Discuss Submain/RCD Selection in the Electrical Wiring, Theories and Regulations area at ElectriciansForums.net

Welcome to ElectriciansForums.net - The American Electrical Advice Forum
Head straight to the main forums to chat by click here:   American Electrical Advice Forum

Reaction score
0
Hi there,
I am new to this forum and I am finding the help very useful. Could someone please advice me on the following. So If I have a 60A Cutout fuse to the property how do you size the main switch, could a 100A be used. Also how would you size the RCD’s, why would one be 63A and another 80A? If I then wanted to add a sub main circuit for a small garage how would you then go onto sizing the main switch and MCB in the Main Consumer unit.
 
As a general rule the rating of any switch and RCD should be above the maximum current the circuit is expected to carry, and in most cases that can be considered as set by the upstream OCPD (i.e. fuse or MCB).

Generally you find that for ratings in the 63A-100A for domestic equipment there is little cost implication so going with a comfortable margin (say just selecting the 100A as worst-case for most setups) is not too difficult.

If you are looking at sub-main (like a garage circuit) your starting point is always: "What is the load going to be like?"

If it is a typical garage it is a junk store :) So you have the occasional 13A socket use and an amp or so lighting, maybe a few amps longer term if there is a freezer out there. But some folk have hobbies or workshop usage and that can push up the load, or massively do if it is an electrical stopping point to an external hot-tub or similar.

Once you know your load, you can think of the cable route, and from the length and environment select a cable that can carry the current with acceptable low voltage drop. You also have to think of earthing arrangements, as if the house is on a TN-C-S system and you have extraneous conductive parts to bond your earth has to be 10mm copper equivalent minimum.

Finally when you have selected your cable, determined the end point Zs for fault clearing current, you can choose the OCPD to protect it against short circuits (and maybe against overload, though the downstream breakers might set that limit). Usually a fuse for upstream OCPD is going to give better selectivity than a MCB, but your use-case and cost/effort/etc might allow that in some cases.

Finally you need to consider Part P building rules: if you are in England/Wales then a new circuit (such as this) is notafiable work so it is generally easier and safer to get a professional electrician in to do that.
 
As a general rule the rating of any switch and RCD should be above the maximum current the circuit is expected to carry, and in most cases that can be considered as set by the upstream OCPD (i.e. fuse or MCB).

Generally you find that for ratings in the 63A-100A for domestic equipment there is little cost implication so going with a comfortable margin (say just selecting the 100A as worst-case for most setups) is not too difficult.

If you are looking at sub-main (like a garage circuit) your starting point is always: "What is the load going to be like?"

If it is a typical garage it is a junk store :) So you have the occasional 13A socket use and an amp or so lighting, maybe a few amps longer term if there is a freezer out there. But some folk have hobbies or workshop usage and that can push up the load, or massively do if it is an electrical stopping point to an external hot-tub or similar.

Once you know your load, you can think of the cable route, and from the length and environment select a cable that can carry the current with acceptable low voltage drop. You also have to think of earthing arrangements, as if the house is on a TN-C-S system and you have extraneous conductive parts to bond your earth has to be 10mm copper equivalent minimum.

Finally when you have selected your cable, determined the end point Zs for fault clearing current, you can choose the OCPD to protect it against short circuits (and maybe against overload, though the downstream breakers might set that limit). Usually a fuse for upstream OCPD is going to give better selectivity than a MCB, but your use-case and cost/effort/etc might allow that in some cases.

Finally you need to consider Part P building rules: if you are in England/Wales then a new circuit (such as this) is notafiable work so it is generally easier and safer to get a professional electrician in to do that.

Ok thanks for that, clarified what I was thinking. So usually when a consumer unit is supplied with 2 different RCD ratings is this for diversity ? Also a bit off topic as I’m struggling to find information, about TT Systems. In relations to fault current a RCD is needed I believe due to the high impedance. The RCD will detect the earth leakage faults but not short circuits so how do the MCB’s function with not a high enough fault current.
 
Ok thanks for that, clarified what I was thinking. So usually when a consumer unit is supplied with 2 different RCD ratings is this for diversity ? Also a bit off topic as I’m struggling to find information, about TT Systems. In relations to fault current a RCD is needed I believe due to the high impedance. The RCD will detect the earth leakage faults but not short circuits so how do the MCB’s function with not a high enough fault current.
It depends on the board arrangement. Are you talking about a split-load board with two RCD feeding two (probably different) sets of MCBs, or a TT style of board where the incomer is a RCD, and there are one or more RCD downstream for personnel protection?

When you have RCD in series you need to separate things to achieve selectivity:
  • The upstream one has to be several time the trip rating of the downstream ones (so for small growing overloads the lower trip ones always go first)
  • The upstream ones have to have a delay-trip so the downstream ones can clear on a high current fault before troubling the supply to other circuits
For most domestic TT you have a 100mA S-type (delay) incomer and then instant 30mA RCD or RCBO for protection of personnel.

For industrial cases the incomer might be 300mA, or even higher, and you can get MCCB style of RCD with adjustable trip times and currents to allow more than one "layer" of coordinated protection (say 500mA 0.5s delay, then 100mA 0.2s, and finally 30mA "instant" or similar).
 
Last edited:
It depends on the board arrangement. Are you talking about a split-load board with two RCD feeding two (probably different) sets of MCBs, or a TT style of board where the incomer is a RCD, and there are one or more RCD downstream for personnel protection?

When you have RCD in series you need to separate things to achieve selectivity:
  • The upstream one has to be several time the trip rating of the downstream ones (so for small growing overloads the lower trip ones always go first)
  • The upstream ones have to have a delay-trip so the downstream ones can clear on a high current fault before troubling the supply
For most domestic TT you have a 100mA S-type (delay) incomer and then instant 30mA RCD or RCBO for protection of personnel.

For industrial cases the incomer might be 300mA, or even higher, and you can get MCCB style of RCD with adjustable timp times and currents to allow more than one "layer" of coordinated protection (say 500mA 0.5s delay, then 100mA 0.2s, and finally 30mA "instant" or similar).

I was thinking of a split load Board with RCD covering two sets of MCBS. Does there need to be a Time delayed RCD before the board? And how would this RCD work with live to neutural faults on the MCB as there wouldn’t be sufficient Fault current to trip, the mcb would there ?
Thank you for your help it’s much appreciated
 
I was thinking of a split load Board with RCD covering two sets of MCBS. Does there need to be a Time delayed RCD before the board? And how would this RCD work with live to neutral faults on the MCB as there wouldn’t be sufficient Fault current to trip, the mcb would there ?
Thank you for your help it’s much appreciated
The time delay is generally for RCD in series.

Typically you only see that in TT setups (due to the earth impedance being too high to clear a fault with the OCPD used) or in some industrial or agriculture cases where you have circuits needing leakage protection against fire from insulation fault (e.g. rats nibbling cables) and the total leakage would be too high, and/or needs for independence of fault clearing too great, for a single upstream RCD for personnel protection to be reliable.

In a split-load board the RCD are on two parallel branches, each feeding a separate bank of MCBs, so in the event of an earth trip you don't lose everything. Here the RCD is only for earth leakage current trip, for over-current you rely on the down-stream MCB to clear an overload.

As an aside, if your total down-stream MCB total was, say, 32A+6A+6A+6A = 50A then you could easily use, say, a 63A RCD even with a 100A up-stream fuse as overload protection for the RCD is still in place (as a bus fault between the RCD and MCB feed is very unlikely).
 
I was thinking of a split load Board with RCD covering two sets of MCBS. Does there need to be a Time delayed RCD before the board?
If it is TT there is an argument for an up-front delay RCD for overall protection as in a TT system even if all final circuits have 30mA RCD protection (split board or all RCBO). It is not a requirement, as the RCDs should be able to clear any fault to earth, but you become dependent on a single RCD for fault clearing and they are more complex and less reliable than a MCB.

An up-fornt RCD also covers the low risk of am earth fault on the busbar side of the CU which would probably not clear on a TT system and would leave all earth bonded metalwork at mains voltage so a hazard if any chance to come in cantact with the real Earth potential as well.
 
Agree with pc1966 above.
Does there need to be a Time delayed RCD before the board?
Not necessarily, is it a TT earthing system?

And how would this RCD work with live to neutural faults on the MCB as there wouldn’t be sufficient Fault current to trip, the mcb would there ?
A final circuit live neutral fault will be interrupted by the mcb. Upstream of the main switch by the incoming OCD.
The RCD is not there to detect live neutral faults.
 
Just re-posting the above with some editing so it reads properly:

If it is a TT supply there is an argument for an up-front delay RCD for overall protection, even if all final circuits have 30mA RCD protection (split RCD board, or all RCBO board).

It is not a requirement in that case, as the downstream RCDs should be able to clear any fault to earth, but you become dependent on a single RCD for fault clearing and they are more complex and so less reliable than a MCB.

An up-front RCD also covers the low risk of an earth fault on the busbar side of the CU. That would be very unlikely to clear on a TT system as it would require an astonishingly low earth rod impedance, hence leaving all earth-bonded metalwork at mains voltage which is a serious hazard if there is any chance to come in contact with the real Earth potential as well.
 
Just re-posting the above with some editing so it reads properly:

If it is a TT supply there is an argument for an up-front delay RCD for overall protection, even if all final circuits have 30mA RCD protection (split RCD board, or all RCBO board).

It is not a requirement in that case, as the downstream RCDs should be able to clear any fault to earth, but you become dependent on a single RCD for fault clearing and they are more complex and so less reliable than a MCB.

An up-front RCD also covers the low risk of an earth fault on the busbar side of the CU. That would be very unlikely to clear on a TT system as it would require an astonishingly low earth rod impedance, hence leaving all earth-bonded metalwork at mains voltage which is a serious hazard if there is any chance to come in contact with the real Earth potential as well.

Thank you for clearing this up for me, also how do the mcbs preform during a shot circuit within a TT System (Live to Neutural) with the overall inpedence of the system being high.
Thanks for your help PC1966.
 
In the TT case a L-N fault is basically identical to the TN-S / TN-C-S cases, it is only the Earth return path that is different for TT.

MCB are generally very good with a number of significant advantages over older fuses:
  • Easy to rest by unskilled (electrically speaking) folk
  • No risk of wrong fuse wire being used
  • Can also be used for isolation without powering off everything (never re-fit a rewirable fuse on to a live fault, it is not going to be a good experience)
  • Trip on small overload better defined
  • Lower let-through energy up to around the ~1kA fault level due to the speed of clearing on the "instantaneous" magnetic trip at 3-5 In (for usual B-curve MCB)
So for supply overload (as opposed to Earth faults) they are identical to TN style and pretty good. However, if you have very large fault currents or are trying to achieve selectivity with cascaded OCPD (e.g. feed to a garage CU, a common thing on these forums) then you can find a HRC fuse performs better.

While they have the disadvantage of being single-shot, and not so easy to replace, the simple HRC fuse can clear massive faults (most supply fuses are rated in the several tens, or even hundred-ish kA range) and has less let-through energy on a fault.

Also the smooth time-current response makes it easier to coordinate OCPD behaviour. With "similar" fuses (say all BS88-2 or so) you can get complete selectivity with a 1:1.5 to 1:2 ratio of protective device rating, and in many case can get adequate selectivity between an upstream fuse and downstream MCB if there is a 1:2 or more ratio in blow/trip level.

The details are more complex than covered here, but you can find some on-line tools that estimate the limit on PFC to get selectivity between devices. MCCB upstream can have fancier trip profiles to match downstream MCB, but they often cost a lot more than a HRC fuse and are often an order of magnitude poorer in terms of limiting worst-case fault energy than an HRC fuse.
 
In the TT case a L-N fault is basically identical to the TN-S / TN-C-S cases, it is only the Earth return path that is different for TT.
MCB are generally very good with a number of significant advantages over older fuses:
  • Easy to rest by unskilled (electrically speaking) folk
  • No risk of wrong fuse wire being used
  • Can also be used for isolation without powering off everything (never re-fit a rewirable fuse on to a live fault, it is not going to be a good experience)
  • Trip on small overload better defined
  • Lower let-through energy up to around the ~1kA fault level due to the speed of clearing on the "instantaneous" magnetic trip at 3-5 In (for usual B-curve MCB)
So for supply overload (as opposed to Earth faults) they are identical to TN style and pretty good. However, if you have very large fault currents or are trying to achieve selectivity with cascaded OCPD (e.g. feed to a garage CU, a common thing on these forums) then you can find a HRC fuse performs better.
While they have the disadvantage of being single-shot, and not so easy to replace, the simple HRC fuse can clear massive faults (most supply fuses are rated in the several tens, or even hundred-ish kA range) and has less let-through energy on a fault.
Also the smooth time-current response makes it easier to coordinate OCPD behaviour. With "similar" fuses (say all BS88-2 or so) you can get complete selectivity with a 1:1.5 to 1:2 ratio of protective device rating, and in many case can get adequate selectivity between an upstream fuse and downstream MCB if there is a 1:2 or more ratio in blow/trip level.
The details are more complex than covered here, but you can find some on-line tools that estimate the limit on PFC to get selectivity between devices. MCCB upstream can have fancier trip profiles to match downstream MCB, but they often cost a lot more than a HRC fuse and are often an order of magnitude poorer in terms of limiting worst-case fault energy than an HRC fuse.
I have to say, you do explain this a lot better than other information. If I could ask you one more question and take some more of your time.
So let’s say I have a TT System with no Upfront RCD. Split Load RCD Board. No Submain. There is a Line to neutral Fault on one of the ring circuits (B32 MCB) e.g line has come loose in a socket outlet and is in contact with neutral. Am I right in thinking this is a short circuit and due to this there is a large current flow that trips the MCB? The Impedance of the circuit has to be low enough to allow this current to flow ? But due to it being a TT System, the Ze is too high to allow this ?
thank
 
Am I right in thinking this is a short circuit and due to this there is a large current flow that trips the MCB? The Impedance of the circuit has to be low enough to allow this current to flow ?
......Yes, and it usually is with a Live-Neutral 'short circuit'.

But due to it being a TT System, the Ze is too high to allow this ?
...... No, the Ze is the earth fault circuit for which pc1966 has explained, you need a RCD(s).
 
So let’s say I have a TT System with no Upfront RCD. Split Load RCD Board. No Submain. There is a Line to neutral Fault on one of the ring circuits (B32 MCB) e.g line has come loose in a socket outlet and is in contact with neutral. Am I right in thinking this is a short circuit and due to this there is a large current flow that trips the MCB?
As DefyG said Ze does not matter here as the fault return is via the neutral, not earth path.

What exactly happens depends on the magnitude of the fault. Using their on-line tool, a Schneider iC60H 32A B MCB and as 63A Gg curve BS88 fuse (to approximate the typical 60A domestic cut-out fuse as they don't list all fuse families) it gives a selectivity limit of 1.9kA

So if your fault impedance at that socket leads to a fault current (typically called the PSSC here instead of PFC) below 1.9kA then the MCB clears the fault before the upstream fuse as got past the pre-arcing I2t limit, and it lives to fuse another day.

If you have the slightly unusual case of a PSSC at the socket outlet above 1.9kA (so house close to substation, socket fairly close to the CU) then by time the MCB has cleared the fault the upstream fuse has already melted and has entered the arcing phase. The fuse is past the point of no return and has blown.

If, hypothetically, your PSSC was massive, say someone for God knows what reason put one in the substation from a similar fuse and a 13A socket right next to the board, you might expect this to destroy the MCB. If you look at the curves for such a fuse you should see one curve showing the peak let-through current:

Submain/RCD Selection fuse-let-through-curves - EletriciansForums.net

The vertical scale is peak let-through current, the horizontal is PSSC as symmetric RMS value (i.e. as you would measure with MFT, or compute with Umax/Zs). The diagonal line is the no-action curve, i.e. what the peak is on a fault for as given PSSC.

The curves for each fuse coming off that shows the point they start to break in less than 1/2 cycle and limit the peak fault current. So even for a frightening 120kA fault situation, the 63A fuse limits the peak let-through to that of about a 5-6kA symmetric fault, so the MCB might be saved to trip another day.
 
Last edited:

Reply to Submain/RCD Selection in the Electrical Wiring, Theories and Regulations area at ElectriciansForums.net

OFFICIAL SPONSORS

Electrical Goods - Electrical Tools - Brand Names Electrician Courses Green Electrical Goods PCB Way Electrical Goods - Electrical Tools - Brand Names Pushfit Wire Connectors Electric Underfloor Heating Electrician Courses
These Official Forum Sponsors May Provide Discounts to Regular Forum Members - If you would like to sponsor us then CLICK HERE and post a thread with who you are, and we'll send you some stats etc
This website was designed, optimised and is hosted by Untold Media. Operating under the name Untold Media since 2001.
Back
Top
AdBlock Detected

We get it, advertisements are annoying!

Sure, ad-blocking software does a great job at blocking ads, but it also blocks useful features of our website. For the best site experience please disable your AdBlocker.

I've Disabled AdBlock