Controlling a motor in presence of phase shift
Bookmark this page:
 Yahoo!
 Google
 Windows Live
 del.icio.us
 digg
 Netscape
|
|
|||||||||||||
|
Posted by fssg on July 3, 2006, 12:09 am
Please log in for more thread options I am trying to control speed of a single phase fan motor using triac phase control, across a optoisolated boundary. The motor could be of shaded-pole or capacitor-run type, between 30 and 200W. My circuit looks like this: 4K7 1N4001 47K 4N25 .-|___|----- 5V ___ | Active --|<--|___|------------' '------------ To uC INT - |/ From uC 120R -----. '------o-----o----o------o--- Active | | | | | | | 470R.-. | | | | | | | | .-. | | | | | | | | | | | MOC3021 360R '-' | | | | | | ___ | | '-' | | | .----|___|-----o | | | | | _|_ | ' | .-.M | V -> V_A | _|_ | | |O ---
- / | | V_A | | |V --- 100nF
| '--------------)--- / | | '-' | ----' | | --- | | | | --- | | --- | 100N| | | 47nF --- | | | | | | | | ___ | '------o-----o----o--UUU-o------- To load 100uH T405-600 or BT137 (created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de) The 4N25 optocoupler detects the zero crossing, a microcontroller turns on an optotriac LED after the required delay, and holds it on until about 20deg before the end of the half cycle. I have noticed that 100% power (where the voltage across triac is almost always zero) is at about 25deg for a 173W capacitor-run motor, 60deg for a 50W shaded pole motor. This is giving me two problems: 1. If I turn on the triac eariler than this point, I start to get 'half-waving' (?) but I can't see why this happens? Shouldn't it just continue to run at 100%? 2. Since I don't know what motor my users will plug into it, how do I prevent this happening? My zero crossing comes straight from the mains - could I put the zero crossing detector across the triac to detect when the triac turns off and have my timing referenced to that point instead? I also notice that on the shaded-pole motor, the circuit doesn't seem to turn on the triac properly at high firing angles. Starting at 140deg, I can see where the triac is being turned on, the voltage across the triac dips to zero briefly but then rises back up to mains voltage. At about 125 deg it starts to do the negative half cycles properly, then positive half cycles start working from about 92deg. Does anyone know why this might happen? Thanks for any help. | |||||||||||||
|
Posted by Ban on July 3, 2006, 6:44 pm
Please log in for more thread options fssg wrote: > I am trying to control speed of a single phase fan motor using triac
> phase control, across a optoisolated boundary. The motor could be of > shaded-pole or capacitor-run type, between 30 and 200W. > > My circuit looks like this: > > > 4K7 > 1N4001 47K 4N25 .-|___|----- 5V > ___ | > Active --|<--|___|------------' '------------ To uC INT > - |/ > ^ -> -| > Neutral ------------' |> > '---- Gnd > > > > > From uC 120R > -----. '------o-----o----o------o--- Active > | | | | | | > | 470R.-. | | | | > | | | | .-. | | > | | | | | | | | > | MOC3021 360R '-' | | | | | > | ___ | | '-' | | > | .----|___|-----o | | | | > | _|_ | ' | .-.M | > V -> V_A | _|_ | | |O --- > - / | | V_A | | |V --- 100nF > | '--------------)--- / | | '-' | > ----' | | --- | | > | | --- | | > --- | 100N| | | > 47nF --- | | | | > | | | | ___ | > '------o-----o----o--UUU-o------- To load > 100uH > T405-600 > or > BT137 > (created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de) > > > The 4N25 optocoupler detects the zero crossing, a microcontroller > turns on an optotriac LED after the required delay, and holds it on > until about 20deg before the end of the half cycle. > What you want is IMHO impossible with your circuit. You will need a variable frequency synthesized from a DC intermediate voltage similar to a switched mode power supply. SCRs are absolutely unsuitable for this. I do not think your electronic understanding will be up to this task. Your zero x-ing detector should not be done with a 1N4001, but with a 1N4004 to -7. You might consider doing an incandescent light control with your circuit instead. > I have noticed that 100% power (where the voltage across triac is
> almost always zero) is at about 25deg for a 173W capacitor-run motor, > 60deg for a 50W shaded pole motor. This is giving me two problems: > 1. If I turn on the triac eariler than this point, I start to get > 'half-waving' (?) but I can't see why this happens? Shouldn't it just > continue to run at 100%? > 2. Since I don't know what motor my users will plug into it, how do I > prevent this happening? My zero crossing comes straight from the > mains - could I put the zero crossing detector across the triac to > detect when the triac turns off and have my timing referenced to that > point instead? > I also notice that on the shaded-pole motor, the circuit doesn't seem > to turn on the triac properly at high firing angles. Starting at > 140deg, I can see where the triac is being turned on, the voltage > across the triac dips to zero briefly but then rises back up to mains > voltage. At about 125 deg it starts to do the negative half cycles > properly, then positive half cycles start working > from about 92deg. Does anyone know why this might happen? > > Thanks for any help. You will blow up either the motor or your triac. -- ciao Ban Apricale, Italy | |||||||||||||
|
Posted by Phil Allison on July 3, 2006, 8:16 pm
Please log in for more thread options
"Ban" >
> What you want is IMHO impossible with your circuit. You will need a > variable frequency synthesized from a DC intermediate voltage similar to a > switched mode power supply. ** Complete drivel. > SCRs are absolutely unsuitable for this.
** Triacs are widely use to control the speed of AC fans. > I do not think your electronic understanding will be up to this task.
** Nor yours. > Your zero x-ing detector should not be done with a 1N4001, but with a
> 1N4004 to -7. ** Sure - but that detector is dodgy anyhow. > You might consider doing an incandescent light control with your circuit
> instead. ** A cheap pulse fired dimmer does that. > You will blow up either the motor or your triac.
** Very helpful - not. ....... Phil | |||||||||||||
|
Posted by fssg on July 3, 2006, 10:33 pm
Please log in for more thread options
>
> "Ban" > >>
>> What you want is IMHO impossible with your circuit. You will need a >> variable frequency synthesized from a DC intermediate voltage similar to >> a switched mode power supply. >
> > ** Complete drivel. > > >> SCRs are absolutely unsuitable for this.
>
> > ** Triacs are widely use to control the speed of AC fans. > > >> I do not think your electronic understanding will be up to this task.
>
> > ** Nor yours. > > >> Your zero x-ing detector should not be done with a 1N4001, but with a
>> 1N4004 to -7. >
> > ** Sure - but that detector is dodgy anyhow. Hi Phil. Why is my detector dodgy? It's cheap, it causes some timing error but I can remove that in software. I don't know if you remember my last thread but I bought the CRO you suggested. It's really bad :) At least it won't be such a great loss if I blow it up while working on the mains. BTW how did you know I was in Australia or was it just coincidence? (I am using a US newserver). >
> >> You might consider doing an incandescent light control with your circuit
>> instead. >
> > ** A cheap pulse fired dimmer does that. > > > >> You will blow up either the motor or your triac.
>
> > ** Very helpful - not. > > > > > ....... Phil > > > > > > | |||||||||||||
|
Posted by Phil Allison on July 3, 2006, 10:52 pm
Please log in for more thread options
"fssg" >>
>> ** Sure - but that detector is dodgy anyhow. >
> Hi Phil. Why is my detector dodgy? It's cheap, it causes some timing > error but I can remove that in software. ** Then you agree it is dodgy. > I don't know if you remember my last thread but I bought the CRO you
> suggested. It's really bad :) ** It is a nice CRO - particularly for the money. > At least it won't be such a great loss if I blow it up while working on
> the mains. ** Never use it on mains without the probe switched to 10:1. Even then, be VERY careful. > BTW how did you know I was in Australia or was it just coincidence? (I am
> using a US newserver). ** I used it as an example of what is available. DSE sell all over the world anyhow. ........ Phil | |||||||||||||
Neutral ------------' |>