18
\$\begingroup\$

There are some devices that look like ordinary light dimmers, but are sold expressly for ceiling fan speed control. For example: Lutron Diva Quiet 3-speed control. Unlike an ordinary dimmer, this has a three-position switch for low, medium, and high speeds, instead of a continuously variable knob. The manual shows typical wiring, which is like an ordinary dimmer:

wiring diagram

(the unconnected red/white wire is for an optional 3-way circuit)

I've heard people say "don't run a fan on a dimmer; you will start a fire", but don't say why. Yet, that Lutron sells this device expressly for fans suggests it is something other than an ordinary triac light dimmer, and that it's totally safe, but they also don't say why.

I also know when I bought this house, a different, uglier fan controller of a similar type was installed. Could it be that this ceiling fan is designed for this type of control, where others are not?

Can someone tell me, as an electrical engineer, exactly what these fan speed controllers do, electrically? How are they different from ordinary triac light dimmers?

\$\endgroup\$
  • \$\begingroup\$ Probably something in the frequency domain, but that's purely speculation... \$\endgroup\$ – Matt Young Dec 4 '13 at 3:50
  • 1
    \$\begingroup\$ Ceiling fans are supposed to have multiple winding taps, like an autotransformer. So the 3-speed switch may just select from the three available taps. From wiki: " They may have multiple taps near one electrical end of the winding, which provides variable speed and power via selection of one tap at a time, as in ceiling fans." \$\endgroup\$ – Li-aung Yip Dec 4 '13 at 5:11
  • 9
    \$\begingroup\$ Note to closers: if "please help me understand the operating principle of X" is not about electrical design, I don't know what is. Maybe I have to phrase the question as "how do I design a fan speed controller?" \$\endgroup\$ – Phil Frost Dec 4 '13 at 12:51
  • \$\begingroup\$ @Li-aungYip I know that the speed controller has only two wires coming out of it, so it's in series with the fan, and couldn't be switching between multiple windings. \$\endgroup\$ – Phil Frost Dec 4 '13 at 12:52
  • \$\begingroup\$ @PhilFrost: can you find a manual for that fan speed controller? \$\endgroup\$ – Li-aung Yip Dec 4 '13 at 12:55
12
\$\begingroup\$

I was recently (today) helping someone with their home renovations, including replacement of the three-speed fan controllers. The old one, which we ripped out and replaced, is shown below. Note the crude circuit diagram on the black box object (the fan speed controller.)

enter image description here

I infer that the fan controller works by inserting a capacitance into the fan's power supply circuit. The slow speed is obtained by using the 4.3uF capacitor (purple, P), the medium speed is obtained by using the 2.1uF capacitor (red, R), and the high speed is obtained by direct connection (no capacitor.)

Here's a circuit diagram for a very similar three-speed motor controller by Clipsal.

enter image description here

I don't understand single-phase induction motors well enough to state with certainly exactly how the extra capacitance modifies the speed of the motor. Sorry!

Note that single-phase induction motors aren't self-starting and must include a phase shifted auxiliary winding to provide starting torque. The phase shift is provided by a capacitor, which may be relevant. (If anyone knows more about the workings of single-phase induction motors, please come forward!)

\$\endgroup\$
  • \$\begingroup\$ I believe the way capacitance modifies the speed of the motor is similar to using a capacitor as a transformerless power supply. For example, see the Microchip document: ww1.microchip.com/downloads/en/appnotes/00954a.pdf Basically, it is using the capacitor to form a high-pass filter, where the cutoff is within the 60Hz AC frequency. \$\endgroup\$ – Zuofu Jan 2 '14 at 21:06
  • 2
    \$\begingroup\$ The fan can be viewed as a complex load. Adding capacitance changes the power factor, changing how much power the fan dissipates as heat. More power dissipated as heat = less power in spinning the motor = slower. That excess heat is what also causes them to start fires/hum/etc \$\endgroup\$ – Steven Goldade Jan 2 '14 at 21:34
  • 1
    \$\begingroup\$ @StevenGoldade: I'm not convinced that's the answer. Changing the power factor shouldn't significantly change the amount of power dissipated as heat - just the amount of power dissipated total. \$\endgroup\$ – Li-aung Yip Jan 2 '14 at 22:42
  • 6
    \$\begingroup\$ Surely it's just the reactance of the capacitor? 4u3F has a reactance of 616 ohms at 60Hz. \$\endgroup\$ – user207421 Jan 3 '14 at 12:30
  • \$\begingroup\$ I think this answer has the slow and medium speeds swapped. The smaller capacitor will have greater impedance, giving a slower speed. \$\endgroup\$ – pericynthion Feb 4 at 3:25
1
\$\begingroup\$

My understanding of light dimmers is that they do not control voltage but cut off the voltage during part of the cycle. A triac is different from a transistor in that it has only off/on states, you can turn it off, but a zero crossing resets it to on. So with a R/C timer you can turn it off sometime after the zero crossing and it will turn on again after the next zero crossing. Thus you are reducing the amount of time per cycle the power is on. This works fine for a light which will glow through the cycle and smooth out the power curve. A motor will vibrate because you are turning it on and off through each cycle and the coils/stator jump and create noise.

I don't know why this would cause a fire. Light dimmers are usually rated at 1500W so way more than the fan motor. But there may be some heat caused by the switching of an inductive load - just guessing, or heat in the motor coils from the irregular wave form?

\$\endgroup\$
  • 2
    \$\begingroup\$ you have your triac action backwards. When a triac is triggered it stays on until the current through it falls below its hold-on current (pretty close to zero). Standard mode of operation is to delay the trigger - the more delay the less on-time - and let the normal zero-cross of the mains turn the triac off at the end of each half-cycle. \$\endgroup\$ – Transistor Nov 29 '15 at 17:13
  • 1
    \$\begingroup\$ In actuality TRAICs can be and often are used for motor drive, and no the motor is not going to "vibrate" back and forth as a result. However, to do so the TRIAC needs to have the inductive kickback of the motor snubbed, and/or be specifically engineered for greater immunity to it than one intended for light-dimmer service would have. \$\endgroup\$ – Chris Stratton Nov 29 '15 at 23:10
1
\$\begingroup\$

There is a voltage drop across the capacitor. The amount of voltage drop is inversely proportional to the value of its capacitance. Using different values of capacitor in series with the fan motor will mean that different voltage levels are applied to the fan motor. The fan motor speed is determined by the applied voltage. Capacitors are used to achieve voltage drop because they don't create heat as would a resistor which would therefore pose a fire risk.

\$\endgroup\$
0
\$\begingroup\$

I am not an electrical engineer but I had this very problem. I was trying to control 2 ceiling mounted regular room fans with a lutron 3 speed box mounted fan controller (which I found later to be a problem because the combined amperage draw exceeded the switch capacity). The fans were multiple speed tap motors with no capacitor. With the fans switched to high and the wall controller to high they worked fine. When I switched the controller to the medium setting (44 VAC), the fans would quit working, also on the low setting (26 VAC). So then I opened up a celing fan to find that it was a single tap fan motor with a capacitor controled with a black box for the speed. So my theory is that a multi speed tap fan motor will need some sort of capacitor inline with the high speed winding circuit (5uf?) in order to operate at the lower voltages of the medium and low switch positions. Without the capacitor the voltage does not have enough balls (for lack of better term) to keep the fan motor turning at reduced voltage. Also the fan controller has a current rating of 1.5A, and a light controller has a current rating of only 600W, I think that is where the fire hazard comes into effect.

\$\endgroup\$
-1
\$\begingroup\$

I studied the controller that came with a Westinghouse fan I just purchased. Only the power lead passes through the controller and each speed setting changed the A/C voltage on the outlet side of the controller as follows:

  1. 120 VAC
  2. 90 VAC
  3. 80 VAC
  4. 70 VAC
  5. 60 VAC

Since A/C motor speed can only be controlled by changing the frequency of the supply, I have to assume that the transformer in the controller was the first stage of power control and the fan has a rectifier and a D/C motor. This is a guess, but it is the only thing that makes sense and it would be a very efficient way to control a fan.

\$\endgroup\$
  • 1
    \$\begingroup\$ The fan motor will be an induction motor. When such a motor is driving a fan a special case exists in that the torque required by the load drops rapidly with rotational speed. The torque of of an induction motor drop when you feed it from a lower voltage. The two effects balance well so it is practical to control the speed of an induction motor driving a fan just by varying the voltage. The motor slip will increase as you drop the voltage. \$\endgroup\$ – Kevin White Apr 9 '15 at 3:33
-2
\$\begingroup\$

Ceiling fans have a higher current rating than lights, and higher current means more heating (heating caused by current is proportional to the square of the current). Fan controllers would be equipped to handle greater amounts of heat, while light dimmers which typically operate at lower currents would burn due to overheating.

As far as device operation is concerned, both fan speed control and light dimming basically involve controlling the amount of energy that is supplied to the load (lights/fan as the case might be). The circuitry of the regulator/dimmer has the overall effect of a switch, whose action can be controlled to regulate the amount of energy supplied to the load.

\$\endgroup\$
  • 1
    \$\begingroup\$ The average ceiling fan's steady state current is lower than a 60W incandescent light bulb. \$\endgroup\$ – Matt Young Dec 21 '13 at 16:35
  • 1
    \$\begingroup\$ I've been around a light dimmer attached to a ceiling fan. Nothing burned out (although the motor sure did make an annoying amount of noise). \$\endgroup\$ – Ignacio Vazquez-Abrams Dec 21 '13 at 19:29

protected by Nick Alexeev Nov 29 '15 at 20:19

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.