will (3) 1pole MCCBs trip before a 3pole MCCB if only 1 leg shorts?

[robbm]

I was having a discussion recently and wanted to see if I could get some clarification/opinion on a suggestion I received recently.

We have an issue with a 3phase electric heater that is wired in the head and the lack of space has caused the leads to break off and touch the housing sometimes, causing that leg to short. The breakers trip eventually, but not before the the one heating element on that individual leg is entirely fried and the whole unit needs replacing (we're also working on preventing the leads from breaking off, touching the housing, and shorting). Someone suggested that with just 1 leg shorting (and not all 3) that the 3pole molded case breaker is tripping off an RMS average of the current across all 3 legs and causing the unit not to trip soon enough to prevent damage. As a resolution, he was telling me that using (3) 1pole units would cause the individual breaker for the shorted leg to trip quicker since it is only sensing that leg, similar to using (3) fuses.

Any thoughts, is he correct or just spouting thoughts out of thin air?

 

[davidbeach]

Each of the three phases in a 3-phase breaker has its own sensing. More likely to actually get all three open with a 3-phase than with three single phase. Probably a code violation to have three single phase breakers.

 

[jraef]

As bavidbeach said, a 3 pole breaker is 3 x 1 pole breakers with just a common trip mecjhanism so that if one trips, it opens all 3 legs. Each leg must have it's own sensing element. So if there is not enough current flowing to cause the breaker to trip prior to the destruction of the heater element, then the number of breaker poles in the circuit is irrelevant. Besides, if the 3 phase heater elements are connected together, you cannot have a complete circuit without current flowing in at least two of the legs anyway, or flowing to ground. And yes, here in North America it would be a code violation to use non-common trip breakers on a 3 phase load.

Most likely the problem is that the elements are in a Wye configuration, so although they work in conjunction with one another when properly connected, if one goes to ground, the voltage is only 58% of the line voltage, which keeps the current through any one relatively low. Then if the part that goes to ground is on the Wye side, the resistance of the element itself is interfering with the rise in current and the breaker is taking too long to trip. If you cannot fix the causation of the ground fault, install a better ground fault protection scheme.

"Will work for (the memory of) salami"

 

[xnuke]

Will something like this work?

Note: I am unaffiliated with the manufacturer or any distributors of this product.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[robbm]

xnuke,
that looks like something that could work, i'll just have to check on sizing, since some of our heaters range from 8A up to 85A, so there's some large ones it wouldn't work for, and most of our systems are 480VAC, not 208-240VAC, but maybe there's expanded options and sizes/voltages coming...thanks.

Thanks for all the other comments as well, regarding the code violation (I get the part about separate sensing elements for each phase) where do you find that branch circuit protection requires each pole to be connected? From NFPA79 I found 7.2.3 but it doesn't state interconnection of each phase's protection as a requirement. Not something I'm planning on doing anyways, but curious so I have that to point to in the future...

7.2.3 Power Circuits. Feeder and branch-circuit conductors
shall be protected against overcurrent in accordance with
their ampacities as specified in Section 13.5. In power circuits
for motors, devices for detection and interruption of overcurrent,
selected in accordance with 7.2.10, shall be applied to
each ungrounded phase conductor.

 

[xnuke]

If that product won't work due to its available ratings, you can always use an equipment ground fault device with a circuit breaker that has a shunt trip feature. When the equipment GFI (typical 30 mA setting) detects a ground fault, it can close a contact in the shunt trip circuit, causing the breaker to open.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[jraef]

...where do you find that branch circuit protection requires each pole to be connected?

It's indirect. First off, the NEC (National Electric Code) is NFPA 70, not 79. NFPA 79 is for machine control systems, as an adjunct to the NEC.

In the NEC then, article 240.15(B) says that a circuit breaker shall open all ungrounded conductors in a circuit. So if you have only a one pole breaker but it is feeding a load that uses two or more poles, you would be in violation of this. There are specific exceptions, but none of them would apply to your situation.

Any GFI breaker will suffice as long as you have a Neutral in your panelboard to connect to for reference, and you can get them in any size you need from just about every known supplier. But there are also plenty of separate stand-alone GF sensors that can be tied to a CB shunt trip or a contactor as well, and some of them will be able to deal with an ungrounded delta system, if that's what you have. If you have a Resistance Grounded (NGR) power system, you made need more specialized GF protection devices, referred to as "Zero Sequence Ground Fault Protection".

"Will work for (the memory of) salami"