60 Hz waveform distortion

Status
Not open for further replies.

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
160104-2137 EST

How different is your 60 Hz waveform distortion compared to my plots?

My plots probably imply capacitor input power supply loading that starts before the sine wave peak. Possibly two slightly different starting points for different types of power supplies.

Then some big load occurs just after the peak. This is probably some other type of load. Looks to be close to 120 deg from the zero crossing.

Although I have three phase on the pole I don't have it in the house. Thus, I don't know how the other two phases look here.

Full view of the sine wave:
160104-1719-60 Hz-Distortion-400.jpg

Enlarged view near the sine wave positive peak:
160104-1950-60-Hz-Distortion-Enlarged-400.jpg

Both positive and negative peaks are quite similiar.

.
 

Besoeker

Senior Member
Location
UK
160104-2137 EST

How different is your 60 Hz waveform distortion compared to my plots?

My plots probably imply capacitor input power supply loading that starts before the sine wave peak. Possibly two slightly different starting points for different types of power supplies.

Then some big load occurs just after the peak. This is probably some other type of load. Looks to be close to 120 deg from the zero crossing.

Although I have three phase on the pole I don't have it in the house. Thus, I don't know how the other two phases look here.

Full view of the sine wave:
View attachment 14080

Enlarged view near the sine wave positive peak:
View attachment 14081

Both positive and negative peaks are quite similiar.

.


Well, I have 50Hz.
But the distortion is evident.

HomeVrev00_zpsf7551ef6.jpg
 

mgookin

Senior Member
Location
Fort Myers, FL
160104-2137 EST

How different is your 60 Hz waveform distortion compared to my plots?

My plots probably imply capacitor input power supply loading that starts before the sine wave peak. Possibly two slightly different starting points for different types of power supplies.

Then some big load occurs just after the peak. This is probably some other type of load. Looks to be close to 120 deg from the zero crossing.

Although I have three phase on the pole I don't have it in the house. Thus, I don't know how the other two phases look here.

Full view of the sine wave:
View attachment 14080

Enlarged view near the sine wave positive peak:
View attachment 14081

Both positive and negative peaks are quite similiar.

.

Is that a one time event or consistent?
 

Besoeker

Senior Member
Location
UK
Dranetz used to have a software program that you could upload an oscillograph and it would make suggestions about potential causes based on the "signature".

I think this is the link http://www.dranetz.com/product-services/power-quality-software/
I can download scope data to an Excel spreadsheet. I have written a programme to exctract harmonics using the FFT function. Unfortunately, it's on my other laptop which got baptised with half a glass of Merlot...........
 

gadfly56

Senior Member
Location
New Jersey
Occupation
Professional Engineer, Fire & Life Safety
I can download scope data to an Excel spreadsheet. I have written a programme to exctract harmonics using the FFT function. Unfortunately, it's on my other laptop which got baptised with half a glass of Merlot...........

A bad outcome for both the laptop and the Merlot. :eek:hmy:
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
169105-1505 EST

Besoeker:

Your waveform has a nice shape that I would consider to be mostly from similar capacitor input filter power supplies.

We have probably over 100,000 computers within about a 5 mile radius, many MRIs, and lots of various research projects. These would come from over 50,000 students, and 3 major hospitals. To my knowledge there are no big electric are furnaces closer than about 25 miles, and I don't know what their current waveform looks like. At one time the Ford Rouge plant consumed more electrical power than the combined load of Detroit and several other cities. But that won't be true today.


mgookin:

Yes, the waveform distortion is consistent. I just looked at it and it is similar to the previous photos. The distortion is not from anything in my home.

Other analysis on my AC waveform shows high frequency noise (basically an impulse shock exciting a resonant circuit, damped sine wave) resulting from one or more dimmers when they are on. This signal is greatest on the phase the dimmer is on as would be expected. The dimmer turn-on point jumps around a little. If I have the TED system on, then I can see its 125 kHz signal.

.
 

mgookin

Senior Member
Location
Fort Myers, FL
169105-1505 EST

Besoeker:

Your waveform has a nice shape that I would consider to be mostly from similar capacitor input filter power supplies.

We have probably over 100,000 computers within about a 5 mile radius, many MRIs, and lots of various research projects. These would come from over 50,000 students, and 3 major hospitals. To my knowledge there are no big electric are furnaces closer than about 25 miles, and I don't know what their current waveform looks like. At one time the Ford Rouge plant consumed more electrical power than the combined load of Detroit and several other cities. But that won't be true today.


mgookin:

Yes, the waveform distortion is consistent. I just looked at it and it is similar to the previous photos. The distortion is not from anything in my home.

Other analysis on my AC waveform shows high frequency noise (basically an impulse shock exciting a resonant circuit, damped sine wave) resulting from one or more dimmers when they are on. This signal is greatest on the phase the dimmer is on as would be expected. The dimmer turn-on point jumps around a little. If I have the TED system on, then I can see its 125 kHz signal.

.

The flat spots at the extremities could be speculated as resistance in the motion of the turbine or some limiting influence but the jitter on the sides can't.

Maybe show those plots to the poco and see if they can explain it?

Have you gone to a neighbor on the same utility to see if it's consistent on the grid?

If you want to run regression analysis on those plots send me a pm and I'll walk you through it (although I don't know if I could ever teach gar anything!)

POCO would (hopefully) be interested in their R2
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
160105-2230 EST

mgookin:

The distortion is not from the DTE generators, it is definitely load created, and it is not from any loads in my home.

From 12 Feb 1962 is a photo of our AC line voltage. P 49 MP 1. There is no visible distortion. Sorry, but I can not produce a larger photo of the peak, and I may not have one.
PICT3866.jpg

Later I will show some waveforms with distortion from my dimmers.

Found another notebook with a somewhat better scope photo. Later I will setup to copy it with a fixed camera mount instead of hand holding.

.
 
Last edited:

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
160106-2112 EST

The following waveforms were from an experiment:

My home 60 Hz AC power from a 240 V center tapped 50 kVA pole transformer. This transformer supplies two street lights, and two homes. The neighbors are presently gone. Generally the secondary voltage is 123 to 125, and balance is better than a volt under light load.

The scope is a RIGOL DS 2072 A with lots of memory. The curves are an average over the last 64 cycles. On a repetitive signal that is moderately stable this increases the apparent resolution, and reduces random noise.

One phase, channel A, is from one phase, and the second phase is on channel B with channel B inverted. Thus, the two phases can be visually overlaid. The input voltages are from the main panel via unloaded wires, and EGC provides, via the scope AC plug, the common reference to the main panel.

Vertical scope gain and position are adjusted to overlay the two traces. The time base of each trace is common.

There are two dimmers active in the house. One a torchère 300 W with intrgral dimmer adjusted to minumum. Can't go very low for a minimum. The second dimmer is a 50 year old GE (thus an early model) controlling front hall lights, eight 60 W bulbs, adjusted to a very low orange glow. Room temperature resistance per bulb is 21.4 ohms. Eight bulbs in parallel is about 2.7 ohms. At dull orange it will be somewhat higher but still low.

The first pair of overlaying curves show one full cycle. Note: how close the curves match. This is to be expected if the voltage distortion is on the primary side of the pole transformer, and not on one side of the secondary center tap. The dimmer distortion is barely perceptible in this view.
DS2_QuickPrint2.jpg


For the second plots the vertical and horizontal scaling have been expaned a little. Also the two curves have been displaced vertically. The torchère disturbance is nearer the voltage peak and the pulse for the front hall lights is lower down. Note: the dimmers are on opposite phases as can be seen by the magnitude of the pulses on the two phases The front hall lights draw a larger peak inrush current than the 300 W torchère.
DS2_QuickPrint6.jpg



End.

.
 

Besoeker

Senior Member
Location
UK
169105-1505 EST

Besoeker:

Your waveform has a nice shape that I would consider to be mostly from similar capacitor input filter power supplies.

The flattend top is consistent with third harmonic distorting. This is typically caused by the abundance of single phase non-linear loads.
In UK, most domestic and commercial installations is fed at 400/230V and, by and large, loads are single phase. Computers, televisions, LED lighting, dimmers etc.

Something I have posted before. Some years ago, we did an installation of variable frequency inverters at a pumping station. From memory, there were eight in total. The two larger units were around 200kW, maybe a bit more so we made them 12-pulse. We were aware of the harmonic spectrum that VFDs generate and we wanted to mitigate the effects. The station was fed by a 1500kVA 11kV/400V transformer. Part of the contract was to ensure that the distortion at the point of common coupling (PCC) was within the recommendations in force at the time. The PCC is that voltage which feed other customers.

The rules are simple. You must not affect the supply to other users. Of course, this is impossible to achieve so some practical limitations were applied. G5/3 it was then. Part of the contract was to ensure that we complied with this so measurements to had to be made to demonstrate that we did so. The PCC was at the 11kV level. So I set up the instruments to do the measurements at 11kV.

What I found was that the supply distortion exceeded the G5/3 limits with none of our VSDs running. And it was no worse with all of them running. Naturally, that got me thinking. The site was in a residential area. It wasn't our drives that were the problem. It was the domestic loading. And I'm sure that's the case with my third harmonic distortion. Not just mine, but area wide.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
160107-0922 EST

mgookin:

For the previous curves the red curve phase is the one with the torchère lamp and the pulse in the red curve is much more dominate than its corresponding occurance in the blue curve. This is to be expected because of the impdeance from the main panel back into 1/2 of the pole transfomrer secondary.

Lower down on the blue curve is the dominate pulse from the front hall lights. Its corresonding occurance in the red curve is much smaller.

Turning off a dimmer does remove its component from the curves.


A new plot:

This is the front hall lights dimmer phase expanded in time (blue) plus a curve of the impulse response of a 125 kHz series resonant circuit (red) connected to the same phase. The TED power monitor carrier frequency is 125 kHz.
DS2_QuickPrint8.jpg


End.

.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
160107-1029 EST

Besoeker:

In do remember your previous comment on this subject.


Anyone else:

Do you have any plots that show distortion, possibly greater than mine, or lack of any visible peak distortion?

.
 

SG-1

Senior Member
Gar, sorry this took so long, but I did not have time until today. This looks different than last time. I remember a few years ago the top was just flat. The wave now has some more distortion on the sides & top.

The scope is a Tenma 72-8400.
 

Attachments

  • V02 copy.jpg
    V02 copy.jpg
    145.5 KB · Views: 0

meternerd

Senior Member
Location
Athol, ID
Occupation
retired water & electric utility electrician, meter/relay tech
Do you share a POCO transformer with other customers? If so, do they have similar distortion? If so, maybe you could plead your case to the POCO to get your own. Not likely, but like the Bible says, "you have not because you ask not".
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
160109-2015 EST

SG-1:

Your curve is interesting because it appears have a change of slope + and - of the voltage zero crossing. Your flattened top looks similar to Besoeker's curve.

.
 
Status
Not open for further replies.
Top