# Getting problem in triac based fan speed control

I'm using this circuit to control the speed of ceiling fan. The fan speed is divided into 8 steps. (0:off & 8:full speed) with 100us gate pulse.

This circuit works with incancescent light bulb perfectly. But having some issues with FAN.

• if I increase the speed of fan step by step with 1.5 s delay of each step (12 s full speed time) fan works fine at full speed.

• If I start fan at 5th step instantly fan rotate at its speed, and increase each step with required delay of1.5 s fan works fine.

• If I start fan at 5th step instantly fan rotate at its speed, and instantly put fan at 8th step(full speed) fan stops with some humming sound.

• If I start fan at full speed(8th step) instantly, fan does not work. It gives humming sound.

This concludes that if the fan has required torque on current step it works on next step otherwise not. But for first 5 steps it always works fine.

• what is preventing you from ramping up speed? Apr 5, 2021 at 21:16
• 100 mks too short Triac has minimum gate current which comes from main voltage across opened triac and resistor. Make pups 500 mks Apr 5, 2021 at 22:58

Why only 100us pulses?

It sounds like the issue is that when you start at full speed with a 100us pulse at zero crossing there is no AC voltage present - as a result even though there is a trigger the triac does not start conducting.

If you start at a lower speed, eg 50%, there will be significant AC voltage across the triac so it starts conducting and continues until the current drops to zero.

Once started at low speed, even if you change to triggering for full speed with the trigger at zero crossing there will still be current flowing because of the inductance of the motor (it could be as much as 90 deg out of phase).

I think the solution would be to apply much longer gate trigger pulses; maybe just a couple of milliseconds so that the trigger is still present when the voltage rises to the point that the triac can trigger.

• Yes possible reason "Once started at low speed, even if you change to triggering for full speed with the trigger at zero crossing there will still be current flowing because of the inductance of the motor (it could be as much as 90 deg out of phase)." Apr 5, 2021 at 21:52
• ceiling fans are typically noisy from the current/torque pulses vibrating the light weight unsecured structure from winding vibrations

• the only good way is by various sizes of film capacitors which you will find many answers showing examples on this site.

• disregard any Triac solutions, unless you like loud obtrusive hum

• Most discrete ceiling fan controllers are actually tapped inductors. The TRIAC has its shortcoming (current harmonics first of all) but all but obsolete, for small load is still a useful solution Apr 6, 2021 at 7:09