Harmonic and Total Harmonic Distortion 1

[shenyee]

Anyone here can suggest to me how to measure harmonic and total harmonic distortion from a induction motor controller? I can see the distorted waveform through the oscilloscope and the FFT.
But the thing I required must be very detail, for example, the value of 3rd harm, 5th harm, 7th harm and other lower order harm. Any equipment or method can use for this purpose?
Looking forward for your suggestion and idea!

 

[alehman]

http://www.dranetz-bmi.com/

http://www.astro-med.com/products/products.html

If you have the FFT of a waveform, it is easy to calculate THD.

THD: the ratio of (a) the sum of the powers of all harmonic frequencies above the fundamental frequency to (b) the power of the fundamental frequency. Note: The THD is usually expressed in dB

 

[oswaldo100]

you can measure the harmonic distortion with any instrument that perform such measurement.In theory a signal, or wave can be decoupled in harmonmic components, when, all of them got multiple frecuencies, e.g. 60 , 120 ,180, 300 hz... with differents RMS values, this values can be founded by the fourier transform, but you nedd the mathematic function of the wave.

The THD is defined by the following formula:
THD = IH/If

where If : RMS value of the fundamental frecuency
IH = (I2^2 + I3^2 + I5^2 +...In^2)^0.5

I2 , I3, I5, are RMS values of the individual components founded by the fourier analysis...

I Hope it can help you a little

Oswaldo








where terms 2..N are the power levels of the harmonics and term 1 is the power level of the fundamental (the pure tone).

The THD+N is defined by the following formula:

 

[nerella]

For measuring the Harmonics in details my instruments available. Some of the manufacturere for this are Fluke, tektranics, Philips. Fluke 41-b is one instrument from Fluke. It gives the results in detail with respect to their harmonic frequency. It gives the results upto 31st(may be more than this) Harmonic. In this instrument there is provision for storing the waveforms, copying the data to PC and prints can be taken from the PC.

 

[ScottyUK]

Probably the best power analysers are made by Voltech Instruments, also some excellent products are available from Yokogawa. The products from both companies are expensive.

If you aren't doing this sort of work every day, hire the best analyser you can find, rather than buy a cheap one which may give you misleading results.

A poorer alternative to a power analyser might be to use a spectrum analyser with extended low frequency response, which will give you amplitude of individual harmonics, but not relative phase. Make sure the spectrum anz. has good resolution at the frequencies of interest

 

[pragmatist]

There are several electronic power meters which measure harmonics and power. Look into Fluke's type 43B power quality analyzer for $2199.75. It measures up to the 51st harmonic. Page 2015 in Newark Electronics catalog #121.

http://www.newarkinone.com.

Amprobe has the HA-2000 power analyzer for $1048.95. It measures harmonics up to the 31st. Page 2021 in Newark's catalog.

Contact the manufacturer by telephone or on their web site for further information.

pragmatist

 

[TasDevil]

It would be wise to engage specialists to do it, you could hire technician with a power quality meter for less than $500 a day. Some of instruments may be difficult to operate and later extract results. Check local office of Sqare D for a price to do it. You may even ask short report in addtion of graphs and tabulated results.

 

[shenyee]

But do anyone have an economic method for measuring harmonic? Because my lab doesn't have the equipment by the way I'm short of time.

 

[jbartos]

Suggestion: If the Fourier Series analysis poses no problem, and if there is a good skill in applied mathematics, try this approach. Calculate the Fourier series coefficients from the waveform that you see on the oscilloscope screen.

 

[jghrist]

Your first post mentioned FFT. Do you have the results of a Fast Fourier Transform (FFT) analysis? The FFT results are the magnitudes of the harmonic orders. If your oscilloscope can output discrete data, you can perform a FFT to get the harmonic data. If not, you can analyze the waveform as seen on the oscilloscope by classical Fourier Series analysis as suggested by jbartos.

 

[busbar]

shenyee, I mean no personal offense, but you state “I required must be very detail” and then “I'm short of time”. Contact a firm that is experienced in power-quality measurements, and pay them for hourly or daily work. It is likely that besides the necessary instrumentation, they have report-preparation software to offer data in tabular or graphical form. Suggested may be a NETA-member firm, listed at

http://www.netaworld.org/memberlist.cfm

 

[NNG]

Please, take 4 steps:
1- load a free software "Top-the output process" from electrotec web site:

http://www.pqsoft.com/top/

2- save the current or voltage waveforms on your scope in *.CSV format file
3- open this file in above application "Top-process..."
4- you will be available to plot the grapths and tables look-up for all harmonics presented in your measured waveforms from fundamental up to 100's.

Good luck.
p.s. I think that all of the modern scopes have a capability to save the data in *.CSV format.

 

[shenyee]

Hi NNG, I have look for the sotware that you mention. I am using Tektronix TDS3054B oscilloscope, but I don't how to save my voltage or current output in .csv format. Or may be this type of oscilloscope not suitable for the TOP software.
Anyone can help?

 

[jghrist]

The product description in

http://www.navicpmart.com/estore/pr...spec.cfm~~DIGOSC~~TEK~~TDS3054B~~~~ |8~~.html

shows that this scope has FFT analysis. This analysis will give you the harmonics data that you need without any additional hardware or software. You need to review the instruction book to figure out how to use this feature.

 

[electricpete]

As I recall there are two different definitions of THD, depending on whether you look at US standards or international standards.

 

[NNG]

Hi, Shenyee.
I have checked the Tektronix site for information about hard copy capabilty (saving the data in *.CSV format)on your scope TDS3054B and could find any reference to this option, call to Tekt-x for support or otherwise:
1- try to look through the scope menu ("Save/Recall" buttom) if you can to save in this format. On some models they have an option - Speadsheet format or *.CSV (the same formats).
2- Tektr-x offers the software package for a communication between PC and your scope. The "Wavestar" program has a lot off fitures to explore the measured data on your scope but it's also expencive ($400.0). I didn't try "the Open Choice" - free for download but may be it will be useful too.
3 - You have to install the communication card between PC and scope to operate those software programs.

My opinion, you have to get your scope data in numerical format and then plug it into Top-process in-less expencive way.
Thanks, NNG.

 

[NNG]

Hi Shenyee,
looks you can do it with your scope TDS3054B because I've checked one of the scopes in our lab - TDS3014B. It has the same functions as yours and it's from the same Tektronix family: TDS3000.
Here is what you want to do:
1- Select the Ch.(Voltage or Current) on your scope,
2- Push the button "Save/Recall" on main panel,
3- On horizontal menu of Display you'll have option to "save" (#4 button),
4- Select this option,
5- On vertical menu of Display you 'll have option to "file" (#1 button),
6- Select this option and then next screen will come,
7- On vertical menu of Display you'll have three options,
8- Chose the "SpreadSheet format" this is a *.CSV file,
9- Execute this option and scope will save numerical data of your voltage or current waveform in *.CSV format on floppy disk.
10- Open the "Top -Output Process" and "Converting" programs from Electrotec,
11- First convert your *.CSV file to Output Process file in "Converting" program.
12- In "Top-Output Process" use option to plot the Grapth,
you will see the same waveform what you captured on scope.
13- In "Top- Output Process" use option to make a Table.
14- This program calculated all of harmonic stuff what you need.
In my expirience the "Top-Output Process" with data from the regular scope very compatible with those expencive methods:Yokogawa PowerMeters, Fluke 41 and 43, HP Spectrum Analyzers .
You will not spend a cent for this try except your labor.
Thanks, NNG.

 

[shenyee]

Hi NNG,
I follow the steps you mention above to save the voltage/current waveform from oscilloscope in the extension of *.CSV. The problem now is I cannot convert or open it in the TOP software. During my conversion, there is a message coming say, there is a missing value around line one. So can you give further detail about that?
Thank you!

 

[NNG]

Hi Shenyee,
you have to convert your *.csv file first. To make it easy follow by these steps:
1. Open your *.csv file in Excel (MS office program).
2. On top of spreadsheet you will see in first 2 colomns designation for your measured data
[DB].Data.Waveforms.CH 1
S Volts
-0.02048 0
-0.020476 0
where s-time, volts- value of measured data.
3. Delete those columns and leave only numbers in your spreadsheet.
4. Save again in *.csv and close the file.
5. Open Top Converter.
6. In Top Box "Input File" add your file.
7. Name your Output File, for example "test1"
8. In bottom screen leave an upper line for the time values and then selected your measured parameter - voltage or current with proper settings.
9. Execute "convert" button and program will convert your data.
10. Open "Top Processor" and make selection for the Graph.
11. Program will built the plot identical to your scope picture, do not forget about adjusting your time axis.
12. Under option "Table" in "Top Processor" you can select
any harmonic analysis data regarding your measured waveform.
Also check the Help file for any additional information.
Thanks, NNG.