Split Duplex & MWBC...vs not MWBC...What is allowed, when?

Status
Not open for further replies.

vtsoundman

Member
Location
NorCal
Split duplex receptacle application limited space, client concerned about aesthetics. Normally I trust my design engineers on things, but I have two with conflicting opinions and I feel like the code is somewhat nuanced on MWBC, vs multi-circuit in single duplexes.

1. Does code permit two 120V circuits on a split duplex w/separate neutrals w/o a breaker handle tie / 2 pole breaker. (Never/not a MWBC with shared neutral)?
Example: Two 20A 120V circuits with individual breakers feeding a split duplex (w/broken tabs fed - each recept fed by independent neutral & hot.)Answer : ?

2. How does GFCI change the situation : will code allow separate/split GFCI circuits on a split duplex w/o a handle tie?
Example : Split duplex with two 20A GFCI circuits present using separate neutrals & hot, broken tabs. No handle tie on the BC breakers. (Again Never a MWBC). Answer : ?

3. Will code allow split duplex if a 2 pole GFCI breaker is used in a MWBC?
Answer: This is an acceptable workaround and is often used. Code mandates 210.4(b) handle tie/2 pole breaker on MWBC split duplex for conventional circuits.

4. Will code allow split duplex w/ GFCI circuits if a handle tie is used with two single phase breakers & GFCI outlets?
Example: Two 20A breakers w/ handle tie. GFCI function from two upstream Receptacles feeding a split duplex. From the panel, circuit has a shared neutral and is a MWBC. After the GFCI recepts, it now was a pair of neutrals - so is it no longer technically a single neutral MWBC, but now a multiwire MWBC? I think no - 210.4(b) - It is still a MWBC, but can be convinced the other way too.
A GFCI recpt can trip and cause only half of the duplex to be 'dead' - thereby tricking the technician into thinking the duplex is dead. If the case of drowned recept, a hazard still exists...


Answer : ?


My thoughts...skip you don't care.

Per NEC, a MWBC is
A branch circuit that consist of 2 or more ungrounded conductors that have a voltage between them and a grounded conductor that has equal voltage between it and each ungrounded conductor of the circuit and that is connected to the neutral or grounded conductor of the system.
- or more simply -
A grounded conductor (neutral) is shared between a pair of hot conductors.

The Code seems clear about MWBCs for (1) requires Handle ties for preventing overloaded neutrals, (2) neutral pigtails to prevent 240V /loss of neutral.

210-4(b) states; "In dwelling units, a multiwire branch circuit supplying more than one device or equipment on the same yoke shall be provided with a means to disconnect simultaneously all ungrounded conductors at the panelboard where the branch circuit originated."


The following link is what confuses me:

http://www.ecmag.com/section/codes-...eaker-feed-split-wired-receptacles-break-tabs
QUESTION: "Your answer about split-wired receptacles on a multiwire circuit is okay, but what if the split-wired receptacles were fed by two circuits, each with its own neutral?"

ANSWER: Since this would not be a multiwire circuit, there is no requirement for the desired protection either in a dwelling unit or other than a dwelling unit.

I think the reason ECMAG is stating the above is 210.4(b) is permissive and does not specifically require breaker handle ties for independent BC on a shared yoke, but I'm hoping that some code experts & senior sparky's can help me out here.
 
210.7 Multiple Branch Circuits.
Where two or more branch circuits supply devices or equipment on the same
yoke or mounting strap, a means to simultaneously discon-
nect the ungrounded conductors supplying those devices
shall be provided at the point at which the branch circuits
originate
 
210.7 Multiple Branch Circuits.
Where two or more branch circuits supply devices or equipment on the same
yoke or mounting strap, a means to simultaneously discon-
nect the ungrounded conductors supplying those devices
shall be provided at the point at which the branch circuits
originate

Awesome. Thx...did a 2x dumb move.

It is the yoke/strap on the duplex! doh!
...and...
Must have read this section 10x thinking it said Multiwire Branch...

Clarifying statement :
Since the circuit originates at the breaker, GFCI functionality does not need to trip both feeds to a split duplex on a MWB or Multiple Branch Circuit....
 
Awesome. Thx...did a 2x dumb move.

It is the yoke/strap on the duplex! doh!
...and...
Must have read this section 10x thinking it said Multiwire Branch...

Clarifying statement :
Since the circuit originates at the breaker, GFCI functionality does not need to trip both feeds to a split duplex on a MWB or Multiple Branch Circuit....
you cannot split feed afci receptacle so I am confused by what you are asking. If the split receptacle is feed from 2 gfci's that are feed from a dp breaker then there is nothing that states the gfci receptacles must trip simultaneously but I can see why that could be an issue...In that case a dp gfci would be a better install
 
A split-wired duplex receptacle fed from two (non-MWBC) circuits would require a simultaneous disconnecting means but not simultaneous disconnect if one circuit were to trip. With a 2 pole CB the point is moot. :)
 
you cannot split feed afci receptacle so I am confused by what you are asking. If the split receptacle is feed from 2 gfci's that are feed from a dp breaker then there is nothing that states the gfci receptacles must trip simultaneously but I can see why that could be an issue...In that case a dp gfci would be a better install

I am asking about the 2nd half of your statement.
3 conductors feed a pair of GFCI Receptacles (ex: 12/3NM) . Then use pair of GFCI recpt to create two GFCI legs from the MWBC - each has their own neutral/hot from the load side of the gfci recept. I'd post an image from google, but not sure this forum allows image posting like that.

kitchenoutlets.jpg

I was pretty sure the code was silent on this issue (requiring GFCI functionality to break both hot conductors). Seems like an obvious safety oversight, but I better not say that too loud - might end up in the next revision. :eek:
 
In that case a dp gfci would be a better install

Depends...cost can be an issue for some.

Or...
I've run into a couple of situations where a the loads would trip a 2p GFCI on a MWBC (2 pcs of equipment, one on each side) ... individually, a single load would not trip a 1p GFCI breakers or recept. But the pair on the dp GFCI did.

It has only happened to me twice - both times it was with older motor driven gear. Best I could surmise was the GFCI thought the sum of the leakage was enough to cause the trip.

First instance happened after the well after Cx and operating for a while - customer's onsite tech pulled everything to check torque, wiring, etc and said he found no issues. Problem kept repeating, so he swapped the dp GFCI for a conv dp breaker - and swapped in GFCI recept at the POL. The tech claims he saw no issues and did a megger...I advised a service call to look for hazards...

2nd time - different site. Happened during Cx. Had the onsite tech to run both loads briefly on a single GFCI circuit as a quick test...it tripped the GFCI, supporting (but not proving) my hypoth about marginally high equipment leakage. Advised client to have equipment mfr check/service ... client requested GFCI at the POL since the panel was too far away. Advised against it (not wanting folks to simply keep resetting into hazard). Client did it anyway after the Cx was completed. Classic situation of owner/operator thinking the engineer is too conservative.

In the end, who knows...
 
I am asking about the 2nd half of your statement.
3 conductors feed a pair of GFCI Receptacles (ex: 12/3NM) . Then use pair of GFCI recpt to create two GFCI legs from the MWBC - each has their own neutral/hot from the load side of the gfci recept. I'd post an image from google, but not sure this forum allows image posting like that.

View attachment 16208

I was pretty sure the code was silent on this issue (requiring GFCI functionality to break both hot conductors). Seems like an obvious safety oversight, but I better not say that too loud - might end up in the next revision. :eek:

Am still a moderated member...can't edit/fix posts.
Dang...that is the wrong photo. Should have been this one!
3 conductors feed a pair of GFCI Receptacles (ex: 12/3NM) . Then use pair of GFCI recpt to create two GFCI legs from the MWBC - each has their own neutral/hot from the load side of the gfci recept.

kitchenoutlets2.jpg
 
Am still a moderated member...can't edit/fix posts.
Dang...that is the wrong photo. Should have been this one!
3 conductors feed a pair of GFCI Receptacles (ex: 12/3NM) . Then use pair of GFCI recpt to create two GFCI legs from the MWBC - each has their own neutral/hot from the load side of the gfci recept.

View attachment 16209

Editing is time sensitive.

You latest graphic will work just fine.
 
3 conductors feed a pair of GFCI Receptacles (ex: 12/3NM) . Then use pair of GFCI recpt to create two GFCI legs from the MWBC - each has their own neutral/hot from the load side of the gfci recept.

View attachment 16209
Is there a question here?

The supply appears to be a MWBC. A 2P breaker is or two handled-tied 1P breakers are required at the panel. Splitting out the MWBC to two two-wire circuits have no bearing on the requirement.
 
Is there a question here?

The supply appears to be a MWBC. A 2P breaker is or two handled-tied 1P breakers are required at the panel. Splitting out the MWBC to two two-wire circuits have no bearing on the requirement.
Dennis was confused by my question, so I posted a clarifying image about a ckt option. All my questions were answered in the 2nd post by a single ref to 210.7

via pyramid energy waves.
 
Status
Not open for further replies.
Top