Looped Power Feed?

Status
Not open for further replies.
J

JFletcher

Senior Member
Location
Williamsburg, VA
Was doing a LV landscaping lighting system Friday, and was thinking about voltage drop as the runs were fairly long. I was thinking also that running the cable in a loop back to the driver might be a way to reduce VD. Example: you have 15 10W 12VDC landscape lights that use 300' of cable. Say the last light is 20' from the driver; could you run from that light 20' back to the driver to effectively have 160' of run vs 320'? Would this be an NEC violation? Would this be a violation with a string of 120V receptacles? What about interconnected smoke alarms?

Ive heard this was done years ago in communications to improve reliability - if you had an open pair on one end, your system would still work because it was a complete circuit from the other. I'm not talking about using two drivers or breakers to feed one circuit, but a circuit that is complete in a circle-like pattern.
 
There are differences of opinion on this but my own opinion it puts the conductors in parallel and that would be a NEC violation in these wire sizes.

A lot of folks would call it a ring circuit and it either was or is used in some countries for receptacle circuits.

They feed a 16 amp wire loop with a 32 amp breaker and have 16 amp fuses in each receptacle.
 
151205-2359 EST

JFletcher:

If you keep the wire size the same, then you will reduce voltage drop.

iwire has to help you on NEC. See if he can give you any firm answer.

.
 
gar said:
iwire has to help you on NEC. See if he can give you any firm answer
Click to expand...

In my opinion it really comes down to how the local AHJ / inspector interprets 'electrically joined at both ends'.

To me 'electrically joined' has a much different meaning then 'physically joined' at each end. To me 'electrically joined' would mean it would apply to a ring circuit.

That being said, I understand others feel differently and certainly some AHJs and inspectors could be fine with it.



310.10(H) Conductors in Parallel.
(1) General. Aluminum, copper-clad aluminum, or copper
conductors, for each phase, polarity, neutral, or grounded
circuit shall be permitted to be connected in parallel (electrically
joined at both ends) only in sizes 1/0 AWG and
larger where installed in accordance with 310.1 0(H)(2)
through (H)(6).

Exception No.1: Conductors in sizes smaller than 1/0 AWG
shall be permitted to be run in parallel to supply control
power to indicating instruments, contactors, relays, solenoids.
and similar control devices, or for frequencies of 360 Hz and
higher; provided all of the following apply:

(a) They are contained within the same raceway or
cable.

( b) The ampacity of each individual conductor is sufficient
to carry the entire load Current shared by the parallel
conductors.

(c) The overcurrent protection is such that the ampacity
of each individual conductor will not be exceeded if one
or more of the parallel conductors become inadvertently
disconnected.


Exception No.2: Under engineering supervision, 2 AWG
and 1 AWG grounded neutral conductors shall be permitted
to be installed il1 parallel for existing installations.
Click to expand...


Code aside, I think it is a good solution for landscape lighting and some other applications.
 
Last edited:
ty gar, iwire, and ActionDave. re: 310.10(H), would that apply to DC systems? "for each phase, polarity, neutral, or grounded
circuit.." are AC terms. Also, didnt know w/o reading all of Chapter 7 if a ring circuit was permissible for FA or LE installations.

istm if you are required to have smoke alarms in a residence, then interconnected (A to B to C to D to...) would be a step above code, and that a loop or ring wiring (A to B to C to D... back to A) would provide added security... on the red (signal) wire anyway.

For the landscape lights, I was thinking that a loop would reduce the differences in lumen output due to voltage drop rather for any reliability reasons. I could have run 'star topography' and had one or two j-boxes with 8-15 splices to make all cable lengths from driver to lights roughly equal, however that would have been a ton of extra trenching and cable. Dunno if this (ring circuit) would be overkill or not, or if there will be a noticeable difference in the lights at night; I have not seen them in full darkness yet.

I also see where it can be considered parallel conductors, however I think it a bit different here than say trying to run 2 14/3 to a dryer from a 30A breaker (which is obviously a violation).
 
Last edited:
JFletcher said:
Was doing a LV landscaping lighting system Friday, and was thinking about voltage drop as the runs were fairly long. I was thinking also that running the cable in a loop back to the driver might be a way to reduce VD. Example: you have 15 10W 12VDC landscape lights that use 300' of cable. Say the last light is 20' from the driver; could you run from that light 20' back to the driver to effectively have 160' of run vs 320'? Would this be an NEC violation? Would this be a violation with a string of 120V receptacles? What about interconnected smoke alarms?

Ive heard this was done years ago in communications to improve reliability - if you had an open pair on one end, your system would still work because it was a complete circuit from the other. I'm not talking about using two drivers or breakers to feed one circuit, but a circuit that is complete in a circle-like pattern.
Click to expand...

Where I work, a LARGE portion of the work that moves through our shop is 12V or 24V LED light boxes for television studios; plywood box 3-6" deep containing many runs of LED tape, with frosted plexi and sometimes a graphic applied. Our tape comes in 5m rolls, and we almost always feed the LED tape from both ends of the runs to limit voltage drop. Voltage drop is less of an issue with 24V tape, but still can cause issues with consistent brightness.

I had to use some special LED tape on a recent project that required 5V. It came in 2.8m lengths and the manufacturer demands the tape be fed from both ends. Much grumbling from the shop staff ensued, as it significantly increased wiring complexity on the project. It drastically complicated the power supply situation as well.

With your landscape lighting, would it be possible to somewhat mitigate your voltage drop problem by using LED fixtures, or are you already doing that? More less light for less watts. Around here, if it's low voltage and especially if it's a class 2 power supply, the inspectors pretty much ignore it.

To directly answer your question, I think it depends on what you're doing:

If I were doing LV landscaping lights, I'd have no problem doing it; I wouldn't see it as presenting any safety concern, since the voltage is typically 12-24V.

120VAC receptacles are a different story in my book; a ring circuit could absolutely present a safety concern simply because most electricians in N. America wouldn't be expecting it. The circuit could still be live, even if one feed were damaged/disconnected.


SceneryDriver
 
JFletcher said:
ty gar, iwire, and ActionDave. re: 310.10(H), would that apply to DC systems? "for each phase, polarity, neutral, or grounded
circuit.." are AC terms.
Click to expand...

Those are both AC and DC terms, 310.10 applies to both.

Also, didnt know w/o reading all of Chapter 7 if a ring circuit was permissible for FA or LE installations.

istm if you are required to have smoke alarms in a residence, then interconnected (A to B to C to D to...) would be a step above code, and that a loop or ring wiring (A to B to C to D... back to A) would provide added security.
Click to expand...

Keep in mind that the line voltage smoke alarms in a home are not a fire alarm system covered by article 760.

Home smoke alarms with battery back up have some supervision for AC or DC power loss. I don't see much to be gained with a ring circuit here.



For the landscape lights, I was thinking that a loop would reduce the differences in lumen output due to voltage drop
Click to expand...

Yes, I agree it would.

It is something I would want to do but I feel it may be a violation to do so. Not unsafe, just a violation.


A side note, with the LV lighting start with article 411.:)
 
SceneryDriver said:
If I were doing LV landscaping lights, I'd have no problem doing it; I wouldn't see it as presenting any safety concern, since the voltage is typically 12-24V.
Click to expand...

Wouldn't current, not voltage be the safety issue?

Many LV lighting power supplies do put out 25 amps.
 
iwire said:
In my opinion it really comes down to how the local AHJ / inspector interprets 'electrically joined at both ends'.

To me 'electrically joined' has a much different meaning then 'physically joined' at each end. To me 'electrically joined' would mean it would apply to a ring circuit...
Click to expand...
I agree the code is murky here. For 24V landscape lighting I wouldn't care. Other than the transformer and the circuit that feeds it my inspector wouldn't either.
 
I submitted a PI that was rejected that would have permitted small conductors in parallel.
Public Input No. 4435-NFPA 70-2014 [ New Section after 310.10(H)(1) ]
Exception Number 3
Exception Number 3. Conductors of any size or combination of sizes shall be permitted to be connected in
parallel where the circuit overcurrent protective device has a rating equal to of less than the ampacity of the
smallest paralleled conductor.
Statement of Problem and Substantiation for Public Input
Many of the states have adopted energy codes that include mandatory voltage drop requirements. This change
would permit non-conventional methods, such as ring circuits, to be used to satisfy these voltage drop
requirements There is no safety hazard where the OCPD has a rating that is equal to or less than the ampacity of
the smallest paralleled conductor.
Submitter Information Verification
Submitter Full Name: DON GANIERE
Organization: [ Not Specified ]
Street Address:
City:
State:
Zip:
Submittal Date: Thu Nov 06 22:00:02 EST 2014
Committee Statement
Resolution: PI4435 No study has been conducted to prove that this practice does not pose a safety hazard. The
submitted language would permit this practice for installations other than addressing voltage drop.
Existing exception #1 permits this practice for high frequency circuits with small electrical loads and it
provides specific requirements for these installations. Based on the conditions prescribed in exception
#1, and no testing done- the panel will continue to support the existing requirements. PI510 The
submitter failed to provide adequate testing to prove that the paralleling of the grounding electrode
conductor, system bonding jumper and main bonding jumper will not affect performance. The largest
size grounding electrode conductor required by the NEC is a 3/O copper which is only 2 sizes larger
than minimum required for parallel installations. The system and main bonding jumpers are typically
shorter in length and required to be sized 12 1/2% of the circular mil area of the ungrounded
conductors. PI3073 does not add clarity or usability to the Code.
Click to expand...
I fail to understand why you need a study for this. As long as the OCPD rating is equal to or less than the ampacity of the smallest conductor that is connected in parallel there is no safety hazard.
 
Run the loop and cut it in the middle that eliminates the parallel conductor problem.
 
iwire said:
There are differences of opinion on this but my own opinion it puts the conductors in parallel and that would be a NEC violation in these wire sizes.

A lot of folks would call it a ring circuit and it either was or is used in some countries for receptacle circuits.

They feed a 16 amp wire loop with a 32 amp breaker and have 16 amp fuses in each receptacle.
Click to expand...

It’s pretty obvious you have no interest in other regulations. Wrong, correct, wrong, wrong.
 
iwire said:
There are differences of opinion on this but my own opinion it puts the conductors in parallel and that would be a NEC violation in these wire sizes.

A lot of folks would call it a ring circuit and it either was or is used in some countries for receptacle circuits.

They feed a 16 amp wire loop with a 32 amp breaker and have 16 amp fuses in each receptacle.
Click to expand...

It is a 13 amp fuse located inside the cord cap, not a 16 amp fuse within the receptacle as BS1363 plugs are only rated 13 amps max.

The wire used is usually 2.5mm2 twin and earth cable which is good for 27amps clipped direct and 20amps touching thermal insulation.

Ring mains and BS1363 sockets have been discussed extensively in threads you participated in. I bring this up because you insisted RCDs nuisance trip even when applied correctly along with others even though foreign sparks know otherwise.
 
JFletcher said:
ty gar, iwire, and ActionDave. re: 310.10(H), would that apply to DC systems? "for each phase, polarity, neutral, or grounded
circuit.." are AC terms. Also, didnt know w/o reading all of Chapter 7 if a ring circuit was permissible for FA or LE installations.

istm if you are required to have smoke alarms in a residence, then interconnected (A to B to C to D to...) would be a step above code, and that a loop or ring wiring (A to B to C to D... back to A) would provide added security... on the red (signal) wire anyway.

For the landscape lights, I was thinking that a loop would reduce the differences in lumen output due to voltage drop rather for any reliability reasons. I could have run 'star topography' and had one or two j-boxes with 8-15 splices to make all cable lengths from driver to lights roughly equal, however that would have been a ton of extra trenching and cable. Dunno if this (ring circuit) would be overkill or not, or if there will be a noticeable difference in the lights at night; I have not seen them in full darkness yet.

I also see where it can be considered parallel conductors, however I think it a bit different here than say trying to run 2 14/3 to a dryer from a 30A breaker (which is obviously a violation).
Click to expand...

don_resqcapt19 said:
I submitted a PI that was rejected that would have permitted small conductors in parallel.

I fail to understand why you need a study for this. As long as the OCPD rating is equal to or less than the ampacity of the smallest conductor that is connected in parallel there is no safety hazard.
Click to expand...
...one thinks better on the beaches of Hawaii during the conference to study the effects:slaphead:
 
SceneryDriver said:
Where I work, a LARGE portion of the work that moves through our shop is 12V or 24V LED light boxes for television studios;
... Around here, if it's low voltage and especially if it's a class 2 power supply, the inspectors pretty much ignore it.

...
SceneryDriver
Click to expand...

I can't speak for where you are but around here, motion picture production is exempt from permitting & inspections. You need a permit to build a studio and for all the related offices, bathrooms, etc. but for anything on a set, they're exempt, unless anything changed in the past few years.
 
Some feel the intention of 310.10(H) when it says electrically joined at both ends is supposed to mean when done so for the purpose of effectively creating a larger current carrying capacity conductor.

If you have 12 AWG conductor in a "ring circuit" with 20 amp overcurrent protection how is it possible to have more then 20 amps in any portion of the circuit without starting to go into trip curve of the overcurrent device?

You have conductor segments that are parallel to one another but not nesessarily equally "in parallel", which is important if the intent is to create an effectively higher current carrying capacity conductor.
 
kwired said:
If you have 12 AWG conductor in a "ring circuit" with 20 amp overcurrent protection how is it possible to have more then 20 amps in any portion of the circuit without starting to go into trip curve of the overcurrent device?
Click to expand...
Put a 40 amp load in the center of the circuit and feed both ends of the ring with its own 20 amp breaker.

Roger
 
Status
Not open for further replies.
Top