Flatened top sine wave, does it mean anything?

Status
Not open for further replies.
gar said:
The cause of the flatten peak is a result of all the equipment with rectifiers followed by a capacitor input filter that are loads on the entire grid. I think you will see this in most parts of the country. I have scope photos from years ago that display very nicely formed sine waves from the line. Today my supply has flattened tops, and these are not primarily from my loads.
.
Click to expand...


How well do you think synchronous caps would do to restore a nicely formed sine wave?
 
Besoeker said:
The flattened top results from typically third harmonic voltage. That means that there is a 180Hz component. Given that Xc = 1/ω the 180Hz component will increase the loading on the capacitors.
And, according to iwire, capacitors are failing.
And ballasts. Conventional magnetic ballasts, also with capacitors maybe?

Seen it happen. Mostly in industrial applications I have to say. But then again, I don't often get involved in commercial and almost never in residential. And the capacitors don't care where.
Click to expand...

Should be:
Given that Xc = 1/ωC etc.

Apologies, folks.
 
Besoeker said:
Seen it happen. Mostly in industrial applications I have to say. But then again, I don't often get involved in commercial and almost never in residential. And the capacitors don't care where.
Click to expand...

Not really a dwelling unit per say, this is a pretty large place, two utility services, a 1600 amp 208Y/120 and a 3000 amp 480Y/277.

There is a theater, a very large commercial kitchen, small store etc., I don't know how many tenant units they have.

There should be no magnetic ballasts in the entire place, they will be all electronic ballasts and electronic power supplies for low voltage lighting.

http://maps.google.com/maps?hl=en&l...41702&sspn=0.002679,0.005681&vpsrc=0&t=h&z=18


http://www.proprietorsgreen.com/
 
iwire said:
Not really a dwelling unit per say, this is a pretty large place, two utility services, a 1600 amp 208Y/120 and a 3000 amp 480Y/277.

There is a theater, a very large commercial kitchen, small store etc., I don't know how many tenant units they have.

There should be no magnetic ballasts in the entire place, they will be all electronic ballasts and electronic power supplies for low voltage lighting.

http://maps.google.com/maps?hl=en&l...41702&sspn=0.002679,0.005681&vpsrc=0&t=h&z=18


http://www.proprietorsgreen.com/
Click to expand...
Appreciated, thanks.
Our lighting division deals with premises like that. And hotels, public buildings, and some hospitals.
I'd call it commercial.
 
120114-0934 EST

Besoeker:

I was directing my comment to electronic ballasts. Assuming the same RMS voltage for a true sine wave and the flat topped one I do not believe the flat top will have an adverse impact on the electronic ballast. The flat top probably slightly lowers the peak current to the filter capacitor and slightly increases the charging time. Peak charged voltage might be slightly less. These comments are intuitive without calculation.

.
 
The last time I remember seeing flat topping like that was where current limiting was happening from impedance, very common in power amps, where the user cranks up the attenuator far past the available power level of the amp to the point of clipping?
 
gar said:
120114-0934 EST

Besoeker:

I was directing my comment to electronic ballasts. Assuming the same RMS voltage for a true sine wave and the flat topped one I do not believe the flat top will have an adverse impact on the electronic ballast.
Click to expand...

Hi Gar.

I agree with that.
The capacitors on the air conditioning units may be a different story.
 
hurk27 said:
The last time I remember seeing flat topping like that was where current limiting was happening from impedance, very common in power amps, where the user cranks up the attenuator far past the available power level of the amp to the point of clipping?
Click to expand...
My residential supply posted in another thread on this forum:

Onecycle01.jpg


As it happens, one of my guys asked me to look at something in the test lab last week. It's 25 miles from where I live. He was trying to measure the phase shift between primary and secondary voltages of a small transformer. The primary was phase to neutral of the supply here, nominally 230Vac and its shape was pretty similar to that at our house and what I have seen elsewhere on single phase supplies.
The fellow considered it to be a pretty good sine wave. In fairness, not his field.
 
Last edited:
As I look closely at the wave form in the original post it really isn't flat topped as it rises to a point and then it rounds over and then slants ever so slightly up toward the right to its peak. You will note that the wave form then drops sharply. It would be my assumption that there is a capacitive load of some type where the current leads the voltage. There appears to be a sufficient amount of current flowing into the load which does not allow the voltage to rise to the peak that is expected. I am assuming this because current represents the load. If there is sufficient current flowing to cause the voltage to drop that may cause the voltage to peak out where it does before rising very slightly to a peak before the voltage follows the current down repeating itself on the other half of the wave form.
I don't know if the irregular line of the wave is do to the scanning process or it’s the actual wave form.

The wave form that Besoeker showed appears to clearly illustrate a very clearly a small jaggedness which is not unusual. The wave form in the OP doesn’t seem to show as much detail but I don’t know it that really matters when looking at the general form of the wave

It would be interesting if one were to add some inductive load to see it would change the top of the wave form where voltage leads current.

I must say that this is an very interesting topic with out actually knowing what the actual content of the load and making assumptions from the wave form.
 
120114-1925 EST

The capacitive load is all the capacitor input filters fed from rectifiers with very little series inductance or resistance. This is a transient analysis problem at the peak region of the sine wave.

Quickly the I found the following chart p 5 of 15 at http://www.southalabama.edu/engineering/ece/faculty/akhan/Courses/EE334-Fall07/EE 334-chapter-2.pdf labeled Power Supply Circuits, Filters and Regulators. However, no regulator. Note the shape of the current waveform. When you add a small amount of series resistance and inductance, then the current pulse shape changes and ends somewhat after the source voltage peak. The leading edge won't be so vertical and sharp peaked.

At the moment my scope is non-functional and I do not have time to setup something else to get curves.

.
 
When afci first came out Sq. D had some major issues in Denver, Colorado. They found that the power quality from the poco was poor and it had a flattened sine wave. That is about all I can add to this thread. :thumbsup:
 
gar said:
At the moment my scope is non-functional and I do not have time to setup something else to get curves.
Click to expand...
I shouldn't worry about that on my account.
I'm moderately familiar with a number of rectifier and regulator configurations..............
 
120115-1038 EST

Besoeker:

My post you were referencing was not directed to you but to tempkll, and should have been so addressed.

I do not think most of the readers are very familiar with transient analysis. So the comment was also directed at them.

A few nice waveforms from an actual set of components might illustrate why the sine wave peak is flattened, and that is where my scope would fit, but at the moment can not.

If you have the time to do this it might help others to see what is happening.

.
 
gar said:
120115-1038 EST

Besoeker:

My post you were referencing was not directed to you but to tempkll, and should have been so addressed.

I do not think most of the readers are very familiar with transient analysis. So the comment was also directed at them.

A few nice waveforms from an actual set of components might illustrate why the sine wave peak is flattened, and that is where my scope would fit, but at the moment can not.

If you have the time to do this it might help others to see what is happening.

.
Click to expand...
Thanks Gar.
I'll try to dig something up.
I have a digital storage scope in my car that can save waveforms and a serial comms module but computer doesn't have a port for that.
I've got some saved waveforms that I can submit but probably none for single phase.
Let me have a think....but...

I'm in the process of puting together a bid for some big drives and, other than dipping in here now and again, that has to be the focus of my attention for now.
 
I've put together an Excel sheet that I think is kind of interesting, but I can't seem to upload it. I keep getting an "invalid file error." I tried to zip it, and I still get the same error.

Anyhow, the spreadsheet plots a sine wave added together with the 1st four odd harmonics. You can change the amplitude of any of the harmonics. So you can play with different harmonics, and see what the waveform looks like.

Anyhow, with amplitudes of:

1 for the fundamental.
0.17 for the 3rd
and
0.03 for the 5th

produces a wave with a very flat top. In fact, it starts looking like a crude approx. of a square wave.
 
steve66 said:
I've put together an Excel sheet that I think is kind of interesting, but I can't seem to upload it. I keep getting an "invalid file error." I tried to zip it, and I still get the same error.

Anyhow, the spreadsheet plots a sine wave added together with the 1st four odd harmonics. You can change the amplitude of any of the harmonics. So you can play with different harmonics, and see what the waveform looks like.

Anyhow, with amplitudes of:

1 for the fundamental.
0.17 for the 3rd
and
0.03 for the 5th

produces a wave with a very flat top. In fact, it starts looking like a crude approx. of a square wave.
Click to expand...

Like this?

GHA3j.jpg

It's roughly what you get for the line current into a three phase rectifier with a good sized choke or an inductive load.
Level current for 120deg in each half cycle.
 
Besoeker said:
Like this?


It's roughly what you get for the line current into a three phase rectifier with a good sized choke or an inductive load.
Level current for 120deg in each half cycle.
Click to expand...

Yes, similar without all the ringing that the higher harmonics produce.

Where is the 3rd harmonic??
 
gar said:
120115-1038 EST

I do not think most of the readers are very familiar with transient analysis.
Click to expand...

I know I am not and I welcome any input from those that are. :)

Right now the meter is online recording. I will leave it there for seven days, at that point I will down load the data and make a report that will have much more data than what I can show with just a snapshot of the screen. :)
 
Status
Not open for further replies.
Top