6
\$\begingroup\$

I need a single-pole circuit breaker for a 200A, 48V DC battery bank powering a solar inverter. The batteries are LiFePO4 and capable of large short-circuit currents on the order of 10 to 15 kA. Large molded-case (MCCB) breakers are available with 200A rating and typically 20kA Icu (breaking) rating, but I am having difficulty finding a suitable enclosure. Can I use a DIN mounted 4-pole 50A interlocked breaker with the 4 poles in parallel? Even though this seems like a common case, I am unable to find any trustworthy answers.

Large MCCB Breaker Parallel DIN-mount Breaker

The 4P breaker has an Icu rating of 6kA, is non-polarized, and is typically used in high voltage solar PV applications where the DC interrupting voltage needs to be up to 1000 VDC, so the breakers are arranged in series. This means that the breaker has 4x the I2R loss is series. Parallel power loss through 4 breakers would be 4*1/(4^2) = 25% of the series power loss, so that seems fine from a power dissipation (e.g. thermal) standpoint. The question likely comes down to the 6 kA Icu rating and how that works in parallel with interlocked breakers.

Breaker Configurations

\$\endgroup\$

2 Answers 2

Reset to default
20
\$\begingroup\$

No, you shouldn't do this. The problem is that the breakers won't share current equally because they're not totally identical and their contact resistance might vary[1]. This means that one of them will trip much faster than the others, causing their total current-carrying capability to be much lower than expected.

Additionally, the four breakers won't open at the exact same time, which means that one of them will momentarily carry all of the short-circuit current and therefore exceed its rating.

[1] The contact resistance is somewhat random every time the breaker is operated (due to arcing and contaminants being "smeared" across the contact surface when operating) and it will also change with temperature, vibrations, oxidation or just on its own. Here's a bulletin by Schneider Electric about the topic: https://download.schneider-electric.com/files?p_enDocType=Data+Bulletin&p_File_Name=0600DB1901.pdf&p_Doc_Ref=0600DB1901

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Thank you for the reply. Since the parallel breakers are manufactured together, I would suspect (but need to verify) that the contact resistance would be close enough. The datasheet for a Langir breaker lists the resistance as 1.4 mOhm for the 50A breaker. \$\endgroup\$
    – Eric
    May 16, 2021 at 0:14
  • 1
    \$\begingroup\$ The real problem, as you mention is the Icu limit which may be exceeded for a few milliseconds while the arc is extinguished between the parallel breakers. I know there are breakers designed to be installed in parallel, I wonder if they all are still rated for the total Icu rating or if they just do better quality control to ensure they are matched. \$\endgroup\$
    – Eric
    May 16, 2021 at 0:28
  • 9
    \$\begingroup\$ @Eric That's sadly not the case, the contact resistance is somewhat random every time the breaker is operated (due to arcing and contaminants being "smeared" across the contact surface when operating) and it will also change with temperature, vibrations, oxidation or just on its own. Here's a bulletin by Schneider Electric about the topic: download.schneider-electric.com/… \$\endgroup\$
    – Jonathan S.
    May 16, 2021 at 0:30
  • 1
    \$\begingroup\$ S . - thank you for the link. A bit hand-wavy, but still useful. I guess with low-resistance connections, the what initially seems like an outrageous +/- 50% contact resistance tolerance is actually probably true. It doesn't affect the overall circuit performance when in series, but in parallel a nominal 50A through each of the 4 breakers in parallel could vary from 33A to 99A with anything over 50A causing it to eventually thermally trip. \$\endgroup\$
    – Eric
    May 16, 2021 at 1:31
  • 3
    \$\begingroup\$ @eric Breakers designed to switch high currents typically have a main set of contacts, and an higher resistance arching contact. Their internal structure switches the main contact off first, and then the arching ones. An arch will form near the arching contact, while the main contacts stay clear. 2 sets of contacts need some more space, hence the bigger package. \$\endgroup\$
    – Ferrybig
    May 17, 2021 at 10:47
8
\$\begingroup\$

Don't use cheapos or misapply equipment when the right stuff is readily available.

Schneider's Square D "QO" load centers (service panels) and breakers are, frankly, the finest residential service panel offered in North America. They are specifically rated for 48 volts DC and Square D can supply breakers rated as high as 120 volts DC.

As far as enclosures, they provide the enclosure for you - not only for the main breaker but for branch circuit breakers which tap the 200A supply.

The neat thing is, they're readily available in any home store in the US or Canada for sane costs. For instance:

The North American style "spaces" system provides a master "bus" which then delivers bussed power to each of the branch breakers. Thus there is no need for wired interconnections between the main and branch breakers, or even "comb buses". Further, it totally satisfies enclosure needs, and the breakers simply "snap in" (and you twist out a section of the metal cover to allow the breaker handle to stick out the front).

The common residential panels provide for two separate "hot" phases - so for instance this could be used in a DC system to separately breaker-protect both positive and negative (breakers on each). Or it can be wired with a "common", and all the needed busing and terminals are provided both for common and separately, safety ground.

The separate common/neutral, and ground, is not a feature of every North American panel, but it is for QO.

What about certification?

Thanks to trade treaties, many countries honor certain other countries' safety certifications, provided they are reasonably comparable.

In North America, there are two parts to it and UL has two roles. First as author of the standard (UL White Book as comparable to IEC standards, in fact they are working to harmonize them. UL 67 is the part that applies to service panels).

Second (and UL has competitors here*) as the physical inspector of the production. First, the manufacturer sends test articles to UL for physical inspection and exhaustive functional testing (including setting components on fire to see if they emit toxic smoke). And second, UL does spot inspection of factories to assure that actual production does match the approved test articles. Only this results in a UL listing.

And then the "UL Listed" logo, with the inspection file number, is placed right on the equipment label.

enter image description here

So it's rather easy to get the file; just read the file number off the equipment label and ask UL or Schneider (Square D) for it. UL has http://www.ul.com/database behind a signup-wall, but Home Depot has this.

Schneider can also tell you to what extent their panels meet IEC standards. Noting that UL 67 is a great deal more encompassing than any one IEC standard, since it covers enclosures as well. They market products all over the world.

* A list of competent testing labs is curated by US OSHA, and is relied on by agencies in many countries. Some of these correspond to trade treaties; for instance CSA, BSI and TUV are "national" testing labs of Canada, UK and Germany, and they are on the list. Likewise Norwegian Nemko, Swiss SGS, etc.

\$\endgroup\$
8
  • 2
    \$\begingroup\$ >Don't use cheapos or misapply equipment when the right stuff is readily available. Yep, fully agree on this. Unfortunately I am in New Zealand and the ABB DC breakers are $1,200 each which hurts when I need 6 of them for 3 battery banks and 2 inverters. Shipping stuff from North America typically doesn't help since it is normally UL listed and everything here must be IEC 60947-1 and -2 certified. \$\endgroup\$
    – Eric
    May 16, 2021 at 9:39
  • 1
    \$\begingroup\$ And just to avoid fears of ultra-cheap parts, I do have to provide the current IEC 60947 test reports for the electrical inspector, so they will be legitimate parts. \$\endgroup\$
    – Eric
    May 16, 2021 at 9:50
  • 1
    \$\begingroup\$ @Eric good point about the inspections, but the US stuff is perfectly well documented (UL is harsh) and UL-listed equipment from Schneider (i.e. Square D) will likely please your inspector a lot more than J random Chinese where you have to go find an IEC file. See edit. And like I say, the beauty of "QO" is it's cheap for DC rated equipment every bit the equal of ABB... you benefit from the huge economies of scale of residential service panels. \$\endgroup\$
    – Harper - Reinstate Monica
    May 16, 2021 at 20:30
  • 2
    \$\begingroup\$ Your post is going to be very useful for anyone doing this in North America or in places where certification requirements are not required. Sadly here in New Zealand UL certification is moot as inspectors are required to verify the relevant AS/NZS certification. They do not have the authority to authorize an alternative standard -- only a chartered electrical engineer could do that. \$\endgroup\$
    – Eric
    May 16, 2021 at 23:14
  • 3
    \$\begingroup\$ @Eric I am confused, because you seem confident that you can get the requisite certification letters for the two el-cheapo breakers off Alibaba that you mentioned in the question. You're saying the cheapo maker can supply a letter and Schneider can't which seems implausible. UL is eminently capable of certifying to IEC standards. One of their stamps is "UL Listed/AU", even! I'm just saying it's worth contacting Schneider to see if any of their "QO" kit is ULAU listed or IEC compliant. Heck, Schneider even makes DIN rail gear. \$\endgroup\$
    – Harper - Reinstate Monica
    May 16, 2021 at 23:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.