Double feed an outlet loop?

[matt_i]

So I'm wiring 3ph outlets in the new shop. Its a 1" trunk line going around the perimeter of the attic with a bunch of 3/4 drops teed into it that drop down to the floor. One common circuit right now feeding all outlets, they all go hot at the same time.

Shop is 25 x 40 so for the very last outlet, it goes thru roughly 100 more feet of conduit after the first outlet.

But I was thinking it would be easy to cut in another tee at this point to make a continuous loop in the trunk line and then I could "double feed" the loop from both ends to reduce voltage drop. That would put the middle outlet roughly 50 feet from the first one, no matter which direction of travel is considered around the trunk line.

The mechanical analogy would be a shop air system where it would be desirable to feed a loop from both ends to reduce starvation of flow.

Seems logical to me although maybe not traditional. I do recall a 1600A x 480vac welding bus being fed from both ends via tie-breaker disconnects just in case of problems but I can't recall if the tie breakers were ever both "connected" at the same time.

For my case, the extra work involved is minimal at this point. Wanted to see about violations or other functional problems with the idea. All the phases are carefully identified don't want to "cross the streams" you know....

 

[sberry]

What is it you are going to do that you think starvation is going to be a problem? It's early and I havnt had a coffee yet so I havnt thought this thru as I never considered it but I am amazed at some stuff we can come up with here.

 

[Norcal]

So your wanting a version of the British ring circuit? Not quite the same but I would not do it.

 

[wyliesdiesels]

So what kind of 3ph do you have?

 

[nsula_country]

I believe he wants to make the conduit in a continuous loop and split the wiring at the 1st tee to reduce wire. Not "Double Ended Feed" in the traditional sense for redundancy.

CT

 

[matt_i]

I was just thinking of reducing the potential voltage drop across the total length of wiring a single closed-ended run vs. energizing the "loop" and letting the current find the natural path of least resistance (the 100' vs 50' to the "farthest outlet"). Its all #10 awg copper stranded THHN.

A side benefit of the full conduit loop is that I can potentially save wire for a future "home run"- unbroken circuit using higher awg wire, but thru the same trunk line. I don't have a specific need planned as of today, but life is hard to get 100% correct because the crystal ball is too hazy.

Its Phase Perfect "generated" 3 phase 240vac, nothing happens until the converter is turned on by yours truly.

 

[teamextreme]

I was just thinking of reducing the potential voltage drop across the total length of wiring a single closed-ended run vs. energizing the "loop" and letting the current find the natural path of least resistance (the 100' vs 50' to the "farthest outlet"). Its all #10 awg copper stranded THHN.

A side benefit of the full conduit loop is that I can potentially save wire for a future "home run"- unbroken circuit using higher awg wire, but thru the same trunk line. I don't have a specific need planned as of today, but life is hard to get 100% correct because the crystal ball is too hazy.

Its Phase Perfect "generated" 3 phase 240vac, nothing happens until the converter is turned on by yours truly.

That is a common misconception. Electricity does not take the "path of least resistance". The current flow through 2 paths of differing resistance will be different, with the path of least resistance seeing proportionally more current that the more resistive path, but it won't be a either/or situation, with the current flowing 100% through the lower resistance path and 0% through the higher path.

Your proposed arrangement is intriguing and I've never seen anything set up like this other than redundant rings for networking, etc. Or as someone mentioned, a double-fed switchgear, but those generally always have tie-breakers or ATS that only allow one feed at a time. I'm not sure if this arrangement would be legal or not.

 

[ForceFed70]

OP: I wouldn't do it. It would confuse someone else in the future and while I'm no code expert, I suspect code wouldn't allow it.

That is a common misconception. Electricity does not take the "path of least resistance". The current flow through 2 paths of differing resistance will be different, with the path of least resistance seeing proportionally more current that the more resistive path, but it won't be a either/or situation, with the current flowing 100% through the lower resistance path and 0% through the higher path. .

While what you are saying is correct "Electricity takes the path of least resistance" is also still correct and not a misconception. The better way to look at this is that as electricity flows through a conductor it's effective resistance grows. When 2 paths exists, that effective resistance can result in electricity taking the 2nd path as well. In this case, the flow through each path balances such that each path has identical effective resistance.

 

[Radix2]

OP:

While what you are saying is correct "Electricity takes the path of least resistance" is also still correct and not a misconception. The better way to look at this is that as electricity flows through a conductor it's effective resistance grows. When 2 paths exists, that effective resistance can result in electricity taking the 2nd path as well. In this case, the flow through each path balances such that each path has identical effective resistance.

How is your made up gobbledygook "effective resistance" a better way to look at it? Resistance grows?

Given a parallel path, the voltage across each conductor set is the same, the current through each path will be I=v/r. The total current is the sum from each conductor. What is " balanced" is voltage... because it is the same...

What you are trying to say with effective resistance.... seems to be some backwards kinda idea that goes back to voltage drop and all, but that is surely a confusing way to think about what is going on... imo.

Sorry.


I don't know if there is any code against parallel feeds, maybe one of the guys can supply a reference if any. What occurs to me, but is unclear from the op is what would be the circuit length from the feed to the point where the loop would attach? Would you simply be better off feeding some outlet from that direction and save the wire from the other direction?

 

[wyliesdiesels]

Im still trying to figure out why the OP wouldnt just feed the further outlets with a separate shorter run of wire since he said the opposite end of the loop is shorter wire...

This is another case of making something simple harder than it has to be.... over thinking and over engineering wastes time and money...

 

[matt_i]

While what you are saying is correct "Electricity takes the path of least resistance" is also still correct and not a misconception. The better way to look at this is that as electricity flows through a conductor it's effective resistance grows. When 2 paths exists, that effective resistance can result in electricity taking the 2nd path as well. In this case, the flow through each path balances such that each path has identical effective resistance.

I mean I get that it's a voltage divider and current will flow proportional to (micro) resistance....as opposed to a "digital" flow path that's either on or off. I probably misrepresented with my english description.

Im still trying to figure out why the OP wouldnt just feed the further outlets with a separate shorter run of wire since he said the opposite end of the loop is shorter wire...

This is another case of making something simple harder than it has to be.... over thinking and over engineering wastes time and money...

Since the outlets are parallel daisy-chained there's no big saving of actual wire other than the last piece that makes up the "ring" from the "open ended loop". The wire just goes from outlet to outlet. Hopefully I didn't misrepresent that. I know diagrams would help and I don't have any posted here.

Anyway, I've decided I'll loop the conduit but not install the last piece of wire. Then I can do some simple voltage drop testing with a machine hooked up to the very last outlet with a mechanical load. If there's an issue I can detect I'll hook up the last piece of the ring.

My labor is still accounted for as "free" for my own projects

 

[rust in the eye]

OP is referring to it as a loop so I suspect not intending to split the loop.
Another has suggested this arrangement, if legal, may confuse others in the future and with this I agree wholeheartedly.
Connecting another feed and interrupting the loop at the furthest point to create two shorter runs would be how I would do it.

 

[75gmck25]

It seems that a simpler scheme would work, and be less confusing for anyone troubleshooting in the future.

- Main feed from panel running to a junction box
- Split the feed at that box and go left and right around the attic.
- Left and right feeds meet (almost) at the far side of the attic, with a gap between the two farthest junction boxes (no connection).

Any receptacle/box will be no farther than 1/2 the distance around the attic from the first junction box, which should limit voltage drop. It also gives you the option of adding a 2nd feed from the panel and turning it into two separate (left and right) circuits.

Bruce

 

[ddawg16]

It sounds to me like the OP wants to feed his loop from both ends.

In other words....load center to point A....then point B.....then point C....back to load center.

Bad idea IMO

 

[Norcal]

It sounds to me like the OP wants to feed his loop from both ends.

In other words....load center to point A....then point B.....then point C....back to load center.

Bad idea IMO

It's pretty similar to a British ring circuit, starts & ends at the 32A circuit breaker in the " consumer unit" all the items plugged in have fused plugs.

 

[ddawg16]

It's pretty similar to a British ring circuit, starts & ends at the 32A circuit breaker in the " consumer unit" all the items plugged in have fused plugs.

Those fused plugs were a PIA.

And most outlets also had an on/off switch.....that people turned off when they went to bed.