Post by thread:[PIC]: Converting PIC port output to +-5V @ 20 ma

I hope this isn't embarrasingly simple. I need to have a PIC port drive a signal, referenced to ground, at 5V or -5V, either sourcing or sinking up to 20 ma (per bit, 8 bits). Furthermore, it would be prudent, but not necessarily required, to have galvanic isolation between the input and output, as the output may be subject to static electricity. This would be easy to build with discrete parts, using two FETs arranged similarly to a CMOS output, but with the lower one connected to -5V instead of ground. Depending on which FET is conducting, the driver output gets "hooked" to +5V or -5V. (And opto FETs could be used for isolation if needed.) To save parts count (16 FETs!) and $$$, I'd rather go for an integrated approach. There are many quad 1:2 analog multiplexers out there, which is what the discrete circuit above really is. A couple of these could switch eight outputs between +5V and -5V inputs very easily. Unfortunately, none seem to be able to handle the -5V swing on an analog input port. The TI SN74CBT3257, for example, has absolute max ratings of -0.5V to 7V analog input swing: http://www-s.ti.com/sc/psheets/scds017l/scds017l.pdf I've also thought of using a comparator, with +-5V supplies and an input reference set between the PIC high and low output voltage. The LM339 can only sink 16 ma maximum, typical is 6 ma, and in general, 20 ma seems to be on the high end of the current sinking capability of common comparators. Any suggestions? Am I completely missing some other, simpler approach? Does someone make an "Octal +-5V input swing analog multiplexer using CMOS level inputs" part that I can order from Digikey? Thanks, Johnathan -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spam_OUTlistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

At 05:10 PM 2/4/2003 -0800, you wrote: {Quote hidden}>I hope this isn't embarrasingly simple. > >I need to have a PIC port drive a signal, referenced to ground, at 5V or >-5V, either sourcing or sinking up to 20 ma (per bit, 8 bits). >Furthermore, it would be prudent, but not necessarily required, to have >galvanic isolation between the input and output, as the output may be >subject to static electricity. > >This would be easy to build with discrete parts, using two FETs arranged >similarly to a CMOS output, but with the lower one connected to -5V >instead of ground. Depending on which FET is conducting, the driver >output gets "hooked" to +5V or -5V. (And opto FETs could be used for >isolation if needed.) > >To save parts count (16 FETs!) and $$$, I'd rather go for an integrated >approach. There are many quad 1:2 analog multiplexers out there, which >is what the discrete circuit above really is. A couple of these could >switch eight outputs between +5V and -5V inputs very easily. >Unfortunately, none seem to be able to handle the -5V swing on an analog >input port. The TI SN74CBT3257, for example, has absolute max ratings >of -0.5V to 7V analog input swing: You could consider 3 74HC4053s. No isolation, but < 80 cents total. Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" .....speffKILLspam@spam@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

Just off the top of my head - could you just use a MAX 232 driver chip - possibly with modified supply rails/charge pump circuit???? May not be viable but should be OK for +/- 20mA, & Static protected to a degree. although not isolated. Or a different line driver chip?? Richard P Johnathan Corgan To: .....PICLISTKILLspam.....MITVMA.MIT.EDU <jcorgan@AEINE cc: T.COM> Subject: [PIC]: Converting PIC port output to +-5V @ 20 ma drive Sent by: pic microcontrolle r discussion list <PICLIST@MITVM A.MIT.EDU> 05/02/03 14:10 Please respond to pic microcontrolle r discussion list I hope this isn't embarrasingly simple. I need to have a PIC port drive a signal, referenced to ground, at 5V or -5V, either sourcing or sinking up to 20 ma (per bit, 8 bits). Furthermore, it would be prudent, but not necessarily required, to have galvanic isolation between the input and output, as the output may be subject to static electricity. This would be easy to build with discrete parts, using two FETs arranged similarly to a CMOS output, but with the lower one connected to -5V instead of ground. Depending on which FET is conducting, the driver output gets "hooked" to +5V or -5V. (And opto FETs could be used for isolation if needed.) To save parts count (16 FETs!) and $$$, I'd rather go for an integrated approach. There are many quad 1:2 analog multiplexers out there, which is what the discrete circuit above really is. A couple of these could switch eight outputs between +5V and -5V inputs very easily. Unfortunately, none seem to be able to handle the -5V swing on an analog input port. The TI SN74CBT3257, for example, has absolute max ratings of -0.5V to 7V analog input swing: http://www-s.ti.com/sc/psheets/scds017l/scds017l.pdf I've also thought of using a comparator, with +-5V supplies and an input reference set between the PIC high and low output voltage. The LM339 can only sink 16 ma maximum, typical is 6 ma, and in general, 20 ma seems to be on the high end of the current sinking capability of common comparators. Any suggestions? Am I completely missing some other, simpler approach? Does someone make an "Octal +-5V input swing analog multiplexer using CMOS level inputs" part that I can order from Digikey? Thanks, Johnathan -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspam_OUTTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

I doubt you will get a logic chip that can drive a -+5 voltage swing. Why not stick with the TI chip design, and invert 8 of the outputs into -5V using a couple of chips such as the LMC7660 switched capacitor voltage converter - this chip is designed to switch a +v to a -v with just two capacitors? Just a suggestion ... Trouble is - this design comes up to at least 4 IC's and I am wondering if that would be any cheaper than using 16 FETS in a cascaded Mosfet switch design. Regards Ian {Original Message removed}

Spehro Pefhany wrote: > You could consider 3 74HC4053s. No isolation, but < 80 cents total. These come really close. 10V Vcc-Vee is the absolute max, though, I'd hate to push this. Also, the ON resistance is fairly high (typical 70, max 160). The signal must drive a line that ultimate terminates in two diode drops to ground, so a total of about 200 ohms needs to be used in series to drop from 5V Vcc to 1.2V nominal at 20 ma. The output FET would then be about a third to two thirds of this--wouldn't this make for a lot of spread in current values between parts? -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

At 07:25 PM 2/4/2003 -0800, you wrote: >Spehro Pefhany wrote: > >>You could consider 3 74HC4053s. No isolation, but < 80 cents total. > >These come really close. 10V Vcc-Vee is the absolute max, though, I'd >hate to push this. http://www.fairchildsemi.com/ds/MM/MM74HC4053.pdf +/- 7.5 seems to be abs. max. >Also, the ON resistance is fairly high (typical 70, max 160). The on resistance is highest well inside the Vcc- Vee voltage range, and is lowest at the rails, right where you'd use it. Typical is more like 20-30 ohms, you can probably scale that safely by the typ:max ratio. See the curve on page 7. >The >signal must drive a line that ultimate terminates in two diode drops to >ground, so a total of about 200 ohms needs to be used in series to drop >from 5V Vcc to 1.2V nominal at 20 ma. The output FET would then be >about a third to two thirds of this--wouldn't this make for a lot of >spread in current values between parts? I'll leave those calculations to you, as I don't know your application. Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" KILLspamspeffKILLspaminterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

Spehro Pefhany wrote: > http://www.fairchildsemi.com/ds/MM/MM74HC4053.pdf > > +/- 7.5 seems to be abs. max. > > The on resistance is highest well inside the Vcc- Vee voltage range, and > is lowest at the rails, right where you'd use it. Typical is more like > 20-30 ohms, you can probably scale that safely by the typ:max ratio. > See the curve on page 7. I was looking at the TI version, this Fairchild part is considerably better, and in fact meets the specs for what I need. A couple of these does the trick. Thanks. > I'll leave those calculations to you, as I don't know your > application. The output drives a PIN diode RF switching circuit that uses the 20 ma to get the ON resistance of the PIN diodes down to a fraction of an ohm. Driver -> RF Choke -> PIN Diode -> Coax -> PIN Diode -> RF Choke -> GND The two diodes drop about 1.2V total, so 3.8 volts/20ma = ~200 ohms series resistance needed. This would be split between the ON resistance of the output FET and a series resistor. If the ON resistance of the FET is only 30 ohms, then this is only 15% of the total, and part variances would be limited to affecting the current by only this amount. If the ON resistance were 160 ohms (TI part worst case), then part variances would affect the current dramatically. The -5V reverse biases these same PIN diodes, as well as forward biases a different set of PIN diodes acting as shunts across the coax to provide even further isolation. -Johnathan -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

Johnathan Corgan <TakeThisOuTjcorganEraseMEspam_OUTAEINET.COM> wrote: > I hope this isn't embarrasingly simple. I think so. > I need to have a PIC port drive a signal, referenced to ground, at 5V or > -5V, either sourcing or sinking up to 20 ma (per bit, 8 bits). > Furthermore, it would be prudent, but not necessarily required, to have > galvanic isolation between the input and output, as the output may be > subject to static electricity. Somebody already suggested MAX232 (or similar), but if you've already got the +/- power supplies, why not a plain old 1488 driver? Four per chip, and they're pretty rugged. -- Dave Tweed -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

Johnathan Corgan wrote: > > I hope this isn't embarrasingly simple. > > I need to have a PIC port drive a signal, referenced to ground, at 5V or > -5V, either sourcing or sinking up to 20 ma (per bit, 8 bits). No isolation, but 2 cheap ideas for low chip count; * use a CMOS 8 buffer chip good for 10v supply rails, connect it between +5 and -5 rails. * use a UCN5804 chip (8 darl drivers), and 8 pullup resistors. Obviously some resistors are needed for biasing etc but resistors are cheap. -Roman -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservEraseME.....mitvma.mit.edu with SET PICList DIGEST in the body

> www.fairchildsemi.com/ds/MM/MM74HC4053.pdf > +/- 7.5 seems to be abs. max. I was looking at the TI version, this Fairchild part is considerably better, and in fact meets the specs for what I need. A couple of these does the trick. Thanks. You may want to get CD4053B parts rather than the 74HC parts. The 4000 series has always had a wider supply range than the 74HC part, and I expect the analog multiplexor voltages go up accordingly... (And since fairchild makes a 4053, I'll bet there 74HC part is the same as that, where as TI's part is perhaps true 74HC technology.) BillW -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestEraseMEEraseMEmitvma.mit.edu>