Post by thread:[PIC:] safely driving a pic input from a PC Parall

Exact match. Not showing close matches.
PICList Thread
'[PIC:] safely driving a pic input from a PC Parall'
2003\10\24@174008 by James Newton, webhost

I've been thinking about how things get connected between PC's and PIC's lately and overall what I can see is: USB: Nice! But not the lowest possible cost and not supported on some old PC's Also, requires some driver development on the PC side... not hard, but still. RS232: Nice... But you can run out of ports, some current laptops don't even have serial ports (!) and you have to do all the darn configuration on the PC to make sure all the settings are right. Has anyone been able to get a USB to RS232 adapter to work right? I can never get the PC application to see the port. Parallel: Sigh... NT and up driver problems (IF you don't simulate a printer with the PIC) but there are solutions for that. One the one port usually and I've seen laptops without parallel ports. USB to parallel adapters are really shaky... has anyone used one to talk to anything other than a printer? The two I've tried don't show up as LPT ports on the PC and so old applications don't see them. Of all the above, probably the most reliable, simple, and cheapest for us little guys is the parallel port (this is IMHO and I'm sure the point can be argued 20 ways from Sunday). So, if I want to put out something that people can setup from a PC, a temporary connection via the parallel port is probably best. Now the question is how do I keep people from bread boarding something that connects via my device that will blow up the PC (as well as me probably) when they screw up? I went on a hunt for parallel port isolation and, not being the most brilliant analog electronics engineer (ok, so I'm a complete moron in that area) I really don't seem to understand most of what I found. I understand that the port can sink 24mA and only source 2.6mA so obviously its a good idea to drive your loads from the port, through the load, then to +5v rather than ground. Safe current limiting can be had with a series resistor of 1.6k or more right? (4.2V / 2.6mA) But that doesn't leave much drive... so you can add a driver, but wouldn't it be better to prevent the pin from sourcing anything by putting a diode long with the resistor in series and then reducing the resistor to 175 or more ohms (4.2V / 24mA)? +5 | / \ / 220R 1N4148 \ >-o---,/\/\/'----|<-----o-----=> _|_/ //_\ 5.1V | GND A pull-up resistor is added to the output to drive TTL or other loads and a Zener could be added to ground to prevent over voltage or driving the pin in reverse. That seems to me to be a nice little, very low cost, easy to work with, circuit to provide some basic protection for kids or students to tack things onto. Not wanting to assume I have a clue, I searched the internet for other solutions. Tomi Engdahl says: "The following circuit uses two 1N4148 diodes to protect parallel port against higher than +5V signals and also against wrong polarity signals (power on the circuit is accidentally at wrong polarity. " Diode 1N4148 4.7K parallel >-|>|-+--\/\/\/--etc... port data | pin +-|<|-+ 1N4148 | parallel >-----------+ port ground | Ground "Adding even more safety idea: Replace the 1N4148 diode connected to ground with 5.1V Zener diode. That diode will then protect against over-voltage spikes and negative voltage at the same time. " Ok, he has his series diode in the exact opposite direction. Could somebody please tell me if I'm just being stupid? Other than that (and the much larger series resistor) the circuit is the same. I also found on piclist.com: "Another way to protect against over-voltage is to put a regular diode with the cathode on the signal line and the anode on the power supply line. Keep in mind that this will cause the power supply to receive the over-voltage and may result in damage to other components on the board." Standard diodes are cheaper... any reason why I can't put ONE Zener in the power supply and use regular diodes on the 8 pins? x8: +5 | +5 / _|_ \ _\ /_ / | 220R 1N4148 \ >-o---,/\/\/'----|<-----o-----=> x1: +5 _|_/ //_\ 5.1V | GND --- James Newton, webhost piclist.com (former Admin #3) spam_OUTjamesnewtonTakeThisOuTpiclist.com 1-619-652-0593 fax:1-208-279-8767 PIC/PICList FAQ: http://www.piclist.com or .org -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2003\10\24@174426 by Josh Koffman

Hi James. I don't have an actual example for you, but in "The Robot Builder's Bonanza" by Gord McComb, he had a parallel port buffer for the hobbiest so you wouldn't fry your computer. My copy is unfortunately in another city, so I can't scan it for you. You could try your local library though. Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams "James Newton, webhost" wrote: <big snip> -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2003\10\24@192313 by Bob Blick

Hi James, My own experience with PC parallel ports has been that they are relatively tough, and you can communicate to and from a PIC pretty much one-to-one. The real damage comes from ground voltage differences. If your PC is not grounded (like if the third prong is missing from the equation somewhere) then the PC will be 60 volts AC away from ground, and that is where there can be some issues. People hooking their PC into their TV entertainment system do sometimes blow up their sound cards. However, since people get away with hooking printers and PIC programmers up to their PCs, day in and day out, with hardly ever a problem, I'd suggest doing the least possible. Like just hooking them together. Or using a series resistor, and optional zener + diode. How much this adds to the cost of your device should be a factor. Of course, there is EMI compliance. Do you need to look at that? Because filters and protection can be combined. There are various standard-setting agencies each with multiple levels of test limits and you could build to suit the one you choose. Asking if something is protected enough is hard, because, like doctors and lawyers, the "safe" answer is "not 100 percent safe". Cheerful regards, Bob -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2003\10\24@204202 by James Newton, webhost

source= http://www.piclist.com/piclist/2003/10/24/192313a.txt? Bob Blick says: "The real damage comes from ground voltage differences. If your PC is not grounded (like if the third prong is missing from the equation somewhere) then the PC will be 60 volts AC away from ground, and that is where there can be some issues." Any suggestions on protecting against that sort of thing? Fuse the ground connection between the PC and device? "...Of course, there is EMI compliance. Do you need to look at that?..." No, thank goodness. Just looking (as you suggest) to do the minimum and hope to stop 90% of the standard junk that happens. Thanks for the feedback... I'd actually just really like to know why I have the series diode going one way and Tomi has it going the other way... http://www.piclist.com/piclist/2003/10/24/174008a --- James Newton: PICList.com webmaster, former Admin #3 .....jamesnewtonKILLspam@spam@piclist.com 1-619-652-0593 phone http://www.piclist.com/member/JMN-EFP-786 PIC/PICList FAQ: http://www.piclist.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2003\10\25@054054 by Rob Hamerling

Hi James, James Newton, webhost wrote: > So, if I want to put out something that people can setup from a PC, a > temporary connection via the parallel port is probably best. Now the > question is how do I keep people from bread boarding something that connects > via my device that will blow up the PC (as well as me probably) when they > screw up? > > I went on a hunt for parallel port isolation and, not being the most > brilliant analog electronics engineer (ok, so I'm a complete moron in that > area) I really don't seem to understand most of what I found. I do control some modelrailroad devices via the parallel port of my PC, and use optocouplers to protect my PC. See for schematics: http://www.robh.nl/art/s88lptw.gif. It might be a beginning. Regards, Rob. -- Rob Hamerling, Vianen, NL phone +31-347-322822 homepage: http://www.robh.nl/ -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@090449 by Dennis Crawley

James, how about optocoupling the port?. I think It can be done with: 1.- an extra cable from the PC to assure 5v from supply . 2.- buffers like 7404, some resistors (cable balance). 3.- 6n317 opto, VERY fast! 4.- A power supply on the project side. 5.- and I think an extra logic to deal with bi-directionality it will be needed. With this structure not only the port is protected, but also the ground plane is isolated. I did something similar for RS232 And it works ok! connecting a PIC with a PC with out Max232. Dennis. ----- Original Message ----- From: "James Newton, webhost" Sent: Friday, October 24, 2003 9:38 PM Subject: Re: [PIC:] safely driving a pic input from a PC Parallel port. > > Any suggestions on protecting against that sort of thing? Fuse the ground > connection between the PC and device? -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@121643 by Bob Axtell

James, it is NOT that hard to interface safely to PC's. Data TO THE PC: The PC expects either OPEN COLLECTOR inputs or tri-state drivers of the Schotty LS (74LSXX) variety, which is extremely durable. The PIC MAY drive it properly, but is not as durable. I recommend a TTL driver chip. Data FROM The PC: The PC sends signals of 0.6V max (low) and 5.0V max (high) but in some cases the HIGH output might be as low as 3.5V. So it needs a TTL interface as well. Overall, I suggest a I/O chip like the 74LS245 which can handle both directions of the databus. If the databus is NOT being used bidirectionally, the 74LS244 can act as an interface chip; all input signals of the LS244 are Schmitt-triggered. Small current limiting resistors, such as 33ohm, between signals of each systems further increases overall reliability. -- To handle NT/XP/Win2K issues, I use Winsoft.Sk IO drivers (C++/Delphi); they have a very good one for parallel ports that make interfacing Win32 I/O like the good 'ole days. http://www.winsoft.sk . Erik is an exceptional programmer and makes a product that REALLY works; I've used IOPORT and COMPORT sucessfully for three years now. --Bob At 02:37 PM 10/24/2003, you wrote: {Quote hidden}>I've been thinking about how things get connected between PC's and PIC's >lately and overall what I can see is: >USB: Nice! But not the lowest possible cost and not supported on some old >PC's Also, requires some driver development on the PC side... not hard, but >still. > >RS232: Nice... But you can run out of ports, some current laptops don't even >have serial ports (!) and you have to do all the darn configuration on the >PC to make sure all the settings are right. Has anyone been able to get a >USB to RS232 adapter to work right? I can never get the PC application to >see the port. > >Parallel: Sigh... NT and up driver problems (IF you don't simulate a printer >with the PIC) but there are solutions for that. One the one port usually and >I've seen laptops without parallel ports. USB to parallel adapters are >really shaky... has anyone used one to talk to anything other than a >printer? The two I've tried don't show up as LPT ports on the PC and so old >applications don't see them. > >Of all the above, probably the most reliable, simple, and cheapest for us >little guys is the parallel port (this is IMHO and I'm sure the point can be >argued 20 ways from Sunday). > >So, if I want to put out something that people can setup from a PC, a >temporary connection via the parallel port is probably best. Now the >question is how do I keep people from bread boarding something that connects >via my device that will blow up the PC (as well as me probably) when they >screw up? > >I went on a hunt for parallel port isolation and, not being the most >brilliant analog electronics engineer (ok, so I'm a complete moron in that >area) I really don't seem to understand most of what I found. > >I understand that the port can sink 24mA and only source 2.6mA so obviously >its a good idea to drive your loads from the port, through the load, then to >+5v rather than ground. > >Safe current limiting can be had with a series resistor of 1.6k or more >right? (4.2V / 2.6mA) But that doesn't leave much drive... so you can add a >driver, but wouldn't it be better to prevent the pin from sourcing anything >by putting a diode long with the resistor in series and then reducing the >resistor to 175 or more ohms (4.2V / 24mA)? > > > +5 > | > / > \ > / > 220R 1N4148 \ > >-o---,/\/\/'----|<-----o-----=> > _|_/ >//_\ 5.1V > | > GND > >A pull-up resistor is added to the output to drive TTL or other loads and a >Zener could be added to ground to prevent over voltage or driving the pin in >reverse. That seems to me to be a nice little, very low cost, easy to work >with, circuit to provide some basic protection for kids or students to tack >things onto. > >Not wanting to assume I have a clue, I searched the internet for other >solutions. Tomi Engdahl says: > >"The following circuit uses two 1N4148 diodes to protect parallel port >against higher than +5V signals and also against wrong polarity signals >(power on the circuit is accidentally at wrong polarity. " > > Diode > 1N4148 4.7K >parallel >-|>|-+--\/\/\/--etc... >port data | >pin +-|<|-+ > 1N4148 | >parallel >-----------+ >port ground | > Ground > >"Adding even more safety idea: Replace the 1N4148 diode connected to ground >with 5.1V Zener diode. That diode will then protect against over-voltage >spikes and negative voltage at the same time. " > >Ok, he has his series diode in the exact opposite direction. Could somebody >please tell me if I'm just being stupid? Other than that (and the much >larger series resistor) the circuit is the same. > >I also found on piclist.com: > >"Another way to protect against over-voltage is to put a regular diode with >the cathode on the signal line and the anode on the power supply line. Keep >in mind that this will cause the power supply to receive the over-voltage >and may result in damage to other components on the board." > >Standard diodes are cheaper... any reason why I can't put ONE Zener in the >power supply and use regular diodes on the 8 pins? > >x8: > +5 > | > +5 / > _|_ \ >_\ /_ / > | 220R 1N4148 \ > >-o---,/\/\/'----|<-----o-----=> > >x1: > > +5 > _|_/ >//_\ 5.1V > | > GND > >--- >James Newton, webhost piclist.com (former Admin #3) >jamesnewtonKILLspampiclist.com 1-619-652-0593 fax:1-208-279-8767 >PIC/PICList FAQ: http://www.piclist.com or .org > >-- >http://www.piclist.com hint: PICList Posts must start with ONE topic: >[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads -------------- Bob Axtell PIC Hardware & Firmware Dev Tucson, AZ 1-520-219-2363 "I lose some on each sale but make it up in volume." -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@150924 by James Newton, webhost

source= http://www.piclist.com/piclist/2003/10/25/054054a.txt? Rob Hamerling says:I do control some modelrailroad devices via the parallel port of my PC, and use optocouplers to protect my PC. See for schematics: http://www.robh.nl/art/s88lptw.gif. Rob, what optocoupler did you use? What part number I mean. --- James Newton: PICList.com webmaster, former Admin #3 .....jamesnewtonKILLspam.....piclist.com 1-619-652-0593 phone http://www.piclist.com/member/JMN-EFP-786 PIC/PICList FAQ: http://www.piclist.com -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@151723 by James Newton, webhost

source= http://www.piclist.com/piclist/2003/10/25/090449a.txt? Dennis Crawley says: "3.- 6n317 opto, VERY fast!" Are you sure of that part number? I can't find it anywhere. http://www.findchips.com/avail?part=6n317 --- James Newton: PICList.com webmaster, former Admin #3 EraseMEjamesnewtonspam_OUTTakeThisOuTpiclist.com 1-619-652-0593 phone http://www.piclist.com/member/JMN-EFP-786 PIC/PICList FAQ: http://www.piclist.com -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@152754 by James Newton, webhost

source= http://www.piclist.com/piclist/2003/10/25/121643a.txt? Bob Axtell says: "Overall, I suggest a I/O chip like the 74LS245 which can handle both directions of the databus. If the databus is NOT being used bidirectionally, the 74LS244 can act as an interface chip; all input signals of the LS244 are Schmitt-triggered. Small current limiting resistors, such as 33ohm, between signals of each systems further increases overall reliability." Based on the pricing, that might just be the best thing... 48 cents in some packages quantity one from digikey. But I don't think that would protect against the "AC between grounds" problem that was mentioned. Optoisolators sound best, but so far the prices are killer. The 244 may just be the "sweet spot." But, I STILL don't understand why my diode ended up in the opposite direction... Tomi: {Quote hidden}> Diode > 1N4148 4.7K >parallel >-|>|-+--\/\/\/--etc... >port data | >pin +-|<|-+ > 1N4148 | >parallel >-----------+ >port ground | > Ground > Me: {Quote hidden}> +5 > | > / > \ > / > 220R 1N4148 \ > >-o---,/\/\/'----|<-----o-----=> > _|_/ > //_\ 5.1V > | > GND Huh? huh!? Why? WHY!? WHY AM I BACKWARDS????? Err... maybe I shouldn't put it exactly that way... --- James Newton: PICList.com webmaster, former Admin #3 jamesnewtonspam_OUTpiclist.com 1-619-652-0593 phone http://www.piclist.com/member/JMN-EFP-786 PIC/PICList FAQ: http://www.piclist.com -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@153008 by Ishaan Dalal

Some optocouplers I have seen are quirky in that they have open collector outputs, which may make things tough if you are going to be sourcing any current; they also often use bipolars at the output, giving the usual source/sink asymmetries. I prefer solid-state optical relays instead, that are "true" switches just like standard relays, and can handle ~50-250 mA in any direction. Check out the excellent HSR312/HSR412 single SSRs from Fairchild semi, I've been very happy with them. NAIS/Aromat also make dual SSRs (8-pin packages) which I have successfully used too. Check them out at http://www.aromat.com/photomos.htm ; I'm not sure, but they might make triple or quad relays in 14-pin pkgs too. The NAIS relays, I remember, had rise/fall times of <~1 uS or so. If you are looking for mad fast comms, check out the 10-mbps optocouplers from Agilent or Fairchild - it's xx-2631, a look through the list on Fairchild Semi's opto page should give you the xx - these have rise/fall times of <~100 ns. Cheers, Ishaan {Original Message removed}

2003\10\25@153624 by Ishaan Dalal

I think he meant the 6N137 (aka HCPL-2630/2631), 10 mbps optocouplers. Note that these have open-collector outputs. Cheers, Ishaan {Original Message removed}

2003\10\25@154507 by Ishaan Dalal

From: "James Newton, webhost" <@spam@jamesnewtonKILLspamPICLIST.COM> > Based on the pricing, that might just be the best thing... 48 cents in some > packages quantity one from digikey. Well, any of the fancy SSRs I mentioned is going to set you back at least $1.25 or so per SSR in small quantities (i.e., a 4-pack would be ~$5). You might want to try the good old 4N29 oc's (darlington outputs), 6-pin DIP, $0.14 small qtys from Arrow. Or, I just found the Fairchild H11A617As, 4-pin DIP, single phototransistor outputs, $0.16 small qtys from Arrow. Cheers, Ishaan -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@160154 by Rob Hamerling

James Newton, webhost wrote: > source= http://www.piclist.com/piclist/2003/10/25/054054a.txt? > > Rob Hamerling says:I do control some modelrailroad devices via the parallel > port of my PC, and use optocouplers to protect my PC. See for schematics: > http://www.robh.nl/art/s88lptw.gif. > > Rob, what optocoupler did you use? What part number I mean. James, Basically the PC817, but since I needed 4 of these, I took a PC847. In my case it isn't very critical, since data transfer is not particularly high speed (approx. 20 KHz). You can find more info (software) by browsing to http://www.robh.nl/mklsoft.htm#S88LPT Regards, Rob. -- Rob Hamerling, Vianen, NL phone +31-347-322822 homepage: http://www.robh.nl/ -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@160404 by Bob Blick

On 25 Oct 2003 at 12:27, James Newton, webhost wrote: > Optoisolators sound best, but so far the prices are killer. The 244 may just > be the "sweet spot." Hi James, The only problem with the '244 or any other interface chip is that it doesn't protect your PC parallel port during the plugging and unplugging process. Only the optoisolators will protect you then. On the other hand, as long as the shell of your connector is wired through to ground, you are pretty safe against ground differences because the shell contacts first and breaks last. Which brings me back to the series resistor by itself, or a series resistor followed by a zener to ground and diode to supply if you want to get fancy. Oh, another thing. PICs are pretty resistant to latchup, more so than many logic families you might be tempted to use for an interface. Although you are safe with LS, HC is highly prone to latchup. Cheerful regards, Bob -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\25@172035 by William Chops Westfield

> On 25 Oct 2003 at 12:27, James Newton, webhost wrote: >> Optoisolators sound best, but so far the prices are killer. The 244 >> may just be the "sweet spot." I'm beginning to see why lots of people don't like using the parallel port. The "PC compatible" parallel port is something of a crock, and propagating that crock into big multi-io chips in irreplaceable spots on expensive motherboards was silly. :-( Serial IO at least has a bit of built-in protection in the form of the drivers (those are still irreplaceable, but more robust.) USB has the big advantage that it was designed well after the problems of putting connectable electronics in the hands of consumers were better understood. USB is wonderful stuff, I think. Add the fact that parallel ports are also disappearing from modern PCS, and the difficulty in accessing the ports from user programs in modern operating systems, and "anything else" starts to look pretty attractive as your protection circuitry costs start to go up. (or - the easiest way to protect a PC's parallel port is to put it at the end of a USB cable. Those run about $25. Does anyone happen to know of (or whether) any of these are usable in the "flexible" fashion of a normal parallel port (being able to set/read individual signals, etc.) Documented?) BillW -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

2003\10\27@055700 by Alan B. Pearce

>Dennis Crawley says: "3.- 6n317 opto, VERY fast!" I think he means the 6N137, which is a Darlington transistor output. However I prefer the HCPL2300 which has an internal opto receiver, and comes out as an open collector digital signal, with a 1k pull up resistor available for connection. Requires 5V supply for the receiver side, and is a heap faster than the 6N137. Both devices are Agilent ones. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email KILLspamlistservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

2003\10\27@085445 by Jim Monteith

I talk to PICs daily through the parallel port on windows 2000. For a manufacturing process, I created a program that writes the code to a PIC, and verifies it, also read and write from/to EEprom. The two keys here are using a drive program like OpenPort which gives you direct access to the parallel port, and then how you write to the port. You have to write to the port address which is usually 0x0378. Beyond this, it depends on what you want to do. You really need to study the pinout of the parallel port, and decide which pins you want to use. It's not that hard to do, and it is a very stable way to talk to the PIC. Jim {Original Message removed}

2003\10\27@100641 by George Sacco

Jim, where can I get the program OpenPort Thanks george -----Original Message----- From: pic microcontroller discussion list [RemoveMEPICLISTTakeThisOuTMITVMA.MIT.EDU] On Behalf Of Jim Monteith Sent: Monday, October 27, 2003 8:52 AM To: spamBeGonePICLISTspamBeGoneMITVMA.MIT.EDU Subject: Re: [PIC:] safely driving a pic input from a PC Parallel port. I talk to PICs daily through the parallel port on windows 2000. For a manufacturing process, I created a program that writes the code to a PIC, and verifies it, also read and write from/to EEprom. The two keys here are using a drive program like OpenPort which gives you direct access to the parallel port, and then how you write to the port. You have to write to the port address which is usually 0x0378. Beyond this, it depends on what you want to do. You really need to study the pinout of the parallel port, and decide which pins you want to use. It's not that hard to do, and it is a very stable way to talk to the PIC. Jim -----Original Message----- From: pic microcontroller discussion list [TakeThisOuTPICLISTEraseMEspam_OUTMITVMA.MIT.EDU]On Behalf Of James Newton, webhost Sent: Friday, October 24, 2003 16:37 To: RemoveMEPICLISTTakeThisOuTMITVMA.MIT.EDU Subject: [PIC:] safely driving a pic input from a PC Parallel port. I've been thinking about how things get connected between PC's and PIC's lately and overall what I can see is: USB: Nice! But not the lowest possible cost and not supported on some old PC's Also, requires some driver development on the PC side... not hard, but still. RS232: Nice... But you can run out of ports, some current laptops don't even have serial ports (!) and you have to do all the darn configuration on the PC to make sure all the settings are right. Has anyone been able to get a USB to RS232 adapter to work right? I can never get the PC application to see the port. Parallel: Sigh... NT and up driver problems (IF you don't simulate a printer with the PIC) but there are solutions for that. One the one port usually and I've seen laptops without parallel ports. USB to parallel adapters are really shaky... has anyone used one to talk to anything other than a printer? The two I've tried don't show up as LPT ports on the PC and so old applications don't see them. Of all the above, probably the most reliable, simple, and cheapest for us little guys is the parallel port (this is IMHO and I'm sure the point can be argued 20 ways from Sunday). So, if I want to put out something that people can setup from a PC, a temporary connection via the parallel port is probably best. Now the question is how do I keep people from bread boarding something that connects via my device that will blow up the PC (as well as me probably) when they screw up? I went on a hunt for parallel port isolation and, not being the most brilliant analog electronics engineer (ok, so I'm a complete moron in that area) I really don't seem to understand most of what I found. I understand that the port can sink 24mA and only source 2.6mA so obviously its a good idea to drive your loads from the port, through the load, then to +5v rather than ground. Safe current limiting can be had with a series resistor of 1.6k or more right? (4.2V / 2.6mA) But that doesn't leave much drive... so you can add a driver, but wouldn't it be better to prevent the pin from sourcing anything by putting a diode long with the resistor in series and then reducing the resistor to 175 or more ohms (4.2V / 24mA)? +5 | / \ / 220R 1N4148 \ >-o---,/\/\/'----|<-----o-----=> _|_/ //_\ 5.1V | GND A pull-up resistor is added to the output to drive TTL or other loads and a Zener could be added to ground to prevent over voltage or driving the pin in reverse. That seems to me to be a nice little, very low cost, easy to work with, circuit to provide some basic protection for kids or students to tack things onto. Not wanting to assume I have a clue, I searched the internet for other solutions. Tomi Engdahl says: "The following circuit uses two 1N4148 diodes to protect parallel port against higher than +5V signals and also against wrong polarity signals (power on the circuit is accidentally at wrong polarity. " Diode 1N4148 4.7K parallel >-|>|-+--\/\/\/--etc... port data | pin +-|<|-+ 1N4148 | parallel >-----------+ port ground | Ground "Adding even more safety idea: Replace the 1N4148 diode connected to ground with 5.1V Zener diode. That diode will then protect against over-voltage spikes and negative voltage at the same time. " Ok, he has his series diode in the exact opposite direction. Could somebody please tell me if I'm just being stupid? Other than that (and the much larger series resistor) the circuit is the same. I also found on piclist.com: "Another way to protect against over-voltage is to put a regular diode with the cathode on the signal line and the anode on the power supply line. Keep in mind that this will cause the power supply to receive the over-voltage and may result in damage to other components on the board." Standard diodes are cheaper... any reason why I can't put ONE Zener in the power supply and use regular diodes on the 8 pins? x8: +5 | +5 / _|_ \ _\ /_ / | 220R 1N4148 \ >-o---,/\/\/'----|<-----o-----=> x1: +5 _|_/ //_\ 5.1V | GND --- James Newton, webhost piclist.com (former Admin #3) jamesnewtonEraseME.....piclist.com 1-619-652-0593 fax:1-208-279-8767 PIC/PICList FAQ: http://www.piclist.com or .org -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservmitvma.mit.edu with SET PICList DIGEST in the body -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservEraseMEEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

2003\10\27@110827 by Jim Monteith

Hi James, Below are 4 sources for different "port openers", and then just a pretty good place to start reading about parallel port programming. I searched for "program windows 2000 parallel port" on google and came up with lots of hits. I was wrong with the name I told you. I am using "userport". It works pretty well. You have to be careful with your timing because what you are doing is passing off your data to the userport driver which then sends it to the port. I have noticed that sometimes I lose data because I'm trying to read or write too fast. Just be aware of that. Hope this helps, Jim http://www.zeecube.com/IOAccess/index.htm http://www.beyondlogic.org/porttalk/porttalk.htm www.embeddedtronics.com/public/Electronics/minidaq/userport/UserPort. zip http://www.logix4u.net/inpout32.htm http://www.lvr.com/parport.htm {Original Message removed}

2003\10\27@121728 by Dennis Crawley

Alan, Yes, I mean that. I've planned to do something similar to James but to serial port oriented. Regards, [OTps:] I can't figured out what lost us that match (Ireland16-Argentina15). I wish I had been at pub just to see the Irish party. From: "Alan B. Pearce" To: <RemoveMEPICLISTspam_OUTKILLspamMITVMA.MIT.EDU> Sent: Monday, October 27, 2003 7:55 AM Subject: Re: [PIC:] safely driving a pic input from a PC Parallel port. > >Dennis Crawley says: "3.- 6n317 opto, VERY fast!" > > I think he means the 6N137, which is a Darlington transistor output. > > However I prefer the HCPL2300 which has an internal opto receiver, and comes > out as an open collector digital signal, with a 1k pull up resistor > available for connection. Requires 5V supply for the receiver side, and is a > heap faster than the 6N137. Both devices are Agilent ones. > > -- > http://www.piclist.com#nomail Going offline? Don't AutoReply us! > email RemoveMElistservTakeThisOuTspammitvma.mit.edu with SET PICList DIGEST in the body -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspamspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

More... (looser matching)
- Last day of these posts
- In 2003 , 2004 only
- Today
- New search...