MCBs and Fuses

<...trippings...> won't happen with <...fuses...>; what will happen is that the <...fuses...> will rupture and require the attention of an Electrician.

<...improper cable laying...>

It sounds as though the whole installation requires investigation and putting right. <...MCB...>s don't just trip for no reason. Find the cause, and put it right, ideally before someone gets hurt/killed.

If in doubt, consult a qualified local Electrician.

<...very old plant with lot of entangled power and control cables which can not be rectified now...>

Then either walk away and find somewhere else to work, or have the <...plant...> demolished ideally before someone gets hurt/killed.

It's not going to get put right by posting stuff like that on CR4!

Duplicate thread, though it does look like some progress is being made.

You really need to go through a checklist to determine what the best approach is for your situation...

https://www.allumiax.com/blog/what-is-nuisance-tripping-and-how-to-prevent-it-in-electrical-power-systems

I rather doubt that voltage spikes by itself will cause MCB trips, unless you have some surge arrestors that protects your equipment the way it should. Current spikes should be equally rare.

Either way, in a plant, why would you think that you'll be better off with fuses? An overcurrent will still cause the fuse to blow. And in some cases, the SOP (Safe Operating Procedure) may still require you to de-energise the circuit in order to replace the fuse. And hope that your qualified and certified electrician (probably 3-phase certified in a cement plant) has the correct fuse/s in his toolbag. If not, a time-wasting trip to the stores is required. All this, as opposed to just resetting the circuit breaker after finding out the real reason for the trip.

• Are your MCBS as per the drawings?
• Does your MCBs have the correct trip values?
• Correct trip curves MCBs selected, depending on the load (resistive vs pump/fan/belt motors)?
• If applicable, DC-specific MCBs on DC circuits?
• El-cheapo uncertified MCBs vs certified to traceable standards?
• As mentioned elsewhere in the discussion, have you put power analysers on the affected circuits and done infrared thermographic analysis?
• Are your motor control cabinets and distribution panels clean and with no leakages, with doors that closes properly and ventilated through filters if specified?
• Are your conveyors running smoothly with no blockages?
• Did the problem start recently or is it an age-old problem?
• Are your ventilation filters clean?
• Have you systematically identified the circuits with the most "spurious" trips and started on identifying root causes? Testing affected cables, motors and transformers during downtime, for example?
• Etc. Etc. Etc. Etc.

It may be time for expert assistance. But even the expert will tell you to sort out the housekeeping first, and I'm not just talking about cables all over the show. But if your cables are failing, guess what, you have to fix that.

First thing to do is monitor and record the amp draw of the breaker.. if it is within the design specifications,monitor each output separately.Look for a conductor or conductors that are out of range.It could actually be too many motors starting at the same time causing an overload.Motors require many more amps when starting than when running.Many large motors require a soft start or two stage mechanism to prevent extreme current draw when starting.3 Phase current/voltage recorders are readily available and cheap compared to the loss of production due to downtime and the loss of quality and quantity and efficiency.You are now working blind without this type of instrumentation.

"Better" depends on the application and requirements, of course. Before addressing the problem of spurious trip investigation, the following apply for ratings up to 1000V approximately, comparing fuses to MCBs.

Fuses

In your multimeter I hope you have a fuse like SIBA's DMI type 50.199.06 10 amps, 10mm diameter 38mm long: which can break 30 kA at 1000 volts AC/DC if you connect meter on amps to busbar volts. To paraphrase a Russian joke, replacing the fuse and [maybe] meter is inconvenient BUT you do not DIE. The same 10 amps 10x38 size SIBA type 50 179 06.10 gives 200 kA breaking capacity at AC 660 V. A 35 mm DIN rail fuse holder for one 10x38 is 78 high x 18 wide x 61 deep.

A fuse does not respond to voltage spikes. If it wears out it will open circuit safely. A good fuse is made of ceramic and metal, it will not burn. High break capacity HBC Fuses are well proven that an up-stream fuse 1.6 times current rating of downstream fuse will never blow first. If you consult with a quality manufacturer, you will probably find that their fuses of same type will ensure 80 amp will blow before 100 amp. Of course a fuse blow can leave a 3 phase circuit "single phasing" but few control circuits are 3 phase and motor control type overloads with single-phase protection are easily available. Fuses under short-circuit faults will usually let through less energy than MCBs, potentially limiting the damage at the point with fault.

MCBs

A [slightly bigger than above fuse holder] DIN rail MCB with 18mm/pole width will give 16kA maximum breaking capacity Icu [only at AC 230/400V], but the "small print" in the standard tells you it was only type-tested for 3 breaks at maximum, so if you know it is breaking the maximum and you do not have test gear to prove it is OK you throw it away after 1 fault break, just like a fuse. Just to be clear, the Ics service S/C breaking current does not have number limit.

A breaker may have electronic [especially if residual current break RCBO] circuits and max. spike withstand of 1 kV 8/20 microsecond - much better if all electro-mechanical.

A worn/old MCB may not trip on fault or may trip but sit with internal arc. It is made of plastic and can burn & start a fire. Documentation I have seen for MCBs indicates you have to have 30 amp MCB upstream of 10 amp MCB to be sure 10 amp trips first.

Finding the problems and Fixing them

1. You write that you think that MCBs are tripping because they are sensitive to spikes coupled from cable to cable. Have you proved this by finding the MCB with most trips , re-cabling the circuit with a cable separated 300mm from all other cables and that stopped the trips?
2. You have not really explained the problem. a) Is it final circuit MCBs that trip or those feeding other MCBs of lower rating? b) Do all MCB distribution boards, have trips or just some? c) Just a certain make or rating of MCBs that trip?
3. Considering the greater ratio of upstream to downstream current rating required for MCBs, have fuses been replaced at some time by MCBs of same current rating without thought to short current peaks of normal operation?
4. Have you identified each MCB that tripped and how often? Much better to tackle the worst one and fix it than apply to all a "solution" that may not fix any.
5. You could try replacing the "worst MCB for trips" with fuses. This will show if the fuse vs MCB idea works [so long as you take enough time for MCB trip to have been expected].

I once investigated an electronic governor for trips caused by its output signal relays being connected to relays without diode spark suppression. For testing, an unsuppressed relay coil was connected to the governor contact as normal, but coil switched on/off by an external relay once per second. The voltage on the coil was monitored by an oscilloscope. Each break caused a burst of fast spikes [negative change very quick] ramping up in voltage from first to last: first a few hundred volts, last as much as 2.5 kV. It was noticeable that most of the spike bursts were small in amplitude/duration with no effect, but one in 50 or 100 was massive and caused a trip. It was found that 1) all the coil wires of the row of governor internal relays were loomed up by cable ties with all their contact wires which went to plant relays - bad practice for RF/spike isolation by the relay 2) The governor's "I am failed" output relay was normally energised, de-energised = trip the set. 3) the coupling via collector- base of trip relay drive transistor and the loomed-together relay wiring was enough for a negative spike to charge an 0.1 microfarad base to emitter capacitor of that drive transistor to -7 volts turning it off. 4) The drive resistor to the transistor base was such that it took ~20 ms to pull the base positive again after the spike, this was enough for the governor's trip relay to de-energise. Thus a capacitor intended to make the trip output insensitive to spurious short trip states within the governor was the cause of tripping by an external spike!

This governor was tested to Military Standards for immunity to electromagnetic interference EMI. So be warned, most equipment and control panels are not as well constructed to stop EMI as they should be. If you do not tell a builder to keep certain wiring runs or input/output separate, all wires will be mixed together!

This, to me, the 1 in 50 big spike, explained why it was so difficult to relate any particular switching event in the installation to a trip. Since most switchings occur during start or stop the easiest way to limit the possible causes of trip is to have an alarm/trip system that records the time of start and time of trip. Modern SCADA systems can do this [but if you have multiple independent systems you may have to search all SCADA records for events simultaneous with a trip - having all their clocks matched helps, else you have to compare clock times and adjust your search time.