Branch circuit breaker ineffective at preventing mains trip

Question

At my place of work, we occasionally deal with appliances which have faulty grounding resulting in an earth leakage trip. Some of these faulty devices have also been tripping the MCB, probably due to a short in the appliance. When the trip happens, the rest of my office trips. These are the 2 models of breakers that pop respectively:

CE463B, earth leakage: https://docs.rs-online.com/ddfc/0900766b8057a6bb.pdf

MT120A, MCB: https://hager.com/intl-en/products/information/mt120a-mcb-1p-6ka-b-20a-1m

I tried splicing a RCBO into a power extension to solve this issue, and I was hoping this would trip in its branch without tripping the mains. But now, both circuit breakers in the branch and mains trip together. Did I select the circuit breaker wrongly? If so, what do I need to look out for? RCBO in power extension: https://docs.rs-online.com/2ee5/A700000008685742.pdf

Answer 1

This is one of those "if you don't already understand what is wrong you probablly aren't competent to be doing the work" situations. With that said.

Adding an extra RCD will probablly not solve your problem. Your two RCDs have the same trip current and even if you got a more sensitive 10mA RCD there is still no gaurantee it would trip first when presented with a significant earth fault.

I see two possibilities for moving forward. The first would be to get an electrian to provide a supply that branches off prior to the existing RCD and feeds a local RCD at the bench. How difficult this will be will depend a lot on the details of the existing wiring, and possibly what country you are in.

The second option, is to use an isolating transformer. This will isolate any earth faults in your test setup from the RCD protection in the building. Isolating transformers with 230V output will normally be supplied with a floating output, this can have some advantages during troubleshooting work but if you want to simulate a normal supply you would want to add a neutral to earth link on the output of the transformer and a RCD after it (the N-E link must be on the transformer side of the RCD or the RCD will be ineffective).

Finally, it's probablly good practice to perform an insulation test on suspect equipment before plugging it in.

Answer 2

All of these devices are RCDs.

When you have a ground fault, it will tend to trip all RCDs capable of seeing the ground fault. Detection is pretty much instantaneous; tripping takes some number of milliseconds for the hammer to drop. By the time the first one disconnects, all have detected and all hammers are in motion. Thus, the idea of putting an RCD before the main RCD is not going to work.

The root cause is a bad safety culture at your company. Employees resetting breakers over and over and over instead of repairing the root problem. Jury-rigged extension cords with RCDs circuit breakers spliced inline. These are bad, and management should correct them ASAP.

If you are a repair shop who tests customer-supplied equipment, then you need an isolated power supply for that test bench. It should have transformer isolation to prevent the supply RCDs from tripping, and then a local RCD onboard -- both for technician protection and to know when you have repaired any ground faults on the appliance.

Note that European RCDs have a very weak 30 milliamp threshold, and that means that if something is tripping it, the problem is fairly serious. The 30mA threshold is chosen specifically so nuisance trips are unlikely to happen.

Answer 3

The problem is that when two breakers trip, particularly for the same reason (both overcurrent or both earth leakage (ground fault in US terminology), it is impossible to predict which will trip first.

In general with overcurrent trips, a 200A main breaker will not trip as fast as a 15A or 20A branch breaker, and if the overcurrent is < 200A then obviously the main breaker won't trip at all. But there was a recent question about a misunderstood MWBC/split receptacle where the main breaker tripped instead of (or in addition to) the branch breaker.

When it gets to GFCI (earth leakage detection/RCB in Europe), it is even worse. GFCI is designed to trip super fast to avoid a fatal injury. That is different from an overcurrent trip where, depending on magnitude, the trip may be super fast or may be much slower - even as much as seconds or minutes. Since all GFCI devices on a circuit will trip super fast, you can't predict the sequence, which is a big reason (the other is cost) not to put multiple GFCI devices on a single circuit.

Answer 4

Sounds like you occasionally get electrical things to check over, but you're not a full fledged electrical repair shop and have just dedicated a bench for these kinds of inspections.

I suggest you run a dedicated power circuit from the distribution board to each "workspace" No MWBC at all. This gives you a dedicated breaker that will go first and protect the rest of the office.

Since you describe this as a workplace, its probably wise to pass some design liability off to an electrician, rather than doing it yourself.

It MAY be appropriate to leave the GFCI out of this circuit and rely on a lower current RCD. For convenience, you might elect to install a second breaker (the lowest current value) at the bench itself so the upstream breaker in the box is less likely to go. These decisions need to comply with your local regulations, hence getting a sparky to do the install.

---

Also consider a UPS for your critical systems, like server/POS terminals/etc.