This week's question is about replacing blown fuses when only one blows on a three phase circuit. There are many schools of thought on this. Do you replace only the fuse that blew or maybe all three fuses. Do you replace the next fuse upstream in that affected phase or do you replace all fuses upstream. Of course the easiest and least expensive option is to only replace the fuse that blew.

Whether you use fuses or are not responsible for this type of activity, let us know your opinion anyway.

Here it is....

When ONE fuse of a three phase circuit blows, you should normally:

Change the ONE fuse that blew
Change all THREE fuses
Change the ONE fuse plus the next upstream fuse (same phase)
Change all THREE fuses plus the next upstream fuses

Comments and opinions are always encouraged.

Surely this mainly depends on the type of load and the reason (if found) for the failure. Not all 3 phase circuits are actually balanced so the fuses may well age at different rates hence the argument for just changing the one. That's my 2p for you.

It depends on the voltage involved, the difficulty of replacing the fuses, and the type of circuit (critical nature).

When I was working in the Utility market we replaced all three fuses on high voltage transformers unless they were fused with S&C type SMD fuses. If the substation served a very critical load we generally replaced all three regardless of type.

In the Industrial market, the cost of 600V fuses is lower and the circuits are frequently more critical, particularly in the petro-chemical industry. Therefore , replace all three fuses every time.

Another 2p worth.

I said one. However if the fuses happen to be on the primary of a delta-wye transformer and the fault was on the wye side at least one of the other two fuses was damaged. In this case I would replace all three.

I was thinking of motor circuit over current protection. Often the fuse will fail on a short to ground or possibly a phase to phase short within the motor. With many years replacing just the failed fuse in these cases I do not remember where there was a nuisance interruption caused by one of the remaining fuses. However, for utility transformer and service protection I have seen where a fuse failed and there was often additional damage. Often this damage could only be found with an IR inspection. I recommend a follow up scan whenever a utility repair as been done.

First let me say this is a great question. generating a lot of thought around our practices.

We deal mostly with motor circuits and as one of the posts above pointed out have had great success with just replacing the blown fuse. ( assumption is a ground causing the fault)

On larger transformer loads I think we need to look at a more complete evaluation/ cause.

In general a current-limiting fuse (UL 600V Class) can open for one of three reasons:

1. The fuse was undersized, this happens quite frequently on motor and transformers loads. A full voltage start motor circuit should have the fuse sized at a minimum of 150% of the NEC table value, 175% is a far better choice. Secondly, a RK5 fuse can be sized at 125% of a transformer primary, however, a RK1 or J Class fuse must be in the 150%%-175% range to ride thru the inrush current.

To expand on the #1 above. current-limiting fuses can open due to the repetitive short time inrush currents. A fuse element will see long time mechanical fatigue if 60% of the minimum melt curve is repetitively exceeded on a daily basis. To verify, you derate the minimum melt curve by 60% and juxtapose that tcc with the inrush curve.

2. Overload (125% to 500% - approx) - most modern current-limiting fuses experience a resistance change near the minimum melting curvewhich leads/causes the fuse to open, in general, this is a go-no go operation, if the fuse did not blow, it should not be replaced.

3. Short Circuit - Similar to #3, this is also a go-no go operation, when a UL Class fuse experiences current that is more than a range of 10 to 20 time rated current it is operating in less than .008 sec, again, if the fuse did not open, it should not be replaced.

Miscellaneous:

A. Always check the fuse clips for discoloration, which typically indicated the clips have lost spring tension or the fuse is overloaded.

B. As is with 99% of all UL branch circuit devices, the fuse should never be loaded more than 80%. If you use an infrared camera and you have a fuse loaded at 80% of the label rating, in a reasonable ambient you will see a temperature close to 150 deg F (UL allows a max of 167F) on the clips. One caveat, shiny copper has an emissivity of approx 25%, therefore, the IR camera may not see the true temp. In cases like that aim for a crevice or hole, the temp may read correctly or use some white-out when the part is de-energized, the IR camera will then give you an accurate reading.

C. Using a clamp-on CT has become a lot safer with the introduction of the Fluke Connects system, this allows a wireless CT to either log or send data back to a remote meter with the cabinet door closed.