Hi all. As I continue to slog through my list of 104 buildings on campus I need to do studies on I discovered yet another over-dutied situation yesterday. Common scenario, 208Y/120V "service" with 15-ish kA available at the main panel and a 100A MLO branch panel right next to it (very short feeder) with 14-ish kA available. Square D QOB (10KAIC) branch breakers. No main breaker. The main panel is Cutler Hammer so no series ratings exist between CH and SQ D.

I've come across this several times over the course of the first 50 buildings' studies and have been changing the breakers out to 22KAIC rated, QOB-VH (or whatever, brand specific) breakers to solve it if no series ratings exist already (existing or easily replaceable feeder breaker or fuse in combo with the branch panel breakers).

I was looking at the Square D series chart and it appears to me there is a series rating of 22KAIC available with a QOB-VH main and QOB branches.

Hmmm. Just move the feeder conductors from the main lugs over to a new 3P80 or 3P90 or 3P100 branch mounted QOB-VH breaker. Simple and easy enough if the feeder reaches. If not, it is only about 4' long to start with so it is easy to replace.

Not sure why it never occurred to me before now to look for a branch mounted main solution for situation like this, Square D or Siemens or GE or Cutler Hammer or otherwise.

Probably commonly done by others on here but if not, now you have another weapon in your arsenal to consider in certain situations. If you have, say, an existing 42 ckt panel with 30+ circuits in-use to deal with adding a branch mounted main like this should be a lot easier and cheaper with less down time than replacing all the breakers.

Good info. Thanks.

Series ratings have been evolving in practice and with NEC requirements. A long long long time ago, it was common to use current limiting fuses and their let-thru charts to predict the let thru current for downstream breakers. Early in my career (rewind to 1981 with a major equipment manufacturer) problems began so the manufacturer's UL and others got together and began testing series ratings. This was to account for some of the unknowns that were occurring by just using graphs. Some of the testing lead to an "ah ha" moment when it was found certain non-current limiting combinations could end up working as a series rating. i.e. the example that you cite in this thread.

Manufacturers soon began to publish their tested ratings and the NEC soon had language that accepted only tested combinations. Fortunately the NEC later added language that permits engineered ratings for existing systems where test data may not exist however, there are many requirements that need to be met when using this method.

Thanks bbaumer for shedding more light on all this.