Branch Circuits: What is and isn't?

[physicsboard]

I'm evaluating an electrical enclosure for a power-plant. Power plants are not required to follow CSA standards, however this electrical enclosure does anyway. Upstream of the enclosure is a rated breaker, and this is what is currently defined as the 'branch.' However, inside of the enclosure there are several elements being powered in parallel, with supplementary breakers for over-current protection. The issue was raised as to whether these parallel loads also required branch rated breakers, and my research into the issue says this by and large depends on whether these parallel loads are considered branches or not. Branches require rated breakers for short circuit protection is the long and short of it.

My biggest problem with the definition is the fact that it almost always makes reference to a breaker, or an outlet. Since the code requires all branch circuits to have branch rated protection, it gets confusing when that same protection enters the definition of the branch. I'll give you a few examples

The portion of an electric wiring system that extends beyond the final overcurrent device (such as a fuse) protecting the circuit.

Wiring between the last overcurrent device and the branch circuit outlets.

A portion of a wiring system in the interior of a structure that extends from a final overload protective device to a plug receptable or a load such as a lighting fixture, motor, or heater.

The portion of an electrical running between the breaker or fuse and the devices it powers. Can serve a single device or several.

When I found that last one it seemed like assuming the entire enclosure was a single branch was legitimate. But I haven't found anything reaffirming that, and I haven't managed to find an adequate definition in any of the codes I've looked at, primarily the Canadian Electrical Code, which defines it as:

That portion of the wiring installation between the final over current device protecting the circuit and the outlets.

Fairly similar to the previous.

Can anyone shed some light on this definition?

 

[FOIWATER]

it always seemed to make sense to me, maybe I'm over simplifying it though, or possibly you're over complicating it.

I always took the branch circuit to be the actual wiring of the circuit exterior to the panel.

Since the breakers or fuses or any overcurrent protection device is inclusive to the panel, the branch circuit means, to me, to be the physical parts of the circuit external to the panel all the way to the outlet.

EDIT: I think the reference to breakers consistently is to drive home the fact that branches must be protected, to distinguish it from any other circuit honestly, and it also sets the boundaries for what the branch is inclusive of. For example, without any breakers feeding the branches (ludacris I know) how could we define the branch circuit? or where it begins? The codes are written by very well trained lawyers who often go into tremendous detail to outline everything in such a way it cannot be misunderstood, which ironically leads to just that.

 

[physicsboard]

it always seemed to make sense to me, maybe I'm over simplifying it though, or possibly you're over complicating it.

I always took the branch circuit to be the actual wiring of the circuit exterior to the panel.

Since the breakers or fuses or any overcurrent protection device is inclusive to the panel, the branch circuit means, to me, to be the physical parts of the circuit external to the panel all the way to the outlet.

This makes a lot of sense to me. Before I looked into this definition I was assuming any time the circuit 'branched' off to a load, it was a branch circuit, so these definitions left me somewhat confused.

In this case, there's no outlet, so would you simply consider the branch to be the stuff between the device and the panel?

 

[FOIWATER]

This makes a lot of sense to me. Before I looked into this definition I was assuming any time the circuit 'branched' off to a load, it was a branch circuit, so these definitions left me somewhat confused.

In this case, there's no outlet, so would you simply consider the branch to be the stuff between the device and the panel?

Yes, any parallel feeds off the breaker as well

Everything really, They reason the 'lawyers' say Outlet, would tell me they don't want to illustrate any load you connect as being part of the residential branch circuit, as they don't inspect the wiring for it.

 

[FOIWATER]

That is to say, any parallel feeds from one breaker, all combine to constitute one single branch circuit**

Welcome to the forums by the way..

 

[physicsboard]

Thank you for taking the time to answer my question, and thanks for welcoming me to the forums. Since I understand how these branches are actually defined, I'm sure I can explain this to others when branch circuits come into question again.

 

[FOIWATER]

Well I'm glad I could help sincerely.

 

[jim hardy]

Sometimes we can clarify such questions by thinking backward.

What we are protecting with the 'branch breaker' is not just the wires leaving it.

A fault on the branch must not be allowed to collapse the source on upstream side of branch breaker, else all the other branches will be affected. Computer industry learned this the hard way. Co-ordination is the buzzword for that concept.

So while multiple individual loads on a branch circuit may have their own overcurrent protection that doesn't negate the 'branch' designation imho. Further those individual loads are obliged by same logic co-ordinate with their branch breaker.

It is sort of like a fractal. Which distribution drawings resemble.

Does that make any sense ?

 

[FOIWATER]

Jim you are saying if you have a panel breaker feeding multiple parallel circuit, and each parallel path was equipped with say a fuse, the terminology lends itself towards calling this part of one (1) single branch circuit?

I agree with you, if I understand you

 

[physicsboard]

So while multiple individual loads on a branch circuit may have their own overcurrent protection that doesn't negate the 'branch' designation imho. Further those individual loads are obliged by same logic co-ordinate with their branch breaker.

The question I am trying to answer is whether these parallel loads are a branch off a branch, or whether they are simply part of the first branch. If they are a part of the first branch, all of these parallel loads would not require branch rated protection. If they are considered branches then rated protection is required.

I don't really follow what you mean here... Do you think that each branch should have it's loads evaluated and if deemed necessary to split into further branches?

 

[jim hardy]

I don't really follow what you mean here... Do you think that each branch should have it's loads evaluated and if deemed necessary to split into further branches?

yes.
If you can afford to lose power to everything on the branch then no further downstream co-ordination is needed.
Say there are only two loads on your branch. One of them is very important and the other trivial. The trivial load needs to not trip the branch breaker when it faults for that would interrupt power to the vital one, making what should have been a trivial event into a non-trivial one..

To my simple thinking that warrants at least evaluating the "what if's" and either co-ordinating or making another branch.
If you can co-ordinate the [STRIKE]fuse[/STRIKE] overcurrent device that's inside your trivial load so it doesn't trip the branch breaker you needn't make a separate "branch" for it. But you should place on that device a placard "Use only MDL-3/8 fuse" or whatever.

I appreciate you have a legal definition of "branch" to deal with. I am only thinking of the logical concept. Failure to co-ordinate can wreak havoc with operations.

Shawmut form 101 fuses are very fast if you need to co-ordinate with fast breakers or inverters.
Electronic inverters are very difficult to co-ordinate because of their limited fault current, only 150 to 200% nameplate. We are accustomed to transformers that'll deliver 10X that.

old jim