harmonics in a 208V service
- Thread starter bwyllie
- Start date
- Status
Re: harmonics in a 208V service
Good question, it brings up a very good point. I get a laugh about hearing people using 200% neutral busses and over sizing the neutral circuits with a standard service where the service neutral is half the size of the phase conductors.
No it does not make since to oversize unless you use a "K" rated delta input/wye output transformer.
Good question, it brings up a very good point. I get a laugh about hearing people using 200% neutral busses and over sizing the neutral circuits with a standard service where the service neutral is half the size of the phase conductors.
No it does not make since to oversize unless you use a "K" rated delta input/wye output transformer.
john m. caloggero
Senior Member
Re: harmonics in a 208V service
On a 3-phase, 4-wire wye system,where the major portion of the line to neutral loads are nonlinear loads, such as electric-discharge lighting, computers/data processing, electronic welders ect., there will be harmonic currents in the shared neutrals. A survey of what type of loads will be served and then a decision can be made as to whether an oversize neutral is needed.
On a 3-phase, 4-wire wye system,where the major portion of the line to neutral loads are nonlinear loads, such as electric-discharge lighting, computers/data processing, electronic welders ect., there will be harmonic currents in the shared neutrals. A survey of what type of loads will be served and then a decision can be made as to whether an oversize neutral is needed.
Re: harmonics in a 208V service
this will be a office/lab IT space with computers and fluorescent lighting. I agree with how unuseful the 200% neutrals would be if the service neutral and some feeder neutrals are 100%.
[ December 02, 2004, 10:30 PM: Message edited by: bwyllie ]
this will be a office/lab IT space with computers and fluorescent lighting. I agree with how unuseful the 200% neutrals would be if the service neutral and some feeder neutrals are 100%.
[ December 02, 2004, 10:30 PM: Message edited by: bwyllie ]
jtester
Senior Member
- Location
- Las Cruces N.M.
Re: harmonics in a 208V service
I disagree with derekbc. If you don't install your portion of the project properly, you will have magnified the problem. Your portion will not add to the voltage drop, to the overheating, and to the final corrective action.
There will still be problems, however the system will work better than if you install undersized neutrals also.
Jim T
I disagree with derekbc. If you don't install your portion of the project properly, you will have magnified the problem. Your portion will not add to the voltage drop, to the overheating, and to the final corrective action.
There will still be problems, however the system will work better than if you install undersized neutrals also.
Jim T
- Location
- Massachusetts
Re: harmonics in a 208V service
I am not an engineer by any stretch, that said running oversize neutrals has always struck me as a waste most times.
Before you say I am an idiot
I am not an engineer by any stretch, that said running oversize neutrals has always struck me as a waste most times.
Before you say I am an idiot
- Location
- Illinois
- Occupation
- retired electrician
Re: harmonics in a 208V service
Are there really documented cases of this being a real problem? Cases that have been investigated to rule out things like poor connections and other types of non-harmonic related problems. I've talked with many design engineers who do large offices and schools that have never seen this problem and do not use oversized neutrals or K rated transfomrers. Almost all of the information that I have ever read on this subject has come from sources with an economic interest in correcting the "problem".
Don
Are there really documented cases of this being a real problem? Cases that have been investigated to rule out things like poor connections and other types of non-harmonic related problems. I've talked with many design engineers who do large offices and schools that have never seen this problem and do not use oversized neutrals or K rated transfomrers. Almost all of the information that I have ever read on this subject has come from sources with an economic interest in correcting the "problem".
Don
jtester
Senior Member
- Location
- Las Cruces N.M.
Re: harmonics in a 208V service
In reply to iwire, I am an engineer who readily admits that my profession regularly does things like oversize way too many neutrals. To admit less would be insincere.
That being said, there are instances where it is a problem. I worked on a data center at a local bank where the harmonic currents in the neutrals were horrendous. We ended up installing new feeders and K rated transformers dedicated for data.
More than 90% of the projects that I design don't need additional neutrals, K rated transformers, isolated grounds, etc. However I am just a small time engineer.
Jim T
In reply to iwire, I am an engineer who readily admits that my profession regularly does things like oversize way too many neutrals. To admit less would be insincere.
That being said, there are instances where it is a problem. I worked on a data center at a local bank where the harmonic currents in the neutrals were horrendous. We ended up installing new feeders and K rated transformers dedicated for data.
More than 90% of the projects that I design don't need additional neutrals, K rated transformers, isolated grounds, etc. However I am just a small time engineer.
Jim T
ron
Senior Member
- Location
- New York, 40.7514,-73.9925
Re: harmonics in a 208V service
Bob,
I admit that I regularly specify double neutrals for feeders out of a power distribution unit (PDU) to the power panel (RPP), as I do not know what kind of the load the client will have in the future. I do not specify IG circuits anymore (for the last 5-6 years or so).
I have found problems in older sites that were designed for a smaller load density, which is now being used for higher.
For example, who would have thought 10 years ago, when designing a datacenter, that some of the loads would be 10-12kW per server rack. Now we are happy that we oversized at that time.
For that same reason, I try to design a project with spare underground conduits, and redundancy, because you never know .....
Bob,
I admit that I regularly specify double neutrals for feeders out of a power distribution unit (PDU) to the power panel (RPP), as I do not know what kind of the load the client will have in the future. I do not specify IG circuits anymore (for the last 5-6 years or so).
I have found problems in older sites that were designed for a smaller load density, which is now being used for higher.
For example, who would have thought 10 years ago, when designing a datacenter, that some of the loads would be 10-12kW per server rack. Now we are happy that we oversized at that time.
For that same reason, I try to design a project with spare underground conduits, and redundancy, because you never know .....
Re: harmonics in a 208V service
I have designed DATA centers for 20+ some years, and collected load data on all of them. I can only think of 2 maybe 3 PDU's where the neutral current exceeded 100%, but never over 110%. I use UPS with "K" rated PDU's out on the floor to limit VD. Of those 2 0r 3 the phase current was above 90%. My experience tells me neutral current exceeding phase current is a very rare bird.
[ December 03, 2004, 05:16 PM: Message edited by: dereckbc ]
Jim, I think you just made my case. If the feeder's or service neutral isn't 100% or more, you have a wink link in the chain. I agree going at least 100% is good design practice for most office work, as it will aid in VD, but no more than 100% unless you have something special like a DATA center. I only use 150% on cable. But I sqeall about it to my company stating it is a waiste.
I have designed DATA centers for 20+ some years, and collected load data on all of them. I can only think of 2 maybe 3 PDU's where the neutral current exceeded 100%, but never over 110%. I use UPS with "K" rated PDU's out on the floor to limit VD. Of those 2 0r 3 the phase current was above 90%. My experience tells me neutral current exceeding phase current is a very rare bird.
[ December 03, 2004, 05:16 PM: Message edited by: dereckbc ]
ron
Senior Member
- Location
- New York, 40.7514,-73.9925
Re: harmonics in a 208V service
Unfortunately I don't have hard and fast data to report. Just anecdotal instances.
Unfortunately I don't have hard and fast data to report. Just anecdotal instances.
Re: harmonics in a 208V service
Have never specified a neutral larger than the phase wires. Have never specified a K-Rated xfmr - (a 75KVA that says 45KVA on the nameplate ? ha).
Properly sized, properly bonded, & properly grounded (earthed) 208Y/120V. systems do the job just fine IMO.
Have never specified a neutral larger than the phase wires. Have never specified a K-Rated xfmr - (a 75KVA that says 45KVA on the nameplate ? ha).
Properly sized, properly bonded, & properly grounded (earthed) 208Y/120V. systems do the job just fine IMO.
- Location
- Windsor, CO NEC: 2017
- Occupation
- Service Manager
Re: harmonics in a 208V service
I'm going to whip it out, lay it on the table. My ignorance. What are harmonics, in a nutshell, and what happens on account of them?
I'm thinking that it has a relationship to unbalanced loads since you guys are talking about 200% sizing a neutral.
Another thought that may or may not be related to what ya'll are talking about: In a much simpler setting, let's say I pull a 14-3 home run to a duplex. Appliance A pulls 11 amps, Appliance B pulls 8 amps. This tells me the neutral is carrying 3 amps back to the panel. When the two circuits spent load collide in the receptacle neutral, that is where that subtraction takes place, right? They cancel each other out there?
I bask in the light of your collective knowledge. Sorry for dumbing this thread down a touch...
I'm going to whip it out, lay it on the table. My ignorance. What are harmonics, in a nutshell, and what happens on account of them?
I'm thinking that it has a relationship to unbalanced loads since you guys are talking about 200% sizing a neutral.
Another thought that may or may not be related to what ya'll are talking about: In a much simpler setting, let's say I pull a 14-3 home run to a duplex. Appliance A pulls 11 amps, Appliance B pulls 8 amps. This tells me the neutral is carrying 3 amps back to the panel. When the two circuits spent load collide in the receptacle neutral, that is where that subtraction takes place, right? They cancel each other out there?
I bask in the light of your collective knowledge. Sorry for dumbing this thread down a touch...
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
Re: harmonics in a 208V service
The biggest problems with harmonics tend to occur when AC to DC rectification is use. The devices (such as drives and power supplies) turn on the power from the line only as they need it, and usually for only fractions of a cycle at a time.
Normal loads tend to use the incoming power over the entire cycle.
The biggest problems with harmonics tend to occur when AC to DC rectification is use. The devices (such as drives and power supplies) turn on the power from the line only as they need it, and usually for only fractions of a cycle at a time.
Normal loads tend to use the incoming power over the entire cycle.
Ed MacLaren
Senior Member
Re: harmonics in a 208V service
These sketches may be of assistance to those who are just beginning their study of harmonics in power systems.
Harmonics result from loads that have a rapidly changing impedance, (electronically switched loads) with the result that they draw a current with a wave form that is not linear, meaning it does not match the voltage waveform.
A simple example of such a load is the common lamp dimmer. The triac can be switched on part way through each cycle in such a way that the "on time" versus "off time" can be varied, thus varying the voltage to the lamp.
A 3 phase, 4 wire system presents a real problem because triplen harmonics (a multiple of three times the fundamental) from all three phases are in phase with each other. They add instead of canceling, and become unbalanced load, which must be carried by the neutral conductor.
Ed
These sketches may be of assistance to those who are just beginning their study of harmonics in power systems.
Harmonics result from loads that have a rapidly changing impedance, (electronically switched loads) with the result that they draw a current with a wave form that is not linear, meaning it does not match the voltage waveform.
A simple example of such a load is the common lamp dimmer. The triac can be switched on part way through each cycle in such a way that the "on time" versus "off time" can be varied, thus varying the voltage to the lamp.
A 3 phase, 4 wire system presents a real problem because triplen harmonics (a multiple of three times the fundamental) from all three phases are in phase with each other. They add instead of canceling, and become unbalanced load, which must be carried by the neutral conductor.
Ed
- Location
- Massachusetts
Re: harmonics in a 208V service
George I am not much on theory but I think I can give you it in a 'nutshell'
Some types of loads, like computers for instance are called non-linear loads. On 240/120 service these loads do not make a bit of differers.
With 3 phase service 208/120 these loads act in a way that can increase current on the neutral.
To prevent this from over loading the neutral many times the engineers will specify larger neutral conductors.
A multiwire branch circuit may have three 12 AWG ungrounded and one 10 AWG grounded conductors.
The panel feeding these branch circuits will be ordered with a 200% rated neutral.
So a typical 225 amp panel wil have a 450 amp rated neutral bar. To this panel we will pull six conductors three ungrounded, two grounded all of the same size, 4/0 copper or 250 kcmil would be typical along with a grounding conductor.
Bob
George I am not much on theory but I think I can give you it in a 'nutshell'
Some types of loads, like computers for instance are called non-linear loads. On 240/120 service these loads do not make a bit of differers.
With 3 phase service 208/120 these loads act in a way that can increase current on the neutral.
To prevent this from over loading the neutral many times the engineers will specify larger neutral conductors.
A multiwire branch circuit may have three 12 AWG ungrounded and one 10 AWG grounded conductors.
The panel feeding these branch circuits will be ordered with a 200% rated neutral.
So a typical 225 amp panel wil have a 450 amp rated neutral bar. To this panel we will pull six conductors three ungrounded, two grounded all of the same size, 4/0 copper or 250 kcmil would be typical along with a grounding conductor.
Bob
- Location
- Windsor, CO NEC: 2017
- Occupation
- Service Manager
Re: harmonics in a 208V service
So, moderately understanding this concept, why is it no longer a concern?
In essence, the unbalanced loads could potentially damage or burn up a neutral?
So, moderately understanding this concept, why is it no longer a concern?
In essence, the unbalanced loads could potentially damage or burn up a neutral?
Re: harmonics in a 208V service
Hello-
Hope that these two references will contribute to this thread:
Thomas Gruzs described neutral currents as high as 173% in his 1990 paper titled "A Survey of Neutral Currents in Three-Phase Computer Power Systems", published in the IEEE Transactions on Industry Applications. (Vol. IA-26, No. 4, July 1990)
This paper is the source of the 200% neutral rule of thumb, and should be available in the library at Universities that have an electrical engineering department.
Also, an excellent introduction to harmonics can be found at this link:
http://www.cda.org.uk/megab2/elecapps/pub145lo.pdf
Mark
Hello-
Hope that these two references will contribute to this thread:
Thomas Gruzs described neutral currents as high as 173% in his 1990 paper titled "A Survey of Neutral Currents in Three-Phase Computer Power Systems", published in the IEEE Transactions on Industry Applications. (Vol. IA-26, No. 4, July 1990)
This paper is the source of the 200% neutral rule of thumb, and should be available in the library at Universities that have an electrical engineering department.
Also, an excellent introduction to harmonics can be found at this link:
http://www.cda.org.uk/megab2/elecapps/pub145lo.pdf
Mark
- Status