harmonics

[rattus]

Oops!

Oops!

Neutral current = SQRT(A^+B^+C^-(AB+BC+CA))

10 amps on A and 10 amps on B and 0 amps on C = 10 amps on Neutral (not dealing with harmonics).

Stiff, how right you are! That is obvious from the phasor diagram.

 

[brian john]

This is 120 VAC 3? 4-wire snapshot for a harmonics load.

Note the neutral sine wave the 3 harmonic is the majority of this distortion.

 

[dbuckley]

Thats nice that. An almost perfectly balanced three phase load (I've yet to see a data centre be that well balanced across the phases), with almost as much current flowing in the neutral. Yet still there is a massive body of opinion who believe there is no such thing as "the neutral issue"....

 

[don_resqcapt19]

Yet still there is a massive body of opinion who believe there is no such thing as "the neutral issue"....

And I am among those who don't believe this is an issue in most installations. Sure it is in a data center or other installation with a very highly concentrated non-linear load, but it is not a real world issue in normal office building, schools and similar occupancies. Also when 90% of the documents you see on an problem originate with an entity that has a economic interest in solving the problem, the problem becomes very suspect to me.
Don

 

[brian john]

Don:

I have stated many times that I find this to be at best a minimal issue at most facilities. Even at this site the harmonics were isolated on the UPS output, the voltage waveforms had minimal distortion (3% V-THD). The neutral was sized same as the phase conductors, the UPS was at 67% of FLA.

There was heat associated with the installation but the heat was with in the cunductors ratings.

We have had cases were harmonics resulted in operational problems as stated previously, I was trying to locate a snashot from one of this facilities but could not locate it. We had a site with VFDs and while on generator the V-THD was excessive and messed up voltage regulation, driving the generator crazy, input filters and changing the governor resolved the issue.

 

[dbuckley]

.. at this site ... The neutral was sized same as the phase conductors, the UPS was at 67% of FLA

Ok then, lets introduce a fault, lets lose a phase, so your currents are 212, 227, 0, and then whats the neutral current gonna be? My guess is that your FLA is 300A, and with a lost phase I'd wager your neutral will be comfortably over that. And that's without considering that the three phase SMPS loads will increase their loading on the remaining phases to maintain constant output power.

I agree that harmonics aren't an issue for most locations (numerically, most locations are residentials), but with the increasing prevalence of high efficiency fluorescents and computers, with any commercial space the person speccing the wiring should at least have considered the possibility of the significance of non-linear loads and their effects. Thats very different to dismissing harmonics out of hand as the hogwash of vested interest.

 

[don_resqcapt19]

dbuckley,

Thats very different to dismissing harmonics out of hand as the hogwash of vested interest.

That is exactly what I am going to do until such time as I see some real world evidence of a problem in commerical buildings. I am not going to include dwelling units in this because the harmonics do not cause grounded conductor overloading on single phase systems. And it is still my opinion for those rare installations where there is problem with harmonic neutral current the most cost effective installation would be to use 120/240 volt single phase power in place of the 208/120 wye that is most commonly used.
Don

 

[brian john]

In commercial office buildings in the Washington DC Metro area this has not been an issue, except in specific circumstances. Should we be concerned yes, is it a cause of major problems, not in my area to date.

Lets take your case; due to poor installation and a lack of maintenance, you lose the neutral, with a grounded neutral downstream, now this connection point is carrying the neutral current. FIRE.

In my expierence 90% + of the commercial buildings have a grounded neutral issue and which means there are a lot more grounded neutral issues than there are harmonic issues IN MY AREA.

 

[ghostbuster]

I would have to agree with all you fellows.We have been involved with harmonic issues for the last 30 years.There are some real live problems out there and there are lots of "smoking guns" waiting to go off.There are also groups out there, that are trying to blame everything on harmonics, if something "weird" happens in the electrical system.Unfortunately, there will always be a small percentage of snake oil flowing around.I remember a well respected electrical textbook written in the 1950's,stating that for all intensive purposes, harmonics are a figment of your imagination.Technology has really changed since then, but there still is an element of truth in this statement when the snake oil wagon rolls into town.

 

[don_resqcapt19]

Brian,

Lets take your case; due to poor installation and a lack of maintenance, you lose the neutral, with a grounded neutral downstream, now this connection point is carrying the neutral current. FIRE.

In my expierence 90% + of the commercial buildings have a grounded neutral issue and which means there are a lot more grounded neutral issues than there are harmonic issues IN MY AREA.

I think that that is true in most areas. It seems that a lot of power quality issues go away when the installation is per the NEC. As far as the open neutral, a large amount of triplen harmonic current makes the problem worse, but it would still be a problem even in cases where all of the load is linear.
One thing that would help the grounded neutral issue is for the panel manufacturers to not provide a main bonding screw or strap with the panels. A lot of installers will use that item any time it is provided with the panel when it should only be used on the panel that contains the service disconnect or the first disconnect on an SDS.
don

 

[winnie]

One thing that would help the grounded neutral issue is for the panel manufacturers to not provide a main bonding screw or strap with the panels. [...] it should only be used on the panel that contains the service disconnect or the first disconnect on an SDS.
don

Given the choice of errors, I would prefer the grounded neutral in a subpanel to the ungrounded neutral at the service or SDS.

Not saying that I would want either fault, but I think that having the bonding screw provided is on the whole a safer situation.

-Jon

 

[dbuckley]

Lets take your case; due to poor installation and a lack of maintenance, you lose the neutral, with a grounded neutral downstream, now this connection point is carrying the neutral current. FIRE.

You quoted and replied without reading. I said lose a PHASE, which is far more common and should be an entirely safe situation, but in the case you illustrated above, loss of a phase will almost certainly lead to an overloaded neutral, as it's "only" rated for 100% of phase FLA.

 

[don_resqcapt19]

dbuckley,

You quoted and replied without reading. I said lose a PHASE, which is far more common and should be an entirely safe situation, but in the case you illustrated above, loss of a phase will almost certainly lead to an overloaded neutral, as it's "only" rated for 100% of phase FLA.

Another example of where the use of single phase 120/240 volt systems in place of the 208/120 wye systems would eliminate the problem.
Don

 

[dbuckley]

Another example of where the use of single phase 120/240 volt systems in place of the 208/120 wye systems would eliminate the problem.

You are, of course, in principle, absolutely correct, and having three separate, single phase transformers or one transformer with three separate secondary windings (ie no wye or delta) providing three separately fed distributions, one for each phase, will absolutely alleviate the harmonic current issues, and no part of the system needs to be rated above 100% FLA.

There are reasons why this is not common practice however, mainly economic, one way to look at what you have is a 300% percent neutral capacity, and value engineering will try and remove that.

In the world of UPSs there exist three phase in, single phase out UPSs, and they would be the absolute ideal thing for data centres and other places like TV studios, but they suffer from the massive drawback that you cant bypass the UPS, as you don't have sufficient juice available one on any one input phase to take the whole load, as the input load was spread evenly across three phases. You would need the POCO to supply a tranny and distribution for 3x your real load, whilst only paying for 1x.

Economics will kill this every time, except for odd applications where money really isn't a significant concern.

 

[LarryFine]

. . . having three separate, single phase transformers or one transformer with three separate secondary windings (ie no wye or delta) providing three separately fed distributions, one for each phase, will absolutely alleviate the harmonic current issues . . .

However, since one line of each secondary must be bonded, you will still effectively have a Y system. You can still have separate 2-wire circuits, of course.

 

[winnie]

Don,

I think that you make a very good point with your suggestions of single phase systems. Such would certainly eliminate the problem of overheating neutral conductors.

However I wonder if it would actually benefit the transformers. Power distribution will still be done with three phase, even if the final load is single phase.

Consider the choice between a single three phase transformer with a 208/120Y secondary, or three single phase transformers with 120/240V single phase secondaries. I presume that in both cases, 'best practice' would have a delta primary (either a true delta in the case of the three phase transformer, or a delta bank in the case of the three single phase transformers). Please correct me if I am wrong on this.

In either case, triplen harmonics in the loads would present as circulating currents on the primary. So in cases where a 'K' rating is actually meaningful (rather than simply selling extra copper), I think that the single phase transformers would still need to deal with harmonic induced circulating currents.

Of course, this would not be the case for a single phase system, or for an 'open delta' system; but such systems would present unbalanced loading to the distribution system, and would presumably add costs here.

Are there different transformer layouts which would present balanced three phase loading to the distribution system, but not have problems with triplen harmonics on the secondary side? Or are 'K' rated transformers and primary side harmonic currents more marketing?

-Jon

 

[winnie]

However, since one line of each secondary must be bonded, you will still effectively have a Y system.

If the secondaries are 120/240V center tapped, then you can feed single phase multiwire branch circuits. This lets you feed 2 loads with 3 conductors; not as good as feeding 3 loads with 4 conductors as in a 3 phase 'full boat', but not as bad as running a full separate neutral for each 120V load.

All of the center taps would be grounded, so in essence you would have a 6 phase star system rather than a wye. If you ran all 6 legs with a suitable sized neutral (say 6 #12 conductors and 1 #8 conductor) then the third harmonic would tend to balance out between the 60 degree legs, and the net copper cross section as compared to a pair of 'full boats' would be very similar. Though I suppose the custom 6 phase panelboard would kill any possible cost benefit

-Jon

 

[dbuckley]

In either case, triplen harmonics in the loads would present as circulating currents on the primary.

I don't believe so. But I confess I don't fully understand why, and it's because the whole issue of triplens falls into black magic math for me.

If you have a sinusoidal voltage supply feeding a switched mode power pack, then the current is drawn as pulses, as shown on the 'Scope piccies above. We know that Mr Fourier taught us that any repetitive waveform can be made up of sums of harmonically related sine waves, and we know that the predominant harmonic in the current pulse waveform is the third and other odd harmonics, and 'cos we use 3 phase the harmonics add on the common neutral. The harmonic currents are on top of the existing "real" fundamental 60Hz current, and I think that is the important bit.

On pure single phase the current waveform is still a pulse, but it's identical on the hot and the individual phase's neutral. But because there is no common harmonic current theres no addition, and thus no circulating current. So no need for K rated transformers.

And back to delta bank or separate secondaries; they would have one end of each winding (the neutral end) connected to ground, as its a SDS, but no current flows through that point, its not a neutral "point"; there are three neutral points, and each one is grounded. If you make it 120/240 I think you are removing the benefit of not sharing neutrals.

 

[winnie]

On pure single phase the current waveform is still a pulse, but it's identical on the hot and the individual phase's neutral. [...] If you make it 120/240 I think you are removing the benefit of not sharing neutrals.

I can actually answer this point for _single_ phase systems. I don't really know what happens when you have a three phase bank of single phase systems, though I do have a strong guess (above).

As you mention, any complex periodic waveform can be broken down using harmonic analysis. The nifty thing is that we can then analyze a system for its response to each individual harmonic, and add up the results to get the response of that system to the complext waveform.

Well, it is _not_ true that all harmonics present on the load waveform will add up on the neutral.

Instead only those 'multiple of 3' load harmonics will add up on the neutral. Everything else will balance out. This means that if a particular power electronic load happened to produce lots and lots of 5th or 7th harmonic, this would not cause any neutral loading problem. It is only the multiple of 3 harmonics, in particular the 3rd harmonic itself, that cause problems.

In a single phase center tapped system, the harmonics which would cause problems are multiples of _2_. 3rd harmonic does not cause a problem for the neutral in a single phase system. The characteristics of most power electronic loads are such that they tend to produce very little 2nd harmonics. So in a single phase system, you could have multiwire branch circuits (2 circuits sharing a neutral) without having harmonic problems on this neutral or on the secondary of the transformer.

This would also be true for the six phase 'star' system that I described, although it would have extensive practical problems.

-Jon

 

[zbang]

In the world of UPSs there exist three phase in, single phase out UPSs, and they would be the absolute ideal thing for data centres and other places like TV studios, but they suffer from the massive drawback that you cant bypass the UPS, as you don't have sufficient juice available one on any one input phase to take the whole load, as the input load was spread evenly across three phases. You would need the POCO to supply a tranny and distribution for 3x your real load, whilst only paying for 1x..

As they say, well, it depends. In the early 80's, I worked on just that UPS- 3ph input for the battery charger and 1ph 120vac output. It also had a 1ph input for the static & mechanical bypass. Granted it was only 20kva, so even the single phase bypass wasn't that much in the context of a 12 story office building. If your loads aren't huge, it's entirely possible. For a moderate-sized TV station, when you exclude studio lighting loads, it's entirely possible. IIRC one local station only needs something like 15kva to run mastercontrol and the uplinks. (It's a lot easier now that most stations are going to servers for playback.)