Harmonics,what are they?


by RozzerETO
Tags: harmonics
RozzerETO
RozzerETO is offline
#1
Feb20-13, 02:15 PM
P: 2
so, I've been doing some research on harmonics recently and i can't actually find a good explanation of how they are created or what the effect. I have a feeling that they are something to do with 'clipping' the sinusoidal waveform to get the desired voltage. VSDs?
Phys.Org News Partner Engineering news on Phys.org
PsiKick's batteryless sensors poised for coming 'Internet of things'
Researcher launches successful tech start-up to help the blind
Researchers propose network-based evaluation tool to assess relief operations feasibility
Windadct
Windadct is offline
#2
Feb20-13, 04:23 PM
P: 532
In their purest mathematical sense Harmonics are signals or part of the waveform at multiples of the fundamental frequency. Basically ANY distortion of the pure sinusoid - will result in harmonics. ( It is interesting / important to note - that ALL signals can be broken down into a sum of wave forms - all of which are multiples of some fundamental frequency - for noise this is true within a specific bandwith....)
For VFDs - one important measure of the quality of the drive is the THD - Total Harmoinic Distortion. Ideally this should not look like clipping - but I am sure some drives end up with a waveform that looks clipped.
The typical effect of harmonics are overheating of magnetics ( Motors, Transformers etc) for "low order" harmonics, that is Harmonics between 2nd and 12th my 20th Harmonic. Another issue, more common with higher order ( Starting around the 12th upward), is increased voltage stress on the conductors and insulation systems ( also in the transformers and motors - etc). They can also cause interference with other devices ( EMI ) due to the power / frequency levels involved.
MATLABdude
MATLABdude is offline
#3
Feb21-13, 02:10 AM
Sci Advisor
P: 1,724
I don't know your level of knowledge, but if you're asking from a signal analysis or mathematical point of view, any non-sinusoidal waveform will have harmonics because it's non-sinusoidal. As Windadct alludes to, in order to create that non-sinusoidal waveform, you need the contribution of higher-power sinusoids (determined via Fourier Analysis):


I just wanted to post to share the animation, which probably comes from this site:
http://blog.matthen.com/post/4211270...es-can-be-used

Windadct
Windadct is offline
#4
Feb21-13, 11:32 AM
P: 532

Harmonics,what are they?


Sorry but I have to disagree with lolone's whole post. Harmonics can be completly understood with out this info. Signal theory only, and photons- what?
f95toli
f95toli is offline
#5
Feb21-13, 11:36 AM
Sci Advisor
PF Gold
f95toli's Avatar
P: 2,197
Yes, please ignore lolone's post.
Harmonics is a mathematical concept, you don't need any physics to understand it.

Harmonics specifically refers to multiplies of a fundamental (for a 50 Hz system the harmonics would be 100, 150, 200 Hz etc), not just any frequency as Iolone's post would suggest.
Mordred
Mordred is online now
#6
Feb21-13, 11:50 AM
PF Gold
Mordred's Avatar
P: 1,512
Here is a decent article on harmonics with regards to VFD's and THD.

http://literature.rockwellautomation...p011_-en-p.pdf

http://support.fluke.com/find-sales/...03_ENG_K_W.PDF

the second site has some handy DMM tests for harmonics although those tests won't tell you what harmonic number is involved only that harmonics are present.
RozzerETO
RozzerETO is offline
#7
Feb21-13, 12:46 PM
P: 2
thanks very much for all the replies, they have helped me grasp the subject! in reply to what my depth of knowledge is, I am currently doing a Marine electrical engineering degree! thanks very much for all your help!
lolone
lolone is offline
#8
Feb21-13, 01:15 PM
P: 8
Quote Quote by f95toli View Post
Yes, please ignore lolone's post.
Harmonics is a mathematical concept, you don't need any physics to understand it.

Harmonics specifically refers to multiplies of a fundamental (for a 50 Hz system the harmonics would be 100, 150, 200 Hz etc), not just any frequency as Iolone's post would suggest.
I have messaged the moderator to restore the post. Harmonics is not a mathematical concept. Mathematics is used to describe a physical system, therefore harmonics is a product of the physics.

This should be obvious since you are dealing with charge carriers and photons. This is what electricity is and you cannot measure something that is not a product of these quantum particles.

I'm sure as engineers you have mathematical descriptions of what is happening, but in order to understand it fully everyone needs to understand the physics.
f95toli
f95toli is offline
#9
Feb21-13, 01:43 PM
Sci Advisor
PF Gold
f95toli's Avatar
P: 2,197
Quote Quote by lolone View Post
I'm sure as engineers you have mathematical descriptions of what is happening, but in order to understand it fully everyone needs to understand the physics.
Firstly. I am not an engineer, I am an experimental physicist.
Secondly, harmonics is a mathematical concept (from Fourier analysis) applicable to a very wide range of phenomena, some electrical some not. You can have harmonics in any driven non-linear system, a mechanical system would be an obvious system.
Hence, trying to explain it by refering to details of charge transport (which btw were incorrect) is just silly
lolone
lolone is offline
#10
Feb21-13, 01:54 PM
P: 8
Quote Quote by f95toli View Post
Firstly. I am not an engineer, I am an experimental physicist.
Secondly, harmonics is a mathematical concept (from Fourier analysis) applicable to a very wide range of phenomena, some electrical some not. You can have harmonics in any driven non-linear system, a mechanical system would be an obvious system.
Hence, trying to explain it by refering to details of charge transport (which btw were incorrect) is just silly
Finally, a physicist. Then you should really get my point here. Fourier analysis is just a transform, we are no longer looking the group (or time domain) activity, but the individual frequencies (or rates of change in direction of charged particles) that make up that group activity.

So, in the Fourier analysis, we are not looking at purely mathematical concept, we are looking at how the system really is, albeit with limited resolution. It is the "time domain" that is the mathematical concept.

In terms of describing what harmonics is, its best to look at it in terms of collisions and particle exchanges.
lolone
lolone is offline
#11
Feb21-13, 01:59 PM
P: 8
Actually, let me put that a little bit better:

Time Domain - Voltage - is a summation of group activity
Frequency Domain - Factorization of the voltage

In terms of harmonics, we are interested in the factorization of the voltage as it is these interactions that give rise to harmonics.
berkeman
berkeman is offline
#12
Feb21-13, 04:32 PM
Mentor
berkeman's Avatar
P: 39,612
Thread closed for Moderation...
berkeman
berkeman is offline
#13
Feb21-13, 04:53 PM
Mentor
berkeman's Avatar
P: 39,612
This thread will remain closed.

The OP has gotten good information about his question, and there is too much misinformation in lolone's posts to effectively clean up the thread.

lolone -- check your PMs.


Register to reply

Related Discussions
3rd and 5th harmonics Engineering, Comp Sci, & Technology Homework 3
Harmonics Introductory Physics Homework 1
Odd harmonics Electrical Engineering 5
why harmonics cause such voltage drops? Electrical Engineering 7
harmonics are integer multiples of a fundamental frequency General Physics 12