Double Neutrals

[Bill4477]

I have been told that double neutrals or 200% neutral conductor sizes are required for harmonic mitigating transformers (HMT's). I do not recall ever using a double neutral on any job. Can someone tell me whether double neutrals should ever be required and what the circumstances are that make them a requirement?

 

[GoldDigger]

I have been told that double neutrals or 200% neutral conductor sizes are required for harmonic mitigating transformers (HMT's). I do not recall ever using a double neutral on any job. Can someone tell me whether double neutrals should ever be required and what the circumstances are that make them a requirement?

I have not heard that mentioned, but I can see that on the load side of the HMT both code and theory would tell you that you would need to have an ovesized neutral since that wire may be carrying both the unbalanced current at the fundamental frequency and the sum of the individual triplen harmonic currents.
A factor of 2 is probably conservative.
On the line side of the HMT, with reduced harmonic content, I would expect that the neutral could be normal sized, having to deal only with the unbalanced current at the fundamental frequency.

 

[ron]

By placing a HMT or a K rated transformer (K-20 or larger IMHO), you are acknowledging that significant harmonics exist. You would likely also need to consider the neutrals as CCC and derate the circuit per Table 310.15(B)(3)(a) whether you doubled them or not.

 

[JDBrown]

I don't know if they're generally required by manufacturers or not, but most specs I've run into require double-sized neutrals and neutral buses on the secondary side of HMTs. I agree that it makes good sense to do it, because you need an oversized neutral if you've got harmonics, and why would you go to the expense of buying an HMT if you didn't have harmonics to contend with?

 

[charlie b]

A factor of 2 is probably conservative.

I have a vague "memory" (if I can justify using that word at my age :roll: ) of someone showing that the maximum current you will see in the neutral, including all harmonic content, would be the square root of three (about 1.732) times the phase current. Can anyone confirm or refute that statement? :?

 

[GoldDigger]

Ah, the famous square root of three, which approaches two for large values of three.
:angel:

Tapatalk!

 

[JDBrown]

I have a vague "memory" (if I can justify using that word at my age :roll: ) of someone showing that the maximum current you will see in the neutral, including all harmonic content, would be the square root of three (about 1.732) times the phase current. Can anyone confirm or refute that statement? :?

I've also read this, and my boss makes reference to it quite often, but I've never seen the actual proof. Sorry, I guess that's not much help.

 

[GoldDigger]

If you take the worst possible case, namely a hypothetical load which draws nothing but third harmonic current, then the current in the neutral would be exactly three times the balanced phase currents.
I think that the 2x number comes from a pragmatic estimate of the worst you would get with a real load.
Same for the sqrt(3). I cannot imagine a proof of that without specifying an exact load

Tapatalk!