Electrician Talk banner
1 - 5 of 5 Posts

·
Registered
Joined
·
259 Posts
Discussion Starter · #1 ·
Hi

Could some one please explain what the (Carrier Frequency) is in a VFD and what it does, all i know so far is that you can play with its value to change the noise of the motor


Thanks in Advance :)


PS

Is there any good book about VFD's that you guys can recommend ?
 

·
Registered
Scada Supervisor
Joined
·
3,612 Posts
Hi

Is there any good book about VFD's that you guys can recommend ?
JRaef posted one for troubleshooting VFD's some time back, I downloaded it. PM with your email and I will send it to you in pdf. It has some very good info.
 
  • Like
Reactions: magmash

·
Registered
Joined
·
438 Posts
For what's its worth, on many drives (well abb primarily) that use direct torque control the carrier frequency is not an adjustable parameter, and it is not a fixed frequency either.

ABB drives using DTC vary the carrier frequency continuously as they see fit. They only switch the IGBTs when needed.
 

·
Registered
Joined
·
5,522 Posts
Here is the very loose interpretation I use in my VFD classes. Purists will argue the deatils and be correct, but this is an easy way for people to understand the basic CONCEPT of what is going on.

Assuming that you know what PWM is, the pulses that make up the PWM output going to the motor are typically depicted as looking like solid blocks of DC in one direction or another, separated by intermittent periods of "off" time, like this:

In reality, each of those pulses is actually made up of thousands of smaller pulses strung together. The rate at which those thousands of smaller pulses is fired at is the "carrier frequency" (that picture alludes to it but no CF is below 1kHz). So in the above picture, that represents 1 cycle of pseudo AC power going to a motor (pseudo means 'fake', because the power is actually DC pulses . The motor ACTS UPON it in a way that looks like AC). It appears that the DC is fired in 6 "blocks" of pulses to make that up, 3 in + and 3 in -. But in reality, at a 10kHz carrier frequency, there would be a little less than 10,000 tinier DC pulses involved, with varying wait-states between blocks of them. The carrier frequency however is the "clock rate" at which those sub-pulses are fired or waited for.

Every time one of those little pulses hits the motor windings, they exert a magnetic field that interacts with the laminated plates that make up the motor stator and rotor cores. Those plates vibrate at that carrier frequency, almost at a molecular level, and it is that vibration that we hear as motor noise. Increasing the frequency, like with any other sound, increases the pitch. So by turning the CF up above 10kHz, you move the pitch out of the range of human hearing (unless you are over 50, in which case VFD driven motors seem to get quieter and quieter...). The noise is still there, you just can no longer hear it, although your dog can.

But the carrier frequency also represents the switching losses inside of the transistors of the VFD doing the firing of those pulses. So the faster you switch them, the more losses the transistors see and the hotter they get. Most VFD mfrs will instruct you to de-rate the drive at high carrier frequencies as a result. In addition, the switching rate also contributes to other phenomenons in motor circuit such as reflected (standing) wave generation in the conductors, and capacitive coupling between the stator and rotor that can lead to bearing EDM damage. So turning up the CF to get rid of noise might be doing more harm than good. It should be avoided unless the issue is critical and if so, understand and mitigate the consequences.
 
1 - 5 of 5 Posts
Top