Searching \ for 'AC fan speed control ideas' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power.htm?key=ac
Search entire site for: 'AC fan speed control ideas'.

Truncated match.
PICList Thread
'AC fan speed control ideas'
1999\09\29@172056 by Morgan Olsson

picon face
I recently got a request to speed control a AC fan.
(100W AC fan cost very much less than an DC fan, espesially also counting the power supply...)

If somebody have some experience on this, please give me some aid/tip/stories about how to let smoke out of fans etc ;)

Fan spec:
Large electronic fan type, 100W, 230VAC,
"Motor run capacitor" generating second phase
(What is the english word for that type of motor?)

Speed control spec:
No precision needed, 4-10 steps fron 20% to full speed, no feedback neccessary, must not generate noise nor heat, and must be low cost.

For me,
Cirquit design is not a problem
Programming is not a problem

But *what* shall the gizmo do?

Because of noise, cost, and EMI, I cannot use high frequency switching

Important: This type of motor need a good balance between positive and negative voltage cycle.

I have three ideas so far:

1)      Conventional phase angle. Problem A): Fan noise (this make the technique
unusable) and EMI  B) might be strange current resonances due to capacitor and double windings, and running at low speed compared to fullspeed.  Giving maybe high current, high heating etc?

2)      Pulsing AC bursts: Power on a whole number of AC cycles, then off a whol
e number of AC cycles.  For least noise:  Power on at voltage zero crossing, off at fan current zero crossing (automatically done using triac)  Problem A) Probably we can hardly notice, but the fan vill be saying hum, hum, hum very quickli as we pulse full power att underpeed?  B) The losses in the windings will be high, but the duty cycle is low, also remembering that the torque is proportional to speed squared, so the duty cycle will probably be much less than ten increase in peak losses, = no problen (?)   Probably the on and off cycle numbers shall be about ten?  If total cycles is always 20 I can vary off to on in 20 steps (5%), but the rpm at 80-100% on will probably be about the same as the motor will "lock" to AC, so to save winding losses it is probably better to jump from 75% to full on.
Also the on and off cycles shall be interleaved to avoud unnecessarily low pulsing speed; instead of 15 on, then 5 off, it shall interleave to three on, one off, etc...

3)      Halfwave pulsing:  Example,  turn on a positive half wawe, off during th
e following neg and pos halfwaves, then on at neg halfwave, off att following pos and neg halfwave, and from beginning, giving f/3.

Or a combination of techniques:
3 for Lowest speed
2 for medium speed
1 close to full speed?

A variant of 2 and 3: one cycle on, one cycle off, one half cycle on, one cycle off, and again  (you need to draw this....)
That vill give slightly higher speed than even cycles on, odd off, so by combinaitons we can have high speed resolution, and still pulsing the fan fast enough not to hear "hum-hum-hum...."


Nice thing to use microcontrollers is that we can change algoritm and still use the same hardware  :)

I plan to use a 12C508, trigged by zero crossing.
If needing phase angle i relay on zero crossing + delay using internal 4MHz osc.
Power supplied by a small transformer and controlling the power stage using optocoupler.  I need low voltage side for a couple of sensors and serial com to remote panel (a pic in there too of course... if not an AVR)

Thanks in advance
/Morgan


Morgans Reglerteknik, HŠllekŒs, 277 35 KIVIK, SWEDEN
  tel +46(0)414-446620, fax -70331,   spam_OUTmrtTakeThisOuTspaminame.com

1999\09\30@164351 by l.allen

picon face
> I recently got a request to speed control a AC fan.
> (100W AC fan cost very much less than an DC fan, espesially also counting the
power supply...)
>
> If somebody have some experience on this, please give me some aid/tip/stories
about how to let smoke out of fans etc ;)
>
Most AC fans I come across are induction type motors.. as in squirrel
cage rotors.
The only real way to vary these suckers is to effectively change the
AC frequency being presented too it.

Other methods will cause increased slip but the motor speed will tend
to be very unreliable and even stall easily at lower speeds.
Phase angle is really for brushed motors like hand drills etc.

There are various methods for varying the AC freq but most involve
high freq switching as low freq iron transformers are large,
expensive and freq dependent.Also generation of an sine wave without
switching techniques (high freq) is very inefficient.

I know this is a vague rendering of the facts but you're brief was for
low freq and low cost.. bad luck.



_____________________________

Lance Allen
Technical Officer
Uni of Auckland
Psych Dept
New Zealand
_____________________________


'AC fan speed control ideas'
1999\10\01@073820 by Morgan Olsson
picon face
Lance Allen wrote

>The only real way to vary these suckers is to effectively change the
>AC frequency being presented too it.

Yep, used such controllers.  Far too expensive for this...

>Other methods will cause increased slip but the motor speed will tend
>to be very unreliable and even stall easily at lower speeds.

I have one spare pin, possible to use for feedback using a magnet mounted in center of fan, and a cheap bipolar hall sensor close to it.

>Phase angle is really for brushed motors like hand drills etc.

Therefor I will try pulsing x number of cycles and y cycles pause instead.

Anybody done that?

I will start testing by mounting a temperature sensor in the windings...

>There are various methods for varying the AC freq but most involve
>high freq switching as low freq iron transformers are large,
>expensive and freq dependent.Also generation of an sine wave without
>switching techniques (high freq) is very inefficient.
>
>I know this is a vague rendering of the facts but you're brief was for
>low freq and low cost.. bad luck.

Thank you... ;)

/Morgan
Morgans Reglerteknik, HŠllekŒs, 277 35 KIVIK, SWEDEN
  tel +46(0)414-446620, fax -70331,   .....mrtKILLspamspam@spam@iname.com

1999\10\01@083619 by Reginald Neale

flavicon
face
>Therefor I will try pulsing x number of cycles and y cycles pause instead.
>
>Anybody done that?
>

 Yes, I have tried this to control the speed of a fan motor.
 It ends up being very noisy and has limited range. Finally
 used a DC motor instead.

 But "dropped cycles" is easy to implement with a PIC
 and an SSR if it works for your motor.

 Reg Neale

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...