that will affext the rest of the electrinocs in PICstart however, like the PIC17, serial comm driver, or muxwe etc. Another alternative is to build a adapter with level-translator, but that would cause trouble if bidirectional lines. But as PIC hat TTL level input (?) and rail-rail output probablu just series resistors in th eused lines will do (except for VPP and GND), and we can vary the supply voltage ecternally, gated by the +5V Picstart think it is supplying. (just a fast buffering OP with variable gain to set it to give 2...7 V @ 5V signal in) But it is probably not much harder to build one of the programmers others sypply as kits etc. And for long series it is not fun to run the Picstart in three passes for each chip... But anyway; do somebody have a scematic on the Picstart internals? /Morgan Morgan Olsson ph +46(0)414 70741 MORGANS REGLERTEKNIK fax +46(0)414 70331 H€LLEKS (in A-Z letters: "HALLEKAS") SE-277 35 KIVIK, SWEDEN .....mrtKILLspam.....iname.com ___________________________________________________________

>Hello, Octavio, > >Thanks for taking the time to answer. > >Can you kindly also tell me, and others, if I can test the programmed >pic at 2.5V and 6V in the final product for verification check? >Also, I assume in yours Propic, you are really reading back the >program at the two extreme voltages. I would think running the >program directly serve the same purpose. > >Again, thanks and appreciate your assistance. > >Bill Bass Hi Bill and others, It's my pleasure to help. I think running the code at both extremes is not as efficient as with a production programmer. Also the programmer test if the chip is erased at vdd min to assure a good erasing margin. When you run the code, you could have some parts of it that are called only if an external event occur, like interrupts, etc. You have to be sure to run all parts of it. And yes, the ProPic 2 Gold programmer is a production programmer. Octavio ============================================== Octavio Nogueira - EraseMEnogueiraspam_OUTTakeThisOuTpropic2.com ICQ# 19841898 From the creator of ProPic, ProPic 2 now much better New ProPic 2 homepage: http://www.propic2.com PIC Programmer for Windows with down to earth price ==============================================

Why bother with picstart plus if there is promate2 made for that. Andre bill bass wrote: {Quote hidden}> My understanding on the major difference between a professional > programmer versus a PICSTART Plus is the lack of verification check at > +/-10% of 5V power supply to the PIC. Correct me if I am wrong. > > Why can't we just program the PIC with the PICSTART Plus and then test > the PIC in the final product with the supply adjusted to 4.5 and 5.5 > volt?? It may be time consumming though. > > If the PIC works at these voltages extremes, I would assume the PIC > has been programmed propertly - regardless of the programmer used. > This, if true and acceptable, will greatly help those on small > production run cases. > > Sure hope somebody knowledgeable can clarify on this issue. > > Thanks, > > Bill Bass

When you verify that your product works at the extremes of the supply voltage, can you then be sure that all the instructions read OK? I would not be so sure (full path coverage etc.). BTW there are also low-budget programmers who can verify at various supply voltages (for instance my home-brew 16x84 programmer). regards, Wouter. ---------- {Quote hidden}> From: bill bass <bass_82spam_OUTYAHOO.COM> > To: @spam@PICLISTKILLspamMITVMA.MIT.EDU > Subject: Re: Using Picstart+ for production programming > Date: Thursday, November 19, 1998 18:32 > > My understanding on the major difference between a professional > programmer versus a PICSTART Plus is the lack of verification check at > +/-10% of 5V power supply to the PIC. Correct me if I am wrong. > > Why can't we just program the PIC with the PICSTART Plus and then test > the PIC in the final product with the supply adjusted to 4.5 and 5.5 > volt?? It may be time consumming though. > > If the PIC works at these voltages extremes, I would assume the PIC > has been programmed propertly - regardless of the programmer used. > This, if true and acceptable, will greatly help those on small > production run cases. > > Sure hope somebody knowledgeable can clarify on this issue. > > Thanks, > > Bill Bass > > > > I don't have a production programmer but would like to use the > > picstart+ to burn chips for a short production run (15). Does anyone > > have experience as to how reliable the picstart+ will burn chips. > > During several months of product development I have never had a > > problem with chips burned with the picstart+. If I reburn the chips > > several times will this help? Can anyone recommend a good production > > programmer for doing short production runs (100) for a reasonable > > amount of money? > > > I know one company had to remove 1000 PICs from returned > > > board because of programming error by Picstart+. The boards > > > failed on the field. > > > I would suggest you ProPic 2 Gold. Take a look at > > > http://www.propic2.com > > > > > > Octavio > > > > > _________________________________________________________ > DO YOU YAHOO!? > Get your free @yahoo.com address at http://mail.yahoo.com

Andre Abelian wrote: > > Matt Bonner wrote: > > > Andre Abelian wrote: > > > > > > Why bother with picstart plus if there is promate2 made for that. > > > > > PicStart+: approx $99.00 US > > Promate 2: over $1000.00 IIRC > > No it is not over $ 1000 > > Check microchips web site $ 500 for Promate 2 and 150 for module > I guess that I did not recall correctly. When I bought mine, I also bought 2 modules. That would take it to over $1000 Canadian. I suppose I should have walked down the hall to the file room and looked it up. :-) --Matt

>Why can't we just program the PIC with the PICSTART Plus and then test >the PIC in the final product with the supply adjusted to 4.5 and 5.5 >volt?? It may be time consumming though. |Sorry but you are wrong. It's not +/- 10% of 5V most of them need to be |tested at 2.5 V and 6V. My interpretation of things is that the PIC needs to be verified above/below any voltage level at which it might be used. If a part is only programmed/verified at VDD=5.0 volts, that part can only be relied to run at VDD=5.0 volts. By contrast, if the part is verified at both 2.5 and 6.0 volts, then unless a smaller volt- age range is specified for the desired clock speed the part may safely be used at any voltage between 2.5 and 6.0 volts. Note that, in practice, lower VDD values will cause marginally- erased EPROM memory cells to read as programmed, and higher values will cause marginal cells to read as blank. Since all OTP devices are shipped from the factory thoroughly blank, it is very unlikely that an OTP device which programs/verifies at 5.0 volts would have programming-related problems at any lower voltage. On the other hand, it is not uncommon for EPROMS that haven't been baked long enough to read as blank at 5.00 volts but not at 2.5. If such a device is programmed in a 5-volt-only programmer and then used in a low-voltage application, the PIC may misbehave strangely and biz- arrely. As for the 16C84/16F84, I believe that the current devices massively over-erase and over-program their EEPROM, making VDD/programming problems unlikely. Nonetheless, if a 16x84 is not totally blank before programming, the only way to be sure that it programmed corr- ectly is to verify it at the lowest voltage at which it will be ex- pected to run.

Hello, John Thanks for your time to answer. Your info supplied is something I don't know. Question: (Pre-requisite) If I have a nice 5V regulator supplying the PIC's power and only use brand new 16F84s for programming, can count on and I still use the Picstart+ as a "production programmer"? I hope my understanding on what you say is correct. It also puzzled me a bit why would I test a pic at, say 3V , if the chip will never see 3V supply. Thanks again. Bill > >Why can't we just program the PIC with the PICSTART Plus and then test > >the PIC in the final product with the supply adjusted to 4.5 and 5.5 > >volt?? It may be time consumming though. > > |Sorry but you are wrong. It's not +/- 10% of 5V most of them need to be {Quote hidden}> |tested at 2.5 V and 6V. > > My interpretation of things is that the PIC needs to be verified > above/below any voltage level at which it might be used. If a > part is only programmed/verified at VDD=5.0 volts, that part can > only be relied to run at VDD=5.0 volts. By contrast, if the part > is verified at both 2.5 and 6.0 volts, then unless a smaller volt- > age range is specified for the desired clock speed the part may > safely be used at any voltage between 2.5 and 6.0 volts. > > Note that, in practice, lower VDD values will cause marginally- > erased EPROM memory cells to read as programmed, and higher values > will cause marginal cells to read as blank. Since all OTP devices > are shipped from the factory thoroughly blank, it is very unlikely > that an OTP device which programs/verifies at 5.0 volts would have > programming-related problems at any lower voltage. On the other > hand, it is not uncommon for EPROMS that haven't been baked long > enough to read as blank at 5.00 volts but not at 2.5. If such a > device is programmed in a 5-volt-only programmer and then used in > a low-voltage application, the PIC may misbehave strangely and biz- > arrely. > > As for the 16C84/16F84, I believe that the current devices massively > over-erase and over-program their EEPROM, making VDD/programming > problems unlikely. Nonetheless, if a 16x84 is not totally blank > before programming, the only way to be sure that it programmed corr- > ectly is to verify it at the lowest voltage at which it will be ex- > pected to run. > _________________________________________________________ DO YOU YAHOO!? Get your free @yahoo.com address at http://mail.yahoo.com

bill bass wrote: > > Hello, John > > Thanks for your time to answer. Your info supplied is something I > don't know. > > Question: (Pre-requisite) If I have a nice 5V regulator supplying the > PIC's power and only use brand new 16F84s for programming, can count > on and I still use the Picstart+ as a "production programmer"? > > I hope my understanding on what you say is correct. It also puzzled > me a bit why would I test a pic at, say 3V , if the chip will never > see 3V supply. > Testing the PIC at the extreme voltage supply ranges stresses the PIC, perhaps bringing out other flaws (for example, those caused by running it at the maximum rated temperature extreme - something that a production progammer can't check for). Also (especially with UV devices), the PIC might appear completely erased at 5V, but a few flakey zeros might appear at a voltage (or temperature) extreme. --Matt

Bill Bass wrote: > Question: (Pre-requisite) If I have a nice 5V regulator supplying the > PIC's power and only use brand new 16F84s for programming, can count > on and I still use the Picstart+ as a "production programmer"? Well, not *really*! > I hope my understanding on what you say is correct. It also puzzled > me a bit why would I test a pic at, say 3V , if the chip will never > see 3V supply. Further to other replies, there are a least *three* variables to consider; Vcc, Temperature, and *aging* (of both CPU and regulator, so they interact). Only one of these is *readily* adjustable by the programmer, so a large variation in this is used in the understanding that it will account for the other two. It may be however that a smaller range, such as 4.5 to 5.5V, or 4 to 6V, will account for the others if you really do intend to use the device only at 5V. This can be achieved by inserting diodes in series with the reference leg of the regulator (increases Vcc by 0.6V) and in series with the output (drops it by 0.6V) and arranging manual switching to short or open these diodes in sequence and do a final quick verify in all three conditions. -- Cheers, Paul B.

I don't see why a Picstart+ can't be used for production, as long as you are aware of what you are doing. First of all, you'd never use a PIC in something life-sustaining, no matter what programmer you used. The PIC is not intended for it(OK, some of the new ones that have parity and can read their own code, blah blah). The real question is, how many warranty or service returns are likely to occur? Is is going to cost you enough money to justify buying a production programmer? The 16F84 is probably the least likely to flake out for want of a production programmer, at least two reasons: 1) programming voltage internally generated. 2) timing not very critical when programming. A tiny one-person company making a few dozen widgets(there's that word!) may not find it worth the expense to buy a production programmer. Anything bigger than that, I'd say get a production programmer. Also, there are alternatives to the Picmaster. First, there are ones made by one-person companies(Octavio's ProPic2). If you are anxious enough about verification to consider a production programmer you probably feel anxious about buying one from somebody you can't sic your lawyers on. Second, there's the Needham's EMP-20. I've had one of those for years and it's been very reliable(it also does other micros, eeproms, plds, flash and eproms. A pretty good value($500usd). Don't buy it if you need to use the very newest pics, because they are always 6 months behind with the new personality modules($20 each). If you are doing a zillion PICs, buy a Picmaster. No question. Just my .02 -Bob

> A tiny one-person company making a few dozen widgets(there's that word!) > may not find it worth the expense to buy a production programmer. Anything > bigger than that, I'd say get a production programmer. My WISP programmer can do in-circuit production programming (4.0 - 6.0) for 16c84 and 16f84's when they are fed by a 7805 or similar. The design, fimware etc are available for free: http://www.xs4all/~wf/wouter/wisp I have seen even simpeler designs on the web which can do the same (though not in-circuit I think). I don't have the URLs at hand. But as said, the 16x84's are the least likely to be affected by VCC-related programming problems. regards, Wouter.

| Further to other replies, there are a least *three* variables to |consider; Vcc, Temperature, and *aging* (of both CPU and regulator, so |they interact). Only one of these is *readily* adjustable by the |programmer, so a large variation in this is used in the understanding |that it will account for the other two. Right, but if successful verification of a part at VDDmin (as opposed to something lower) is sufficient to assure operation at VDDmin under all temperature/aging variations, and if verification at VDDmax (as opposed to something higher) is sufficient to assure operation at VDDmax under all temperature/aging variations, is there any reason to expect that verification at VDDnom will not ensure correct operation at VDDnom?