Show us your mains waveform!

[Caliaxy]

Oh, wait... you mentioned solar panels. Geesh, never mind. Who the hell knows what's going on in those proprietary inverter controls.     

That’s easy to figure: repeat the measurements at night. 

[engrguy42]

Oh, wait... you mentioned solar panels. Geesh, never mind. Who the hell knows what's going on in those proprietary inverter controls.     

That’s easy to figure: repeat the measurements at night. 

Yeah, that's a good point. Try measuring at night with all your loads and inverters and stuff off and see what you get. Around midnight or something when most loads in the area are off. 

May help you narrow down the culprit.

And make sure your inverter is disconnected completely.

[ahbushnell]

Our mains are quite distorted.
So much so, that our stove elements buzz at certain times of the day.  The attached image was taken several years ago.
It too shows the flattened tops, but even worse are the small 'steps' which are causing the buzzing I hear.

When I spoke to a friend who works on generation stations across North America, his first words were "this doesn't surprise me".
Apparently non-linear loads are the culprit and are no longer just a large industry problem: https://www.mirusinternational.com/downloads/hmt_faq01.pdf

There are growing demands for PFC in electronic equipment, but hardly anything is said about the severe harmonic distortion in the current waveform that most PFC hardware causes.

PFC input hardware generates a sine current waveform to reduce harmonics and improve power factor.  That's why it's called Power Factor Correction (PFC). 

[engrguy42]

I'm curious whether the o'scope that is making the measurements is also causing some distortion...

Hmm.....

[tautech]

And make sure your inverter is disconnected completely.

And the ATX SMPS PSU in your PC. 

[engrguy42]

And make sure your inverter is disconnected completely.

And the ATX SMPS PSU in your PC. 

Yeah. Though with mine I just tried turning off my 3 desktops and the little distortion I have didn't change. I also turned off my CFL. But not the fluorescents in the kitchen, etc.

But I think the first place to start looking is in your own house. Turn it all off and then measure.

And maybe if someone can measure with a scope that is battery powered or whatever that might help.

EDIT:    Here's a before and after image with mine. Top is before, with 3xATX desktop computers on as well as the CFL and LED light in my office, and bottom is with all that off, but still some CFL's and fluorescents in the house on. And of course the scope I measured with...

[tautech]

But I think the first place to start looking is in your own house.

Nope, distorted flat top mains sine waves are a worldwide phenomenon, period !

Nothing to see or prove here other than place yourself or equipment at risk trying to measure mains. 

And maybe if someone can measure with a scope that is battery powered or whatever that might help.

As above, nothing to gain or see that most everyone experienced already knows about.

Distorted mains is so common that clean mains is entirely uncommon !

Know this, accept it and everyone gets on with life.

The only variable is how bad it is.......none of the screenshots posted thus far are as bad as some I've seen.......the worst of which was in a Uni computer sciences campus with dozens and dozens of PC's each contributing to it. Most of the nearby campus's had the pleasure of having to suffer it too including the power electronics campus when needing to take accurate mains measurements of mains control devices.

Still, if nothing else it was a good introduction to the real world for students ! 

[engrguy42]

Gee Mr. Tautech you're so smart!! We mere students are in awe of your brilliance.

By the way, how about some facts once in a while rather than nonsense like "worldwide phenomenon"?

EDIT: Oh wait, isn't tautech the guy who said the reason ATX power supplies are evil is because he once had a bug climb in one and it went ZAP? 

Yeah, um...forgive us if we don't admire your brilliance.

[tautech]

By the way, how about some facts once in a while rather than nonsense like "worldwide phenomenon"?

Point us to a screenshot already posted from members around the globe that displays a clean mains sinewave.....there are none !

I rest my case.

[engrguy42]

By the way, how about some facts once in a while rather than nonsense like "worldwide phenomenon"?

Point us to a screenshot already posted from members around the globe that displays a clean mains sinewave.....there are none !

I rest my case.

You're free to rest whatever you want, but until we take an engineering approach by removing all possible local causes it's not possible to determine the ultimate cause. It could be the utility, it could be our own equipment, it could be a lot of things.

Handwaving is not engineering.

[tautech]

Handwaving is not engineering.

Nor is your display of limited knowledge of the real things that might affect measurements so they are not what we expect them to be.
Gawd help us when we don't see the result on an oscilloscope that we expect !
What might cause it when all known things we can control are excluded from impacting this measurement ?


Simple deduction in this case of distorted mains, we wrongly consider mains as a infinitely low impedance current source and while it is quite low it is NOT the perfect current source that we might base calculations on therefore it's presumed perfect sinewave is influenced by every mains grid users current loading to be what we measure and see.
As is the case with a low impedance current source, use of voltage measurement techniques like 10M input DMM's and 10X (gawd forbid  ) probes has an infinitesimal influence on such a low impedance current source so for any practical purposes we can disregard them.

Distorted mains is a fact of life, get over it and get on with life.

[engrguy42]

tautech, for some reason you're going ballistic and taking a technical discussion personally and trying to discredit those who disagree. Chill out.

Mains voltage is designed to be very clean. Take a look at IEEE 519 which defines limits for flicker, THD, etc. As I recall those values are somewhere below 3-5%. People go to a lot of trouble to make sure it's clean, and people can expect the distortion to be below those levels. Nobody ever said it's always perfect, but in general there are industry guidelines that limit distortion. 

Now if these waveforms we're seeing are within IEEE 519, then fine. But IEEE 519 doesn't apply in our homes, and stuff we have connected could cause our waveforms to get distorted beyond IEEE 519.

Why get triggered over a technical discussion? Geez. 

[coppice]

Point us to a screenshot already posted from members around the globe that displays a clean mains sinewave.....there are none !

I don't have a scope with me, but a power analyser plugged into the wall beside me currently reads 4.7% THD. I don't know how far away our sub-station is, but I don't think its close, or I would have noticed it. 4.7% is not going to give an obviously distorted waveform on the scale of a scope screen. I've plugged distortion measurement tools into the mains in a number of hotels around the world, and you can often see readings as low as 1% or 2% on a lower floor, so many people's grids are capable of clean results. What tends to make a big difference is the floor you are staying on. On a high floor you might see 15% or 20% THD, from all the activity in the hundreds of rooms below you modulating the ohmic losses in the tower's wiring.

If you are used to looking at power in labs in industrial areas you might well not be used to seeing a clean waveform. Those places tend to have horrible distortion, even close to the sub-station. Residential locations near a sub-station, or the lower floors of towers with a sub-station in the basement or ground floor usually look pretty clean.

[BravoV]

Just do NOT feed the troll ...

[uer166]

I'm with tautech here. When working on designing electricity meters, it's hard/impossible to use the grid to test the meters or do calibration because the grid voltage is constantly fluctuating +- 0.5V RMS with bigger dips whenever someone's dryer turns on. Also the flat tops caused by all the rectifier+cap loads are pretty much universal, no matter the region. Never seen clean mains that look like a sinewave, whether it's residential, or household (at least here).

The measurement I posted is done by a 24-bit metrology front-end, so I have decent confidence it's correct, yet the flat tops being clipped is obvious. It is a fact of life, and all mains-powered things have to be designed to deal with it. Have you ever seen a modified sine inverter waveform? Literally a square wave, yet most loads deal with it just fine.

[coppice]

I'm with tautech here. When working on designing electricity meters, it's hard/impossible to use the grid to test the meters or do calibration because the grid voltage is constantly fluctuating +- 0.5V RMS with bigger dips whenever someone's dryer turns on. Also the flat tops caused by all the rectifier+cap loads are pretty much universal, no matter the region. Never seen clean mains that look like a sinewave, whether it's residential, or household (at least here).

The measurement I posted is done by a 24-bit metrology front-end, so I have decent confidence it's correct, yet the flat tops being clipped is obvious. It is a fact of life, and all mains-powered things have to be designed to deal with it. Have you ever seen a modified sine inverter waveform? Literally a square wave, yet most loads deal with it just fine.

I don't think anyone would question that you have to make equipment that will deal with quite poor mains, as you need to design for the worst you will see in the real world. That can be pretty bad, especially in industrial areas, and some countries with a less than stellar grid quality. Tautech claimed mains power is never clean, and that's not the case.

A lot of loads do very badly with a square power waveform. Many power supplies will go up in smoke, because high enough harmonics will put a huge current through the front end filter caps. Quite a lot of devices using a cap drop power supply object strongly to high harmonics. A well designed cap drop supply puts the resistor well away from anything else, to maximise tolerance if it overheating on rich harmonics, but a lot of cap drop supplies are very poorly designed, and in development are not tested on severely distorted power waveforms.

[joeqsmith]

Nothing to see or prove here other than place yourself or equipment at risk trying to measure mains. 

Distorted mains is so common that clean mains is entirely uncommon !

The only variable is how bad it is.......none of the screenshots posted thus far are as bad as some I've seen.......

Can you post some of your data? 

As I mentioned, I've seen pretty bad signals from unstable gensets (not your little toy 10KW system), over loaded transformers, line conditioners.     

There was one that I am still not sure what the cause was.  I suspect that they had a way to try and balance the loading between phases dynamically or they had some other sort of line conditioner.   Most of the larger places with have a master electrician that will know pretty much all the details.  In this case, I have no idea.   I'll see if I can find it.   

One place had some fairly large equipment that had to be scheduled with the power company when it was cycled.   Their electrician was asking me for some advice about a very high current circuit they were working on.  It sounded so far fetched, I asked to see it.  That was the first time I had seen water cooled cables.   

I've been on a tour of a power plant and that was on a whole different level.     This was the plant I had mentioned where a worker had been killed.  I don't venture out much but have a lot of respect for the people who work in these environments.     

[joeqsmith]

I expect all our industrial electricians will know the cause of two of these but that third ones got me.  If you were to overlay the three phases, it looks like they lay right in line.  This is why I suspected some sort of dynamic load balance.  I've read about them.  That or something else really strange.  If you have seen it before and know the cause, I would like to hear from you. 

[srb1954]

This is basically correct.
Specifically, it is the use of a bridge rectifier feeding into a capacitor input smoothing filter that causes distortion like this. The rectifier only conducts towards the top of the mains cycle and the smoothing capacitor draws a high current until it recharges back up to the peak on the mains voltage, whereupon the bridge rectifier stops conducting until the peak of the next cycle.

[srb1954]

There are growing demands for PFC in electronic equipment, but hardly anything is said about the severe harmonic distortion in the current waveform that most PFC hardware causes.

This is incorrect.
The purpose of PFC circuits is to substantially reduce the current waveform distortion caused by rectifier circuits. They may not produce a perfectly clean waveform but they do considerably improve the distortion compared to non-PFC rectifiers.

[ahbushnell]

There are growing demands for PFC in electronic equipment, but hardly anything is said about the severe harmonic distortion in the current waveform that most PFC hardware causes.

This is incorrect.
The purpose of PFC circuits is to substantially reduce the current waveform distortion caused by rectifier circuits. They may not produce a perfectly clean waveform but they do considerably improve the distortion compared to non-PFC rectifiers.

you are correct. 

[Simon]

tautech, for some reason you're going ballistic and taking a technical discussion personally and trying to discredit those who disagree. Chill out.

Mains voltage is designed to be very clean. Take a look at IEEE 519 which defines limits for flicker, THD, etc. As I recall those values are somewhere below 3-5%. People go to a lot of trouble to make sure it's clean, and people can expect the distortion to be below those levels. Nobody ever said it's always perfect, but in general there are industry guidelines that limit distortion. 

Now if these waveforms we're seeing are within IEEE 519, then fine. But IEEE 519 doesn't apply in our homes, and stuff we have connected could cause our waveforms to get distorted beyond IEEE 519.

Why get triggered over a technical discussion? Geez. 

I think that applies to yourself! what's your problem?

[Simon]

Solar could be contributing. I should measure mine while I am at full power. Solar is constant power so come the peaks there is less current to give unless they are storing it and managing the output carefully over the cycle.

My computers ATX supply is PFC and that is 10 years old.

[nfmax]

To clarify things and answer a few questions people have asked:

• I'm using a Picotech TA041 high-voltage differential probe, which is rated to ±700V operating, and is safety rated CAT III to 1000V. This is connected to a Keysight MSOX-3104T oscilloscope which is properly earthed via the mains protective earth conductor. I'm using appropriate test leads and connectors.
• At the time of the measurement, 14:23 BST, my solar panels were generating 1242W and the house was drawing an additional 8W from the public supply, measured by counting pulses from the supply company's tariff meters. These are 15 minute average values, covering the period 14:15 to 14:30. However, neither the oscilloscope nor the datalogger clocks are NTP synchronised so the times may be off a bit. When I repeated the measurement, at 19:17 BST, the panels were generating 12W and the house was importing 359W. As you can see the waveforms were pretty much identical.
• As I said, I'm very close to the transformer, and I was surprised to see this amount of distortion. It doesn't seem feasible that it is happening because of the loads on the local 240V distribution cables, rather that it represents the waveform on the district 11kV(?) network.
• I did take a quick look at the FFT of the waveform, but didn't make a note of the numbers - I will do this later & post the results.

That distortion looks high enough that it might cause problems, for example with a large induction motor. The harmonic currents generate no torque, only heat, and the impedance of the motor may be much less at harmonic frequencies. Fortunately the only induction motors I have here are the pond pump, the bench grinder, and the fridge. Probably this explains why the replacement central heating circulating pump is now a BLDC motor - the plumber had no explanation for this other than 'they all are now'. Of course, replacing induction motors with BLDC motors will only make the harmonic problem worse.
This has been an educational experience for me, at least.

[uer166]

"replacing induction motors with BLDC motors will only make the harmonic problem worse." Why is that? I'd think any sizeable BLDC will have a PFC front-end, so no harmonics.