Noise free AC fan control using MOSFET

Thread Starter

alpesh_borad

Joined Apr 21, 2017
8
Hi,
I am designing AC FAN Control Circuit using MOSFET.
During Testing I observe the humming noise in FAN Speed and I try different method but noise is not removed.
Initially when FAN is at Speed 1 only noise observed and FAN is in STOP condition then increase to Speed 2 then slightly start the speed with noise and gradually speed increase and noise removing with Speed Increasing.

Can Anyone help me to remove the humming Noise from the FAN?

Please check the Circuit in attachment which I use for the same.

Regards,
Alpesh
 

Attachments

Alec_t

Joined Sep 17, 2013
11,997
Welcome to AAC!
I assume the ground symbol on the mains side of the circuit doesn't represent earth ground? If it did, then Line voltage would be shorted to ground when the FETs conduct.
If your PWM frequency is in the audible range then you will get hum/whine from the motor winding.
Since the fan motor is highly inductive I'm surprised the FETs have survived back-emf spikes.
Why not use conventional phase-angle control with a triac (as per off-the-shelf lamp dimmers)?
 

DickCappels

Joined Aug 21, 2008
7,248
How fast is your PWM and what is the frequency of the acoustic buzzing sound?

You might have to move your PWM up to past the range of normal hearing. That means 20 kHz or more, and that might take a faster driver and MOSFETs with less gate capacitance. If you have pet rodents or dogs they aren't going to like it so then you would have to insert a low pass filter after the PWM stage.Things can become complicated quickly.

It is understood that lamp dimmers make fans very noisy and variacs give good quite control but are expensive and bulky. It would be very nice if you can get your MOSFET PWM to work quietly.
 

Dodgydave

Joined Jun 22, 2012
9,936
You're using two Nfets, which means they're both either on or off at the same time, which is pointless! why don't you use a Triac and opto like normal....
 

DickCappels

Joined Aug 21, 2008
7,248
Good catch Dodgydave!

That means that there will always be a harmonic rich line frequency component to the waveform, hence there will be buzzing.
 

AlbertHall

Joined Jun 4, 2014
11,212
You're using two Nfets, which means they're both either on or off at the same time, which is pointless! why don't you use a Triac and opto like normal....
While they are both on the power is connected to the fan. While they are both off there is no power to the fan. Exactly as required.
 

Thread Starter

alpesh_borad

Joined Apr 21, 2017
8
Welcome to AAC!
I assume the ground symbol on the mains side of the circuit doesn't represent earth ground? If it did, then Line voltage would be shorted to ground when the FETs conduct.
If your PWM frequency is in the audible range then you will get hum/whine from the motor winding.
Since the fan motor is highly inductive I'm surprised the FETs have survived back-emf spikes.
Why not use conventional phase-angle control with a triac (as per off-the-shelf lamp dimmers)?
Hi,

Thanks.

Here Line Ground is not ETH.

First start with 10 KHz Frequency and observe the Humming Noise then change it to 100 Hz but the result is the same.

With using TRIAC, found the Humming Noise and also need the accuracy in Speed controlling so move to MOSFET or IGBT.

I also study and found that "Give the Gate pulse in gradually increasing manner to the MOSFET so it will operate gradually and reduce the Noise" I try with RC filter but there is no change in result.
 

Thread Starter

alpesh_borad

Joined Apr 21, 2017
8
Hi,

As mentioned the frequency use for PWM is 10 KHz and 100Hz and in both case noise observed.

If I am increasing the Frequecy more than 10KHz noise is gradually decrease but FAN is not start.

Please provide any suggested circuit for the 20KHz PWM operation.
 

Thread Starter

alpesh_borad

Joined Apr 21, 2017
8
Hi,

Thanks.

Actually i don't know the frequency of Noise and also don't know to measure it?

Please provide guidance to measure.

I have given PWM of 100Hz of Frequecy.
 

Kermit2

Joined Feb 5, 2010
4,162
You could have a mechanical issue such as a motor winding that has come loose from its varnish and is moving.
This happens in transformers as well and makes the "hum".
 

MaxHeadRoom

Joined Jul 18, 2013
22,823
Can Anyone help me to remove the humming Noise from the FAN?
I am working on a μP triac firing circuit right now, but when done I want to see if a fan can be quietened using Triac firing but in a burst fashion rather than phase angle control, as this is what causes the buzz in typical off the shelf fan motor triac controllers.
 

Alec_t

Joined Sep 17, 2013
11,997
I want to see if a fan can be quietened using Triac firing but in a burst fashion rather than phase angle control
I was wondering if that would work. Any vibration might then be sub-audio. May work better on a large (heavy) fan rather than a small light-weight one. Interested to see the results, Max.
 

AnalogKid

Joined Aug 1, 2013
9,157
I've been puttering around with a phase-control dimmer with MOSFETs instead of a TRIAC, to make a dimmer with total power dissipation under 1 W at all brightnesses with a 200 W load. That means no TRIAC and no bridge. There is nothing wrong with the approach in post #1, and variations have been patented in the past. As long as the driver power is galvanically isolated from the line (and his is), two FETs work very well. You need two for the back-to-back body diode connection. But be aware that this thing will generate *lots* of RF noise, especially when driving an inductive load.

ak
 

Thread Starter

alpesh_borad

Joined Apr 21, 2017
8
I've been puttering around with a phase-control dimmer with MOSFETs instead of a TRIAC, to make a dimmer with total power dissipation under 1 W at all brightnesses with a 200 W load. That means no TRIAC and no bridge. There is nothing wrong with the approach in post #1, and variations have been patented in the past. As long as the driver power is galvanically isolated from the line (and his is), two FETs work very well. You need two for the back-to-back body diode connection. But be aware that this thing will generate *lots* of RF noise, especially when driving an inductive load.

ak
Hi,

Thanks.

This means that If I am controlling AC FAN using MOSFET, it create noise.

If there is any method to remove the noise then suggest it.

Thanks.
 
Top