harmonics

Status
Not open for further replies.
S

sparky59

Senior Member
when you're running a multi wire feeder with 3 20 amp circuits and a shared neutral (120/208Y) to offices. How many computers can be on these circuits before harmonic problems that might overload the neutral would be a concern. In my case there will be 6 computers.....2 on each circuit.
 
When I used to wire for computer rooms we used unshared neutrals and oversized the nuetral in the sub panel. For the most part we never had complaints or burned up parts like other establishments that did not use this technique
 
Look at Holts last few newsletters. One of them included an excellent reference that just described this issue.
 
There is some debate as to whether or not this problem really exists. You'll find opinions of some of the posters here that the entire harmonic issue has been blown out of proportion by the copper industry.
 
Although multi-wire circuits are legal and more convienient for the installer, I always spec no shared neutrals in office power loads. Not so much because of the harmonics but because if something happens to the neutral you affect three circuits- or in your case 6 computers- leaving nothing working. If must run shared neutrals, 2 computers per phase is all I would sit still for. No science to back this claim up just a complete absence of problems since having put this policy into effect several years ago.

Having said that, if you are talking a laser printer anywhere in this configuration, all bets are off. Neutrals for everyone.
 
Harmonics can be a real issue, The cases I have involved in where they harmonics were a real concern, these were Casino's and call centers were there were 100's so electronic loads with switch mode power supplies and the transformers were loaded at 80% or more, the neutral current exceeded the transformer rating.

In general (speaking for my area) typical office building, single phase loads are not a problem, mainly because the typically transformer and branch circuits are lightly loaded. Most transformers are carrying 15%-25% load. The same goes for the branch circuits.

2 PC's on a circuit would not be a major issue and is still common practice for office buildings. Data centers ECT are different.

One issue we do see is voltage drop on copier circuits, but that is electricity 101 and most electricians never give it a thought as they pull a 150" #12 awg for a copier.

Another area were we have had harmonic issues was in very old buildings that installed a large number of VFDs, the utility transformers failed, the loading on these old transformers was very high and they may have failed anyway. The end user installed harmonic filters and the transformers were replaced.

In my experience as stated harmonics can be an issue, harmonics like grounds are a catch phrase for some unexplained issue, like bad hardware or software.
 
greetings to all !!!!

I designed a software park having 1650 computors.

Each 20A feeder , i connected with 4 number of computor .... Total 28 feeders in a load center ....but grounded conductor of incomer in the loadcenter would be double the size of live conductor to withstand with hornonic currents.
 
Brian,
The cases I have involved in where they harmonics were a real concern, these were Casino's and call centers were there were 100's so electronic loads with switch mode power supplies and the transformers were loaded at 80% or more, the neutral current exceeded the transformer rating.
Click to expand...
In those somewhat rare cases would it not be more cost effective to use single phase systems for those loads in place of K rated transformer and oversized neutrals? The harmonics do not add in 120/240 volt single phase systems.
Don
 
don_resqcapt19 said:
Brian,

In those somewhat rare cases would it not be more cost effective to use single phase systems for those loads in place of K rated transformer and oversized neutrals? The harmonics do not add in 120/240 volt single phase systems.
Don
Click to expand...


Would the harmonic currents add in the neutral of a 208Y/120 system with two phase conductors sharing a neutral?
 
Trevor,
Would the harmonic currents add in the neutral of a 208Y/120 system with two phase conductors sharing a neutral?
Click to expand...
Yes they do and not only that, but the neutral conductor current from linear sources on that circuit is about the same as the current on the ungrounded conductors.
Don
 
don_resqcapt19 said:
Trevor,

Yes they do and not only that, but the neutral conductor current from linear sources on that circuit is about the same as the current on the ungrounded conductors.
Don
Click to expand...


The reason that I asked was because I have seen engineers design systems using shared neutrals for computer circuits but only two circuits/neutral. I assumed that the harmonic characteristics were different than have three circuits sharing a neutral.
 
Another associated concern is the high neutral to ground voltage at the computer receptacles.This is created due to high 3rd harmonic current flowing back on the shared neutral.Typically,above .5 volts (neutral-ground voltage) will generally result in accelerated computer component damage.Therefore,dedicated neutrals to each receptacle is highly recommended.:)
 
don_resqcapt19 said:
Trevor,

Yes they do and not only that, but the neutral conductor current from linear sources on that circuit is about the same as the current on the ungrounded conductors.
Don
Click to expand...


Agreed. But if only two circuits share a neutral will the current on the neutral ever be higher than the phase current? In a nonlinear, three phase, 4 wire MWBC the neutral current may be up to 1.73 times the phase current. After doing some limited research I can't seem to find any information regarding two circuits sharing a neutral.

My question is if you only had two nonlinear circuits sharing that neutral would the neutral current exceed the phase current?
 
Having a few, or even more than a few computers on a circuit is not the problem. Having a few hundred of them, office lighting, copy machines, air handlers, and an elevator on the transformer that feeds them is. JMHO... A few moderately loaded 120v 20a circuits from a 4-wire MWBC is not anything to worry about.

BTW I rum them (120/208 4-wire) all the time, and often there is no other way to feed furnature systems.
 
infinity said:
Agreed. But if only two circuits share a neutral will the current on the neutral ever be higher than the phase current? In a nonlinear, three phase, 4 wire MWBC the neutral current may be up to 1.73 times the phase current. After doing some limited research I can't seem to find any information regarding two circuits sharing a neutral.

My question is if you only had two nonlinear circuits sharing that neutral would the neutral current exceed the phase current?
Click to expand...
I am going to surmise... Yes, given the same conditions of the 4-wire MWBC that produced 1.73 times the line conductor current. Using that same setup, one could simply de-energize one of the circuits. The triplen-portion of the neutral current would be two-thirds, while the linear-portion of the current would be the imbalance.

Using an example, say the 4-wire mwbc which yielded the 1.73X neutral current was conducting 20A on the line conductors. Keep in mind this 20A also includes the triplen-portion of the current. Hence the set up produced a triplen current on the neutral of 20A ? 1.73 = 34.6A, assuming (and perhaps I'm mistaken for doing so :grin: ) the linear portion of the line currents canceled. Since triplen currents do not cancel, each circuit is contributing one third, or 11.5A, to the neutral current. Line current has both linear and triplen current totaling 20A, of which we determined 11.5A was triplen current. That would mean 8.5A is the linear portion. The neutral will carry half of this current for each of the two energized circuits, which equates to the same value as one of the circuits, namely 8.5A. Add on triplen current for two circuits and we have:
(2 ? 11.5A) + 8.5A = 31.5A.​
The above is just a demonstration of the basic math regarding the described situation. In real life, I believe the values to be considerably off the mark, except for perhaps under extreme conditions of triplen current.
 
Typically,above .5 volts (neutral-ground voltage) will generally result in accelerated computer component damage.Therefore,dedicated neutrals to each receptacle is highly recommended.
Click to expand...




Ghost: Test have been conducted at 15 VAC between neutral and ground and no resulting damaged or operational issues was recorded. The voltage drop issue is true and will result with or without harmonics; voltage drop will be increased in the neutral due to the additional current with switch mode power supplies, resulting from the non cancellation of phase currents.


Voltage drop in the 1.5 VAC range is not uncommon depending on panel (fed from a sub panel can increase VD), type of loads, amperage of circuit, length of branch circuit, conductor size, and the equipment works adequately. I had a hospital with 7.5 between neutral and ground (not recommended) but the equipment operated without issues. Just had everyone freaked out.

Some basic literature


http://www.dranetz-bmi.com//pdf/harmonicspart1.pdf
 
Last edited:
Trevor,
My question is if you only had two nonlinear circuits sharing that neutral would the neutral current exceed the phase current?
Click to expand...
In my opinion it is even more likely that the grounded conductor current will exceed that of the ungrounded conductors when using 2 phases and the grounded conductor from a 3 phase wye system. Remember that when using two phases with equal currents then the grounded conductor current will also be equal to that of the phase current. This being the case any triplen harmonics will increase the grounded conductor current to above that of the phase conductors.
Don
 
don_resqcapt19 said:
In my opinion it is even more likely that the grounded conductor current will exceed that of the ungrounded conductors when using 2 phases and the grounded conductor from a 3 phase wye system. Remember that when using two phases with equal currents then the grounded conductor current will also be equal to that of the phase current. This being the case any triplen harmonics will increase the grounded conductor current to above that of the phase conductors.
Click to expand...
This makes sense. At least with a 'full boat', the neutral current is only the harmonics.
 
Worse than that:

Worse than that:

don_resqcapt19 said:
Trevor,

In my opinion it is even more likely that the grounded conductor current will exceed that of the ungrounded conductors when using 2 phases and the grounded conductor from a 3 phase wye system. Remember that when using two phases with equal currents then the grounded conductor current will also be equal to that of the phase current. This being the case any triplen harmonics will increase the grounded conductor current to above that of the phase conductors.
Don
Click to expand...

If the PFs are equal on two lines, and the current is zero on the third line , the neutral current would be 1.73 x either line current plus the harmonic currents.

Right? Or am I missing something?
 
Status
Not open for further replies.
Top