Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Harmonic Filter Design

Status
Not open for further replies.

karalahana

Electrical
Feb 24, 2010
52
Hi, I want to learn about the design of harmonic filters. I read through some info about it but unfortunately I could not get the logic of the sizing of the filter. Say I have a system dominated by 7. harmonic current. I want to learn two things.
Firstly, is there any relation between the order of harmonic current and its magnitude or is it possible that a 7. harmonic is half the first order wave?
Secondly, I want to learn to what degree I compensate the capacitor with an inductance(filter) if I am to filter the 7. harmonic totally out of the system? Do we set an assumption and and take some part of the harmonic current out of the system? Namely what criteria we use to find the frequency of resonance?
I think you could show me the way. thanks...
 
Replies continue below

Recommended for you

A common rule of thumb for a 6 pulse (6 diode) three phase rectifier is that the magnitude of the harmonic is equivalent to the inverse of the harmonic order (5th =20%h =14%, 11th =9%,...). That's crude though and depends on other circuit components like source impedance, bus capacitance, voltage balance and any other inductances present. I don;t think you will be able to completely eliminate any specific harmonc with a passive filter. You can dramtically reduce it by targeting the parallel LC branch to very close to the resonant frequency (420hz in this case). Start with 10 to 15 % KVAR of capacitance and convert to C and then solve for L. (420 = 1/(2*PI*sqrt(L*C)). This is where it gets tricky. How much C to start with. You want as much as possible without pushing the system to leading PF. Bet to model in a circuit simulator. Keep in mind that harmonics can come from both directions. That's why manufacturers use a second inductor in one side of the paralel branch in the line to detune the filter (move the resonant freq to a harmless value for that side of the circuit).

Neil
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor