Crest Factor & Harmonic Distortion

Status
Not open for further replies.
GoldDigger said:
Since you are looking at the supply voltage from the grid (yes?), it is possible that the bigger generators of third harmonics along with other higher odd harmonics (such as rectifiers and switching mode power supplies) have a third-harmonic-suppressing distribution transformer or two between them and you. That leaves the fifth harmonic, even though it may have been lower in amplitude at the source, to be dominant.

Plus what Besoeker said. :)
Click to expand...

I am two transformers from the grid. That sine wave is about 12800 volts tall. A substation transformer ( inside ) then through a 30KVA transformer. There is metering at the service. I have not been able to fully access it yet. I can see the volts, amps, PF, THD, through my laptop at any time while at work. I cannot get down into the menus and see the % of each order. That would also allow me to open the incoming breakers !!! I do not even trust myself that much. Lights out, shift is over, everybody go home.

I do not think any third harmonic transformers exist, unless recently installed with specific equipment. I can check on that. The facility is about 35 years old. At the time of construction steel & copper fabrication was not part of the plan.
 
gar said:
130415-0841 EDT

SG-1:

To get a reasonably clean sine wave from my 60 Hz power I used 5000 ohms and 3 ufd as a filter, 3 db point 10 Hz. Then I read the Beckman RMS and Fluke 27 meters simultaneously at the filter output. The input to filter was in the 10 V range. The capacitors were polystyrene which are a very low dissipation factor device.

With a filter like this and the parallel measurement you can see if there is a calibration difference between the meters with a pure sine wave input. If the two meters are different, then the ratio of the readings can be applied to your direct line measurements to see if the difference in readings is a result of the waveform distortion, or just because of the scaling difference.

The tests with and without the filter should be run at about the same voltage. You do not need polystyrene capacitors, just good paper, Mylar, or polypropylene.

.
Click to expand...

This is a high pass filter ? Switch the resistor & cap positions form the low pass filter.

What is really sad, I can generate a pure sine wave at home. I am pretty sure that ability has been wiped out at work, since the calibration services was farmed out.
 
130515-2355 EDT

SG-1:

Capacitor across meter input and the resistor in series. A 0.5 ufd might have been a better choice or cascading several stages. But I got the result I wanted by reducing the harmonics relative to the fundamental. A high Q LC at resonance would be better.

.
 
gar said:
130515-2355 EDT

SG-1:

Capacitor across meter input and the resistor in series. A 0.5 ufd might have been a better choice or cascading several stages. But I got the result I wanted by reducing the harmonics relative to the fundamental. A high Q LC at resonance would be better.

.
Click to expand...

Ok, thanks, The roll off at 10Hz confused me. I was too busy tonight do anything on this project. I still had the ION meter to watch for a while & the harmonics looked the same as before.
 
I was able to construct a low pass filter tonight using a 7500 ohm resistor in series with a .22 mF capacitor. This should produce 95 Hz roll off frequency. First picture is without filtering the meters are reading .8 volts apart at 100volts. The crest factor is 1.4. The second picture with filtering to remove the harmonics above 95 Hz the two meters are reading .1 volts apart at 100 volts. This shows the measurement difference between the two meters in a previous post was caused by harmonic distortion & not a calibration problem.
 

Attachments

  • WithoutFilter.jpg
    WithoutFilter.jpg
    66.5 KB · Views: 0
  • WithFilter1.jpg
    WithFilter1.jpg
    67.6 KB · Views: 0
SG-1:
Since you have harmonic problems, I think you need to use a true RMS meter for measurement of test voltage. Also, check that the crest factor of the test voltage is within [(square root 2) plus or minus 5%]. Then the peak of the applied test voltage= Its RMS value/1.414 for your testing purpose.
 
Sahib said:
SG-1:
Since you have harmonic problems, I think you need to use a true RMS meter for measurement of test voltage.....
Click to expand...
The Fluke 289 is a true RMS meter.

From the Fluke website:

Fluke 289 True-rms Industrial Logging Multimeter with TrendCapture

From SG's post #30:

My next step is to verify the Fluke 289 reading by other means.
Click to expand...
 
130520-0742 EDT

Sahib:

I disagree with this statement
Since you have harmonic problems, I think you need to use a true RMS meter for measurement of test voltage. Also, check that the crest factor of the test voltage is within [(square root 2) plus or minus 5%]. Then the peak of the applied test voltage= Its RMS value/1.414 for your testing purpose.
Click to expand...
If the peak voltage is the most important criteria for SG-1's test, then he needs a peak reading meter.

A flattened sine wave, which is what SG-1 and all the rest of us have, will have a smaller peak value than predicted from 1.414 * the RMS voltage of that waveform. For example on a square wave, a very flattened sine wave, the peak value equals the RMS value. Also I believe you meant to say RMS value * 1.414 .

.
 
gar said:
If the peak voltage is the most important criteria for SG-1's test, then he needs a peak reading meter..
Click to expand...

One of the most important criteria for SG-1's high voltage test is the wave shape must be nearly pure sinusoidal with both half-cycles closely alike.

gar said:
A flattened sine wave, which is what SG-1 and all the rest of us have, will have a smaller peak value than predicted from 1.414 * the RMS voltage of that waveform. For example on a square wave, a very flattened sine wave, the peak value equals the RMS value..
Click to expand...

The results of a high-voltage test are thought to be unaffected by small deviations from a sinusoid if the ratio of peak-to-r.m.s. values equals square root(2) plus or minus 5%, a requirement which can be assumed to be met if the r.m.s. value of the harmonics does not exceed 5 per cent of the r.m.s. value of the fundamental.

gar said:
I believe you meant to say RMS value * 1.414 .
.
Click to expand...

You are correct. It was a typo. The nominal value of the test voltage is defined by its peak value divided by square root of 2.
 
gar said:
Eor my flattened sine wave there appears to be around a 1 V lower peak than predicted from 122.6 V RMS * 1.414 .

.
Click to expand...
It does not matter for SG-1's high voltage test, if its crest factor is within square root of 2 plus or minus 5%.
 
Tonights Sine Waves

Tonights Sine Waves

Some pictures from a PowerXpert meter. Besides the crest factor being within 5% of the square root of two, the wave is considered acceptable if the total harmonic distortion is 5% or less. Tonights sine wave, at least at the time of capture, is well in spec.
 

Attachments

  • X-Pert_1.jpg
    X-Pert_1.jpg
    76.2 KB · Views: 0
  • Xpert_2.jpg
    Xpert_2.jpg
    74 KB · Views: 0
  • Xpert_3.jpg
    Xpert_3.jpg
    68.2 KB · Views: 0
  • Xpert_4.jpg
    Xpert_4.jpg
    59.7 KB · Views: 0
SG-1 said:
Besides the crest factor being within 5% of the square root of two, the wave is considered acceptable if the total harmonic distortion is 5% or less.
Click to expand...

Finally the answer to the question you posed in the OP:

If the crest factor is within 5% of the square root of two, the total harmonic distortion would be within 5% for a nearly pure sinusoidal wave.

This answer is based on page no 30 para 4 of 'High voltage Engineering' in http://bib.convdocs.org/v4887/?download=1
 
Sahib, it is really the other way round per your web reference:

If the total harmonic distortion is within 5%, the crest factor would be within 5% of the square root of two for a nearly pure sinusoidal wave.

So,unfortunately, the total harmonic distortion can not be determined from a given crest factor.

So the answer to the OP's question:no.

SG-1, the crest factor criterion is important: if the crest factor is meeting the spec, but the harmonic content is not (more than 5%), the sine voltage wave is still fit for testing purpose (This follows from above web reference).
 
Sahib said:
Finally the answer to the question you posed in the OP:

If the crest factor is within 5% of the square root of two, the total harmonic distortion would be within 5% for a nearly pure sinusoidal wave.

This answer is based on page no 30 para 4 of 'High voltage Engineering' in http://bib.convdocs.org/v4887/?download=1
Click to expand...

Besoeker actually answered the origional question in post 2. There was a great deal of useful information exchanged in the following posts. I thank all for helping. I still need to continue to monitor our waveform.

I found a later version of Std-4 & the 5% harmonic distortion had been added to the crest factor criteria.
 
Status
Not open for further replies.
Top