Post by thread:[PIC]: PWM output wave form

Remembered to give Roman the gears today but forgot to post my question ;-] So here goes. I keep reading that the pwm output generated from the pic is not that sharp/clean a signal. I've seen the suggestions for cleaning it up which is simple to implement but I was wondering if anyone has a image of the raw pwm wave form and perhaps another after its filtered? It might be a bit longer before I can get something running here to see it myself so I was curious if anyone could take a snap of the scope for me. Thanks. Dave -- http://www.piclist.com hint: To leave the PICList spam_OUTpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

--Message-Boundary-26196 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Content-description: Mail message body > I keep reading that the pwm output generated from the pic is not that > sharp/clean a signal. I've seen the suggestions for cleaning it up > which is simple to implement but I was wondering if anyone has a image > of the raw pwm wave form and perhaps another after its filtered? Not sure why you should think that the PWM output is not sharp/clean because in my experience it is so with a PIC. Nice square edges, 0V when low, 5V when high, duty cycle and fequency just what you tell it to be. A PWM waveform is a "square" wave, where you have the ability to alter the ratio of the high and low periods. See attached diagram for examples. Normally you filter a PWM signal when your intention is to produce an analog value. Ideally the output signal is then a DC voltage (no waveform) that the average value of the PWM waveform. Typically you might use this to produce an analog voltage from 0-5V from a duty cyle of 0-100%. Less than perfect filtering (such as a simple RC filter) can leave some AC ripple at hte PWM frequency on the signal. Does this help? -- Brent Brown, Electronic Design Solutions 16 English Street, Hamilton, New Zealand Ph/fax: +64 7 849 0069 Mobile/txt: 025 334 069 eMail: .....brent.brownKILLspam@spam@clear.net.nz --Message-Boundary-26196 Content-type: Image/GIF; name="PWM.gif" Content-disposition: attachment; filename="PWM.gif" Content-transfer-encoding: BASE64 R0lGODlh9wByALMAAAAAAIAAAACAAICAAAAAgIAAgACAgICAgMDAwP8AAAD/AP//AAAA//8A /wD//////ywAAAAA9wByAAAE/hDJSau9OOvNu/9gKI5kGT4ZiqkXa7kVTMkTLdkIrqf82re/ VzA2nBVrpqRyyWw6n9CodEqtWq/YrHbL7Xq/4LB4TC6bz+i0es1uu9/wuHxOr9vv351vD+QL /USARoJIhDdHh4Y5iIuKen14kZKTlJWWl5iZmpucnZ6foKGio6SlpqeoqZyPfxcBr7CxAYw4 sra0GbayuBi6sbyQrcKBwcRTs71PyK7Kuc3JqiPLFtNL1RTXSdkS2yXd3VgJ4hTi5RPjCAlj 38/MTuzvzl/qEvTpF+P2YfBN/Ez+1uSJsafPQsEvAJUk1CbwX8N55MxVOIjwoUKLDKH1w7iF Yj1y/mYWmhDpjWNJjV4OEgSJbp9JEiRhvpQ2s0q5lulw0sNZEWVAnxeBZnTniVUFX7ByIV3K VCnTp7d6QZ2aVCrVqcCGDSoWravXr2DDih1LtqzZs2jTql3Ltq3bt6iMbtVaiGtduons5sXb SG9fvnLvGhs8l7DgwnATK17MuLHjx5AjS55MubLly5i9Bt579OovRZ4/+w1ddTTpbpv/GuaM +DAUcDWJgoAtlANt2R9un9J9lKYI3th8hwA+gfiIm/ok1tu5Tvhs57mhezCOgHoIiiuX32v+ W3oH6uC927aCXftEMuG7qx8ufkN6KjdZbj9X5v3z9ffZ44+ehSBz+vW1/qeBffzpZ2B+U2Q3 X3I5cXdggQhCKOF0Ajah3HIMNigGgRTu1+GDH0YYSWpNgXYaI6eh5lRoKJ7oSFarZSbjjDTW aOONOOao44489ujjj2WlJiSMrbHmmpFIqlakkkcymeSQL0YJ5JRUVmnllVhmqeWWXHbp5Zdk QOlKikitSOYuVp2J5phqrgkYkU3isFE8tZ2EW1B3DrVbO9Tw2RudeY7EBHLnxGeeR1tYZ5Ki dTIaaHV+LqFgRPO5BGifl/4556MyPTGpQehFWpyo3JAKaaZKJHfhR6GiOqqrpcJ66qaSZpBh q7Rq6lCdnXJa4XXnsfoRT144iqmvHr6aa3Cyah5nkE7mWbqssrsiCyK1P8lB4lJwcsOtlNR8 61ct4r6Zpi7dOqmunGC26+678MYr77z01mvvvfiaIKa546a7b4z/LhlwnP7ma/DBCCes8MIM N+zwwxCXMfCTBYMLcMX9WiwwxnxF7DFbEQAAOw== --Message-Boundary-26196-- -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestKILLspammitvma.mit.edu

On Sat, 20 Jul 2002, Dave King wrote: > Remembered to give Roman the gears today but forgot to post my question ;-] > > So here goes. > > I keep reading that the pwm output generated from the pic is not that > sharp/clean > a signal. I've seen the suggestions for cleaning it up which is simple to > implement but > I was wondering if anyone has a image of the raw pwm wave form and perhaps > another after its filtered? > > It might be a bit longer before I can get something running here to see it > myself so > I was curious if anyone could take a snap of the scope for me. Dave, Back in March when we were talking about this stuff, I wrote an Octave (open source implementation of MATLAB) script to synthesize a sine wave from Pulse Modulate (but not PWM) signal. A PWM filtered signal will look similar. The script: http://www.dattalo.com/swsd.m http://www.dattalo.com/p2.png http://www.dattalo.com/p3.png http://www.dattalo.com/p4.png p2.png and p4.png show two full cycles of the filtered sine, the difference is that the RC time constant is slightly different. p3.png is a close-up shot of p2.png. Roman has some similar screen shots too: http://centauri.ezy.net.au/~fastvid/picsound.htm Scott -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam.....mitvma.mit.edu

Brent, Scott I going to use the pwm to drive a 2N2222A which is in turn going to control a mosfet. I wanted to know if the pwm that is produced is a clean square wave or if it is really dirty and has to be cleaned before it will let the 2222 switch nicely. Of course I read the bit about the dirty pwm output in a app note and now can't find it again. And if I had a pic running right now I could look and not have to ask dumb questions ;-] Was just curious if anyone had actually noticed this or if its another "caution" to just ignore. Thanks Dave -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Hi Brent. Somebody posted it here already sometime ago... You can produce +5 ... -4.5 volts output by using PWM from single +5 source. WBR Dmitry. Brent Brown wrote: > > Normally you filter a PWM signal when your intention is to produce an > analog value. Ideally the output signal is then a DC voltage (no > waveform) that the average value of the PWM waveform. Typically you > might use this to produce an analog voltage from 0-5V from a duty > cyle of 0-100%. Less than perfect filtering (such as a simple RC > filter) can leave some AC ripple at hte PWM frequency on the signal. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> I going to use the pwm to drive a 2N2222A which is in turn going to > control a mosfet. I wanted to know if the pwm that is produced is a > clean square wave or if it is really dirty and has to be cleaned > before it will let the 2222 switch nicely. > > Of course I read the bit about the dirty pwm output in a app note and > now can't find it again. And if I had a pic running right now I could > look and not have to ask dumb questions ;-] Was just curious if > anyone had actually noticed this or if its another "caution" to just > ignore. Dave, It is a really clean square wave as far as I have seen, no nasty glitches or double pulses to cause any problems. I have not heard of any such cautions. It is a push-pull output like most PIC outputs and will drive the 2N2222 nicely with an appropriate base drive resistor. It can also drive a MOSFET directly, perhaps even better than if you used the 2N2222 in between. That is, if the MOSFET does not have too much gate capacitance and if 5V gate voltage is OK. -- Brent Brown, Electronic Design Solutions 16 English Street, Hamilton, New Zealand Ph/fax: +64 7 849 0069 Mobile/txt: 025 334 069 eMail: EraseMEbrent.brownspam_OUTTakeThisOuTclear.net.nz -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

On Sat, 20 Jul 2002, Dave King wrote: > Brent, Scott > > I going to use the pwm to drive a 2N2222A which is in turn going to control > a mosfet. > I wanted to know if the pwm that is produced is a clean square wave or if > it is really > dirty and has to be cleaned before it will let the 2222 switch nicely. The only concern I'd have is that the gate capacitance of the MOSFET will filter narrow pulses. You mention a 2N2222A, which is an NPN BJT, and I assume it's configured with the emitter tied to ground and the collecter tied to the gate of the MOSFET. If so, then this means you're using a resistor to turn on the MOSFET. That resistor and the capacitance of the MOSFET's gate will form a relatively slow RC circuit (unless the resistor is tiny - in which case you'll burn quite a bit of power). I'd suggest getting rid of the 2N2222 and using a MOSFET driver instead: PIC I/O --> MOSFET driver ---> MOSFET Scott -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

>It is a really clean square wave as far as I have seen, no nasty >glitches or double pulses to cause any problems. I have not heard of >any such cautions. It is a push-pull output like most PIC outputs and >will drive the 2N2222 nicely with an appropriate base drive resistor. > >It can also drive a MOSFET directly, perhaps even better than if you >used the 2N2222 in between. That is, if the MOSFET does not have too >much gate capacitance and if 5V gate voltage is OK. Thanks Brent Thats what I wanted to know. I'll have to see about driving the mosfet directly. You realize this might require further dumb questions aka pleading for help to the list ;-] Cheers Dave -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads