Hi All,
People next door want to install a power supply to their large wooden concrete floored shed and a hot tub in the garden too. They want me to give them a hand..
House is on TNS so i will assume a Ze of 0.8 ohms and the cable run is 25 metres.. clipped direct or run in ground..
The hot tub is apparently 32A, so let's round it up to 40A as they might want lights or a small heater in the shed at the same time.
So, for the load, let's say 40A with a cable length of 25 metres.
We will be using a 40A "B" curve MCB
Let's do volt drop first.. there will be lighting so this is limited to 3%
This gives us a maximum volt drop of 230/100 x 3 = 6.9V. Call it 7 volts..
So, it is 7000/40A x 25 metres which = 7
So we want a cable with a value for mV/A/m of LESS than 7..
From the wiring regs, a 10mm 2 core 70c armoured cable has a mV/A/m value of 4.4 at a conductor operating temperature of 70 Centigrade so a 10mm cable will be MORE than good enough from the point of view of volt drop..
Now, to get a 40A "B" curve breaker to cut off in the required time, requires 40A x 5 = 200A
BUT, it is assumed that the voltage will drop to 218.5 volts AND we then also have to make an allowance for cable heating too..
This can either be done by using a max Zs of 0.8 of the theoretical figure or by increasing the required tripping current by 1.25 [1.25 being the reciprocal of 0.8] and this is what I usually do..
So, we have now arrived at 218.5V/250A = 0.874 ohms, and this is our max Zs for our circuit..
Therefore, to achieve our max Zs of 0.874 ohms, the max "R1 + R2" value for our circuit must not exceed 0.074 ohms..
1000 metres of 10mm 2 core armoured cable, using the armour as CPC has an R1 + R2 of 7.83 ohms.
We only want 25 metres and so 7.83/1000 = 0.00783 x 25 [metres] = 0.196 ohms..
This figure is too high, therefore we have two choices..
We can EITHER, use a 50mm 2 core using the armour as CPC and we would then achieve a Zs of 0.867 ohms, [which would be perfect][but deranged!]
or
we can try using a three core cable using an internal core as CPC..
Let us crunch the figures for a 3 core 10mm..
1000 metres of 10mm 3 core cable, using an internal core as CPC has an R1 + R2 of 3.66 ohms.
We only want 25 metres and so 3.66/1000 = 0.00366 x 25 [metres] to give an R1 + R2 of 0.0915 ohms..
Our Ze of 0.8 ohms plus our R1 + R2 equals a Zs of 0.8915 ohms. This is still slightly too high, [we want 0.874 ohms really] BUT we will now have the steel armour in parallel to the copper CPC..
This means that the R1 + R2 for the cable will now be [for a 1000 metre run] 1.83 ohms for the copper and 4.00 ohms for the armour, which, [as they are resistors in parallel] gives us an R1 + R2 for 1000 metres of 1.26 ohms..
We only want 25 metres and so 1.26/1000 = 0.00366 x 25 [metres] to give an R1 + R2 of 0.0315 ohms..
We now have our Ze of 0.8 ohms PLUS our R1 + R2 for the combined copper earth core [AND the armour in parallel] of 0.0315 ohms, to give a total Zs of 0.8315 ohms.
This is lower than our maximum of 0.874 ohms, and so will do nicely..
But, you must not "make up" a CPC from two conductors, say the armour and another separate conductor, as you have no way of knowing which way the fault current will choose to flow, therefore, EACH conductor [be it armour or otherwise] must be large enough on its own.. So that idea is no good..
So, what about a three core 16mm cable.
From the tables,
1000 metres of 16mm 3 core armoured cable using an internal core as CPC has an R1 + R2 of 2.30 ohms. We only want 25 metres and so 2.30/1000 = 0.0023 x 25 [metres] to give an R1 + R2 of 0.057ohms..
The R1 + R2 of 0.057ohms PLUS the Ze combine to give a Zs of 0.857 ohms, well within the maximum allowable figure of 0.874 ohms..
Happy days.. BUT, the house is on TNS, it is VERY likely, [the cut out is very old] that before long the DNO will alter the thing to TNCS. This will very good, as then, with an assumed Ze of 0.35ohms, even with a 2 core 10mm we would have a Zs of 0.546 ohms, which would be brilliant..
BUT, and it is a very very big but.. We would now have a shed out in the garden and a hot tub on TNCS.. Err, no thanks.. Never mind lost neutrals, what about "perceived shock"
I am leaning towards a 2 core 10mm SWA [which is more than good enough from the volt drop point of view] The cable can be installed as TNS and protected at its origin with the 40A mcb AND a 30mA RCD to provide ADS for the CABLE within the required time..
At the other end, will be a CU but with the SWA cable terminated in an insulated gland and the shed and hot tub TT'ed with a second 30mA RCD to provide ADS for the shed.
Idea is that there will be TWO 30mA RCD's in series, one at the origin and one in the shed CU to provide "redundant protection" [stuff discrimination, which is why i would not use a delayed RCD at the house end] as an RCD is not a very reliable device..
The idea of all this is that not only will things be fine as they are, things will still be fine IF the house supply is altered to TNCS..
What you all think????
Thanks all
john..
People next door want to install a power supply to their large wooden concrete floored shed and a hot tub in the garden too. They want me to give them a hand..
House is on TNS so i will assume a Ze of 0.8 ohms and the cable run is 25 metres.. clipped direct or run in ground..
The hot tub is apparently 32A, so let's round it up to 40A as they might want lights or a small heater in the shed at the same time.
So, for the load, let's say 40A with a cable length of 25 metres.
We will be using a 40A "B" curve MCB
Let's do volt drop first.. there will be lighting so this is limited to 3%
This gives us a maximum volt drop of 230/100 x 3 = 6.9V. Call it 7 volts..
So, it is 7000/40A x 25 metres which = 7
So we want a cable with a value for mV/A/m of LESS than 7..
From the wiring regs, a 10mm 2 core 70c armoured cable has a mV/A/m value of 4.4 at a conductor operating temperature of 70 Centigrade so a 10mm cable will be MORE than good enough from the point of view of volt drop..
Now, to get a 40A "B" curve breaker to cut off in the required time, requires 40A x 5 = 200A
BUT, it is assumed that the voltage will drop to 218.5 volts AND we then also have to make an allowance for cable heating too..
This can either be done by using a max Zs of 0.8 of the theoretical figure or by increasing the required tripping current by 1.25 [1.25 being the reciprocal of 0.8] and this is what I usually do..
So, we have now arrived at 218.5V/250A = 0.874 ohms, and this is our max Zs for our circuit..
Therefore, to achieve our max Zs of 0.874 ohms, the max "R1 + R2" value for our circuit must not exceed 0.074 ohms..
1000 metres of 10mm 2 core armoured cable, using the armour as CPC has an R1 + R2 of 7.83 ohms.
We only want 25 metres and so 7.83/1000 = 0.00783 x 25 [metres] = 0.196 ohms..
This figure is too high, therefore we have two choices..
We can EITHER, use a 50mm 2 core using the armour as CPC and we would then achieve a Zs of 0.867 ohms, [which would be perfect][but deranged!]
or
we can try using a three core cable using an internal core as CPC..
Let us crunch the figures for a 3 core 10mm..
1000 metres of 10mm 3 core cable, using an internal core as CPC has an R1 + R2 of 3.66 ohms.
We only want 25 metres and so 3.66/1000 = 0.00366 x 25 [metres] to give an R1 + R2 of 0.0915 ohms..
Our Ze of 0.8 ohms plus our R1 + R2 equals a Zs of 0.8915 ohms. This is still slightly too high, [we want 0.874 ohms really] BUT we will now have the steel armour in parallel to the copper CPC..
This means that the R1 + R2 for the cable will now be [for a 1000 metre run] 1.83 ohms for the copper and 4.00 ohms for the armour, which, [as they are resistors in parallel] gives us an R1 + R2 for 1000 metres of 1.26 ohms..
We only want 25 metres and so 1.26/1000 = 0.00366 x 25 [metres] to give an R1 + R2 of 0.0315 ohms..
We now have our Ze of 0.8 ohms PLUS our R1 + R2 for the combined copper earth core [AND the armour in parallel] of 0.0315 ohms, to give a total Zs of 0.8315 ohms.
This is lower than our maximum of 0.874 ohms, and so will do nicely..
But, you must not "make up" a CPC from two conductors, say the armour and another separate conductor, as you have no way of knowing which way the fault current will choose to flow, therefore, EACH conductor [be it armour or otherwise] must be large enough on its own.. So that idea is no good..
So, what about a three core 16mm cable.
From the tables,
1000 metres of 16mm 3 core armoured cable using an internal core as CPC has an R1 + R2 of 2.30 ohms. We only want 25 metres and so 2.30/1000 = 0.0023 x 25 [metres] to give an R1 + R2 of 0.057ohms..
The R1 + R2 of 0.057ohms PLUS the Ze combine to give a Zs of 0.857 ohms, well within the maximum allowable figure of 0.874 ohms..
Happy days.. BUT, the house is on TNS, it is VERY likely, [the cut out is very old] that before long the DNO will alter the thing to TNCS. This will very good, as then, with an assumed Ze of 0.35ohms, even with a 2 core 10mm we would have a Zs of 0.546 ohms, which would be brilliant..
BUT, and it is a very very big but.. We would now have a shed out in the garden and a hot tub on TNCS.. Err, no thanks.. Never mind lost neutrals, what about "perceived shock"
I am leaning towards a 2 core 10mm SWA [which is more than good enough from the volt drop point of view] The cable can be installed as TNS and protected at its origin with the 40A mcb AND a 30mA RCD to provide ADS for the CABLE within the required time..
At the other end, will be a CU but with the SWA cable terminated in an insulated gland and the shed and hot tub TT'ed with a second 30mA RCD to provide ADS for the shed.
Idea is that there will be TWO 30mA RCD's in series, one at the origin and one in the shed CU to provide "redundant protection" [stuff discrimination, which is why i would not use a delayed RCD at the house end] as an RCD is not a very reliable device..
The idea of all this is that not only will things be fine as they are, things will still be fine IF the house supply is altered to TNCS..
What you all think????
Thanks all
john..