CU or Isolator switch in garage?

[RF Lighting]

I'm not saying there's anything wrong with having a CU in the garage for a setup like that. I've got similar my self, but for one socket and one light as the OP wants, there is just no point in installing one.

If you do install a CU in a garage I'd still steer clear of 'garage CUs' and fit a bog standard on instead.

 

[bernardgreen]

Having the lighting on a 6 amp MCB and the socket on a separate MCB would (probably ) leave the lights working if the machinery tripped its MCB.

 

[ban-all-sheds]

Possibly/probably not, if it's supplied from the house CU - depends how big that breaker is and whether you'd get proper discrimination.

 

[bernardgreen]

Yes there is a risk of MCB in the house tripping. Especially if the feed to the garage is spurred from an existing ring or radial.

Depending on a risk assessment the extra cost of a slow acting 32 Amp MCB supplying only the garage via suitably sized cable may be worth spending. Then the normal fast acting 16 amp and 6 amp MCBs in the garage can trip without taking out the 32 amp in the house.

 

[JohnW2]

Having the lighting on a 6 amp MCB and the socket on a separate MCB would (probably ) leave the lights working if the machinery tripped its MCB.

Indeed, per the extensive (albeit slow) recent exchanges about this with scousespark in another thread, that was really the only potential downside of not having a garage CU (apart from fairly vague statements about 'poor design' and/or 'suboptimal' design) he could come up with.

If that were the only consideration leading to the decision to have a garage CU, and if that were going to have appreciable financial implications (particularly if, as in the 'other thread', it might have the effect of making the job notifiable), it would probably be far simpler/cheaper to have an 'emergency' battery operated light in the garage which activated in the case of power failure.

Kind Regards, John

 

[JohnW2]

Depending on a risk assessment the extra cost of a slow acting 32 Amp MCB supplying only the garage via suitably sized cable may be worth spending.

Do 'slow acting MCBs' exist? I'm familiar with 'time -delayed' ('Type S') RCDs but can't say that I've come across anything similar for MCBs (although I guess it makes sense, analogous to slow-blow fuses). The difference between Types B, C and D MCBs obviously does not necessarily have any effect ion speed of operation for fault currents high enough to operate their magnetic trip.

If one could find, and did use, a 'slow acting MCB', the cable it supplied would presumably have to be sized appropriately for the consequent longer period of potentially carrying fault current before the device operated. Whether that would be different from the size required with a standard MCB would depend upon 'how slow' the MCB was.

Kind Regards, John

 

[SaladFingers]

I think he probably means a type C or D which can handle the inrush. The time curve graphs will show that it does take longer to trip on overload (still a very short time period, but enough to allow the inrush), but indeed a fault current will be high enough to meet requirements.

 

[SaladFingers]

I think he probably means a type C or D which can handle the inrush. The time curve graphs will show that it does take longer to trip on overload (still a very short time period, but enough to allow the inrush), but indeed a fault current will be high enough to trip the breaker in time to meet requirements before the cable could get warm.

Edit: no idea why I quoted myself, perhaps it's because the things I say are never wrong

 

[JohnW2]

I think he probably means a type C or D which can handle the inrush. The time curve graphs will show that it does take longer to trip on overload (still a very short time period, but enough to allow the inrush), but indeed a fault current will be high enough to meet requirements.

As I said, if a fault current is high enough to operate the magnetic trip of a Type D MCB, I'm not at all convinced that it will be appreciably slower in so doing than would a Type B (or Type C) at the same fault current.

Kind Regards, John

 

[bernardgreen]

Yes I did mean type D, I used the description "slow acting" as I doubted that the OP would know what type D meant.

A 16 amp type B at the downstream end of a length of SWA will trip before a 40 amp type D feeding the other end of of the SWA. In the event of a short circuit fault some of the instantanous ( 1/2 cycle ) fault current through the 16 amp type B is supplied by the capacity of the cable and does not pass through the 40 amp type D. If the 16 amp takes more than a half cycle to trip then the full fault current of the next half cycle will pass through the 40 amp type D but it should survive for one half cycle.

 

[JohnW2]

Yes I did mean type D, I used the description "slow acting" as I doubted that the OP would know what type D meant.

OK - but, as I've been saying, do you think that, with a fault current high enough to operate the magnetic trip of a Type D MCB, it would operate any slower than would a Type B with the same current? I personally suspect that there would be no appreciable difference - if the current is high enough to operate the magnetic trip, it's probably going to operate just as quickly whether the actual current is 5 times or 20 times the In.

Kind Regards, John

 

[JohnW2]

A 16 amp type B at the downstream end of a length of SWA will trip before a 40 amp type D feeding the other end of of the SWA. In the event of a short circuit fault some of the instantanous ( 1/2 cycle ) fault current through the 16 amp type B is supplied by the capacity of the cable and does not pass through the 40 amp type D. If the 16 amp takes more than a half cycle to trip then the full fault current of the next half cycle will pass through the 40 amp type D but it should survive for one half cycle.

You slipped all that in after I started typing my recent reply! Maybe you're right, but I'm far from convinced. Given a typical PSCC or PFC of a few hundred amps, I would have expected the Type D to trip (or, rather, reach 'the point of no return' in terms of magnetic tripping - actual disconnection {opening of contacts} will take a finite time beyond that) extremely quickly, quite probably within the first half-cycle. Standard performance graphs etc. do not show enough degree of detail to give us an answer - do you perhaps have access to any more detailed data?

Kind Regards, John

 

[Iggifer]

Type B trips between 3 and 5 time full load current
Type C trips between 5 and 10 times full load current
Type D trips between 10 and 20 times full load current

Assuming a 40a Type D in the house protecting a 16a Type B in the garage, in most situations the 16a is going to trip first.

a typical PSCC or PFC of a few hundred amps

will not trip a Type D 40a

 

[bernardgreen]

Only experience left. The documents stayed behind when I retired.

With nuscience tripping due to inrush currents to motors starting and power being switched onto transformer the "fast" 16 amp would trip open while the slow 40 amp remained closed.

With short circuits faults both might trip but very often the upstream device remained closed.

 

[Iggifer]

With short circuits faults both might trip but very often the upstream device remained closed.

I wouldn't be complaining of two OPD's in a short circuit condition, belts and braces