Why do you need 200% neutral for non-linear loads?
- Thread starter cppoly
- Start date
- Status
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
I believe that the 200% neutral is a job spec, not a code requirement.
When you have non-linear loads, that means that you apply a pure sine voltage to the load and non-sine current flows. Non-sine current can be understood as the sum of the fundamental and harmonic currents. It turns out that for a balanced three phase system, all 3x harmonics _add_ rather than cancel on the neutral.
If you had a perfectly balanced system, with the same line current on each phase but with 33% 3rd harmonic, then the neutral current would equal the line current.
However if you had an absolutely pathological and virtually impossible situation, where the loads were so non-linear that only 3rd harmonic current flows, then the neutral current would be the sum of the line currents, that is for balanced loading the neutral current would be 3x the line current.
(Note: if you apply 60Hz voltage to a load, and it is so pathological that only 180Hz third harmonic current actually flowed, then no real power would be delivered to the load. So unless you specifically design loads intended to damage electrical systems without doing any useful work, the above described situation is impossible.)
-Jon
When you have non-linear loads, that means that you apply a pure sine voltage to the load and non-sine current flows. Non-sine current can be understood as the sum of the fundamental and harmonic currents. It turns out that for a balanced three phase system, all 3x harmonics _add_ rather than cancel on the neutral.
If you had a perfectly balanced system, with the same line current on each phase but with 33% 3rd harmonic, then the neutral current would equal the line current.
However if you had an absolutely pathological and virtually impossible situation, where the loads were so non-linear that only 3rd harmonic current flows, then the neutral current would be the sum of the line currents, that is for balanced loading the neutral current would be 3x the line current.
(Note: if you apply 60Hz voltage to a load, and it is so pathological that only 180Hz third harmonic current actually flowed, then no real power would be delivered to the load. So unless you specifically design loads intended to damage electrical systems without doing any useful work, the above described situation is impossible.)
-Jon
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
You're missing that the third harmonics are in phase w/ each other and add.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
Actually I take by my comment about impossible in practice. Consider a single phase non-linear load where current is flows as a pulse only during the peak of the sine wave (say from 80-100 and 260-280 degrees of the sine function). A set of 3 of these loads fed line-neutral on a 3 phase wye system would _never_ conduct at the same time. All current flowing down line A returns on the neutral, same for B and C. In this case the neutral load would be 3x the line current and real power would be delivered to the load.
-Jon
-Jon
True for single phase non-linear loads.You're missing that the third harmonics are in phase w/ each other and add.
Carultch
Senior Member
- Location
- Massachusetts
To confirm I understand what you are saying:
As an absolute worst case scenario of nothing but 180Hz 3rd harmonics and little (if any) of the fundamental 60Hz, it could be possible to need a 300% neutral. However, it is an extremely improbably situation in practice, and a 200% neutral fits what is typically needed for practical cases of harmonic-intensive loads.
What if all your loads are similar to VFDs of motors, where harmonics are abundant, but triplen harmonics systematically never happen? Would neutral still have to count as a CCC, and would there be any need for a neutral larger than the phase wiring?
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
You are correct in your description of the underlying math/physics. But the language of the code is not that nuanced. It does not refer to triplen harmonics or any particular type of non-linear load. Instead it tells you that if the degree of non-linear loading is high enough (but without a specific calculable threshold) then the neutral must be counted as a current carrying conductor. And the neutral must be correctly sized for the load, which may be larger than 100% of the ungrounded conductor amps in the case of some specific non-linear loads.
Depends on whether the VFDs have a single phase or three phase supply.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
In the absolute worst case you have 3 180Hz sine waves adding together, and yes the neutral current is 300% of the line current.
As was pointed out in a private message however, things get more complicated when you consider RMS current calculations for more plausible scenarios. I posited some sort of rectifier load with conduction only at the peak of the sine waves; in this case the neutral current would be sqrt(3)* the line current. I suspect that this is the worst case practical scenario, a balanced set of single phase rectifier loads with no attempt at harmonic correction.
Interesting thought. I think that any large VFD would have its rectifiers connected line-line and would not put any current at all on the neutral. If you had a large number of small VFDs connected line-neutral, then you might have a problem. If you are describing any sort of hypothetical 3 phase non-linear load, you are almost certainly talking about something without a neutral connection.
-Jon
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
Good thought. Although a VFD may have line to neutral surge suppression elements connected internally, it makes sense that its rectifier bridge be connected line to line to provide a six pulse output instead of the three pulse that you would get running line to neutral with a single diode on each line. Theeoutput voltage would be higher too.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
VFD's typically don't have any neutral connection on the power circuit, control circuit maybe, but the magnitude of current of control circuit is nothing in comparison to the power circuit, pretty rare that any drive controls are not low volts anyway, just some aux contacts that could switch some externally derived controls at the most for line volt controls.
I think for the most part, unless you have a large data center, the chances of needing a 200% neutral are not really all that great.
I think for the most part, unless you have a large data center, the chances of needing a 200% neutral are not really all that great.
- Status