selectivity on submains

[suffolkspark]

Evening all, I've got an upcoming project.... looking for info on selectivity on some submains.

currently the property in question has 2 100A single phase meters to 2 buildings 50m apart.

UKPN are going to be removing these and fitting 1x 160A three phase supply to an adjacent brick outbuilding. this building will eventually house a three phase ground source heat pump.

from this building I need a 100A single phase submain to re-feed the main house, the boards in there were all replaced myself around 5 years ago, they are MK and fully rcbo protected.

I need a second submain to building 2, this one 3phase, because a three phase car charger is being fitted, they are also maintaining a single phase car charger in the same location, so I am taking 80A three phase over there and replacing the current single phase board and spreading the existing loads onto the 2 lesser loaded phases (if both car chargers are charging 1 phase could be drawing 64A)


I was looking at memshield 3 or schneider MCCB panel boards as I need 1x 100A single phase outgoing device, and 1x 80A three phase outgoing device (+one for the ground source later) so 4 or 6 way is fine, but the board is coming to around 1.3k, would i achieve fault protection selectivity instead using a an eaton (or other) high capacity board, that take the larger MCB'S eg EMCS380? this more than halfs the cost, but in the event of a fault downstream (largest downstream device will be 32A type B on car charger) could it operate.... ? and if so ill stick to a MCCB board

 

[suffolkspark]

just to add, ill also fit an SPD in this or adjacent to the main distribution board so there will be a way for that...

 

[mattg4321]

Busbar chamber to feed switch fuses for distribution circuits and also local consumer unit for heat pump etc.

 

[GrantT]

Probably got all that stuff in my garage.
3phase Schneider dbs
dein 3 phase spd devices and all the breakers
I’ll do you a deal.

 

[pc1966]

If you are looking at only 2-3 feeds off the DNO supply then a couple of fused-switches is going to be your best and (probably) cheapest option.

The likes of the Schneider SQB1003K are well made (in the UK!) and also work well on busbar chambers as they have shrouds that also reduce the chances of anything ending up in a busbar chamber.

With a 1.6:1 ratio you should have total selectivity on the fuses (so 160A DNO to 100A main property). To get similar performance with MCB/MCCB you need to be looking at the fancy electronic MCCB with a short delay (unless you hit the high magnetic trip point) and they cost as much as the TPN fused-switches.

 

[suffolkspark]

If you are looking at only 2-3 feeds off the DNO supply then a couple of fused-switches is going to be your best and (probably) cheapest option.

The likes of the Schneider SQB1003K are well made (in the UK!) and also work well on busbar chambers as they have shrouds that also reduce the chances of anything ending up in a busbar chamber.

With a 1.6:1 ratio you should have total selectivity on the fuses (so 160A DNO to 100A main property). To get similar performance with MCB/MCCB you need to be looking at the fancy electronic MCCB with a short delay (unless you hit the high magnetic trip point) and they cost as much as the TPN fused-switches.

I'll have 3x 3phase outgoing and 1 single phase in the end, I'm wondering if in this domestic situation, A) the probability of the upstream device getting tripped is slim anyway, and B) even if it does the inconvenience isnt really so bad. the amount of jobs I've come accross where theres daisy chains of standard 60808 type B MCBS in series and they have never tripped in reality....

I just dont know if the schneider NS100N MCCBS offer better selectivity over the eaton EMCS380 type, cost isn't a massive constraint, but I dont want to overcook what is still really a domestic job.

Busbar and switch fuse option is currently coming out dearer option actually which has surprised me

 

[suffolkspark]

If you are looking at only 2-3 feeds off the DNO supply then a couple of fused-switches is going to be your best and (probably) cheapest option.

The likes of the Schneider SQB1003K are well made (in the UK!) and also work well on busbar chambers as they have shrouds that also reduce the chances of anything ending up in a busbar chamber.

With a 1.6:1 ratio you should have total selectivity on the fuses (so 160A DNO to 100A main property). To get similar performance with MCB/MCCB you need to be looking at the fancy electronic MCCB with a short delay (unless you hit the high magnetic trip point) and they cost as much as the TPN fused-switches.

Do you have a link to an example of those fancy mccbs you speak of, seems the Schneider ones I was thinking of are actually still just thermal and magnetic like a standard breaker

 

[Lister1987]

Stupid question, are fuses not an option instead of MCCB, might give better i2t performance? 160A DNO Fuse, down to your 100A then on to 63A fuse instead MCB/RCCB

 

[suffolkspark]

They are an option, thinking about it it may be my only option depending on my Zs at each load end, I better get a pen and paper out I think?

 

[suffolkspark]

The 3 options I guess really are:
A) busbar chamber into 4x switch fuses
B) MCCB board
C) MCB board (larger type)

but it looks like max zs allowances may now be the deciding factor, thanks for all input so far

 

[pc1966]

If you have long-ish sub-mains then the PSSC might be low enough not to fire a 80A D-curve MCB (or similar) "instant trip".

Schneider make a bewildering range of MCCB with all sorts of variations. This is one example, but expensive as 85kA break capability:

LV429770 | Schneider Electric, NSX Range MCCB 3P 100A, Breaking Capacity 85 kA, Fixed Mount | RS

uk.rs-online.com

The (poor) photo shows some of the adjustment post at the bottom of the body.

You might not go for a bus-bar chamber, it might be OK just to daisy chain a couple of fused-switches together with the appropriate crimped lugs back-to-back so they are all sound and tidy if only 2-3 in total. You might even use a TPN fuse-switch for more than one single-phase feed. As you say, having to isolate to replace a fuse should be very rare.

 

[suffolkspark]

If you have long-ish sub-mains then the PSSC might be low enough not to fire a 80A D-curve MCB (or similar) "instant trip".

Schneider make a bewildering range of MCCB with all sorts of variations. This is one example, but expensive as 85kA break capability:

LV429770 | Schneider Electric, NSX Range MCCB 3P 100A, Breaking Capacity 85 kA, Fixed Mount | RS

uk.rs-online.com

The (poor) photo shows some of the adjustment post at the bottom of the body.

You might not go for a bus-bar chamber, it might be OK just to daisy chain a couple of fused-switches together with the appropriate crimped lugs back-to-back so they are all sound and tidy if only 2-3 in total. You might even use a TPN fuse-switch for more than one single-phase feed. As you say, having to isolate to replace a fuse should be very rare.

100A single phase only needs to go 20m, running 25mm 3core swa

80A 3phase needs to go 70m, running 35mm 5core swa to keep zs low

Just had a look at the charts and I would invisige keeping within limits for Ctype but Dtype will be a no go, I will calc properly though

 

[mattg4321]

Just yesterday I had a commercial customer call me up having lost power to one DB following a DNO outage. Turns out the MCB feeding the board had tripped when the power came back on. Not the best design imo. Fuses are best for distribution circuits.

 

[suffolkspark]

Just yesterday I had a commercial customer call me up having lost power to one DB following a DNO outage. Turns out the MCB feeding the board had tripped when the power came back on. Not the best design imo. Fuses are best for distribution circuits.

I agree, I think I just need to weigh up cost vs benefit with it being domestic

 

[pc1966]

100A single phase only needs to go 20m, running 25mm 3core swa

80A 3phase needs to go 70m, running 35mm 5core swa to keep zs low

If the 3-phase load is basically one (or maybe two) circuits that are not so critical you could consider a combination of a basic 250A TPN board, maybe with 4-pole input switch (don't like non-isolated neutrals here) that has a 3-pole 80A MCB for the chargers

Then use the direct-link things to feed out to a 100A fused-switch to feed the single phase sub-DB and have reasonable selectivity and Zs requirements?

Just had a look at the charts and I would envisage keeping within limits for Ctype but Dtype will be a no go, I will calc properly though

See what you can get away with. In the ideal world you have an end of sub-main PFC & PSSC that is enough to disconnect in 5s or a bit less, but your final circuits are all under 0.4s and PFC/PSSC not enough hit the "instant trip" point of the upstream devices. Usually that juggling act is best if you can use a D-curve upstream so it is not really in the instant region, but firing on the thermal trip.

With MCBs and very little leeway in selection (no adjustment on measurement, etc) it is not always practical to get there. Fuses (if feasible) are a bit more forgiving.

 

[suffolkspark]

If the 3-phase load is basically one (or maybe two) circuits that are not so critical you could consider a combination of a basic 250A TPN board, maybe with 4-pole input switch (don't like non-isolated neutrals here) that has a 3-pole 80A MCB for the chargers

Then use the direct-link things to feed out to a 100A fused-switch to feed the single phase sub-DB and have reasonable selectivity and Zs requirements?


See what you can get away with. In the ideal world you have an end of sub-main PFC & PSSC that is enough to disconnect in 5s or a bit less, but your final circuits are all under 0.4s and PFC/PSSC not enough hit the "instant trip" point of the upstream devices. Usually that juggling act is best if you can use a D-curve upstream so it is not really in the instant region, but firing on the thermal trip.

With MCBs and very little leeway in selection (no adjustment on measurement, etc) it is not always practical to get there. Fuses (if feasible) are a bit more forgiving.

That's actually a good shout, direct links out to a switch fuse for the main house sub main, more food for thought...

The 80a TP is feeding another building that is just used as a games room that also has a car port on where the chargers are, so the worst thing that could happen by the upstream MCB tripping is your car wont be fully charged and you cant see to play snooker ?.

Think I'll spec all options in the coming week now I've got a few more options to explore, thanks

 

[pc1966]

Another thought was having just a 160A isolator switch and then a small Ryfield board for the SPD and 100A & 80A sub-mains. Probably not any cheaper, but likely to be very reliable with complete selectivity with the DNO fuses and very good selectivity with any down-stream MCB/RCBO.

As it is one property you don't really need separate user-switchable isolation per sub-main as you might want in a block of flats, as the main isolator switch would do that in any emergency for the owner, and a sparkey could then pull fuses to isolate any sub-main for longer term work on it.

Also you might be OK for 25mm for the 3-phase 80A run but probably will need a parallel run of 16mm CPC to keep the R2 down (unless going for expensive 5 core SWA) so your end-of-cable Zs is OK.

 

[GrantT]

Another thought was having just a 160A isolator switch and then a small Ryfield board for the SPD and 100A & 80A sub-mains. Probably not any cheaper, but likely to be very reliable with complete selectivity with the DNO fuses and very good selectivity with any down-stream MCB/RCBO.

As it is one property you don't really need separate user-switchable isolation per sub-main as you might want in a block of flats, as the main isolator switch would do that in any emergency for the owner, and a sparkey could then pull fuses to isolate any sub-main for longer term work on it.

Also you might be OK for 25mm for the 3-phase 80A run but probably will need a parallel run of 16mm CPC to keep the R2 down (unless going for expensive 5 core SWA) so your end-of-cable Zs is OK.

I fit these SPD’s in an enclosure along side the DB
if there is space under the cover of the DB I fit these.