Voltage drop

[mbrooke]

The regs say no more then 5% voltage drop, however how do I apply (calculate) this for socket circuits?

 

[ericmark]

From the loop impedance. With a ring final with standard 32A MCB you consider it to have 20A drawn in centre and 12A even spread so the amps used to calculate are 26A not 32A. Also you have to apply correction factors so total cable permitted in a ring final works out at around 106 meters. This will show as a 0.59Ω increase in loop impedance to incoming supply.

 

[mbrooke]

This is a 20amp radial in 2.5mm2 T&E.

My apologies for not stating it earlier.

 

[ericmark]

Many years ago I made this calculator to work out volt drop. I can't remember off hand what readings are line - earth and which are line - neutral the latter is clearly what matters with volt drop. Click calculate without changing from default and you will get 106 meters of cable.

 

[mbrooke]

I did find this:

http://cablesizer.com/iec/

 

[ericmark]

Then 32 meters and around 0.79Ω increase if my old calculator is correct.

 

[mbrooke]

But do I assume a full 20amps of load lumped at one end?

 

[ericmark]

My calculator does not have power factor correction like the one linked to but does have things like Rating factor for grouping Cg. In the main you don't take to the limits. I used formula and data from BS7671:2008 it was first made around then when I was trying to see where the increase to 106 meters came from when we went from 4% to 5%. At that time I was more active with the IET and it was during one of their talks about the then new 17th Edition that I got the data.

The problem is 20A from mid point or end be it a ring or radial so as the MCB rating is reduced from 32A the increase of cable length is not linear.

 

[mbrooke]

I know, but the 5% is applied to the middle, end or evenly spaced apart as an algebraic sum?

 

[ericmark]

It depends what it supplies. Clearly if just one 13A socket at end and one mid way then use 19.5A but if two sockets at end use 20A but really if it's that critical on half an amp then you should be considering a 16A MCB.

If a radial was for example supplying an immersion heater only at 13A even with a 20A MCB you still work it out at 13A. With the earth loop impedance then it does not matter what it supplies 1.44Ω in old money was limit for 32A supply. Less now can't remember figure think to times by 0.95 to get new figures?

But for volt drop it is down to expected max load plus some common sense. Shed down your 20 acre garden (I wish) with a 40 volt volt drop feeding an LED lamp rated 150 ~ 250 volt may not comply, but it's not really a problem. Same job for a client however is a different story in 5 years time he could be demanding to upgrade the cable for free.

No loop impedance tester is that accurate so 0.79Ω increase may be limit, but even at 1Ω increase being sure the volt drop is excessive rather than meter out an bit and cable not a 30°C is near impossible. At 2Ω clearly you are sure some one got it wrong but at 81 meters rather than 32 meters it should be obvious some one has made a big error without the loop impedance reading.

It was this error problem that stopped me doing other versions there seemed little point. I also realised there was little chance of anyone finding my errors and it coming back to bite me. Take two readings and they will be different both at incomer and end enter highest incomer reading and lowest end reading then lowest incomer reading and highest end reading and look at the two cable lengths. And this is with the same meter remember. Once you factor in error then you realise may be good in theroy but in practice not really going to work.

 

[JohnW2]

With a ring final with standard 32A MCB you consider it to have 20A drawn in centre and 12A even spread so the amps used to calculate are 26A not 32A.

As has been observed before, although that is probably not unreasonable, although you very frequently quote those figures, I have never seen them mentioned, or written down, by anyone but you!

Kind Regards, John

 

[bernardgreen]

If you are that worried then use 4 mm² between CU and the cluster of heavy loads on the shorter side of the ring. Or use 4 mm² for the whole ring and thus reduce voltage drop for all soickets.

 

[JohnW2]

But do I assume a full 20amps of load lumped at one end?

If there is a double socket at the end, then you don't really have any choice but to do that, since it is then quite possible that there could be a 20A load right at the end. If it were only a single socket at the end, you could theoretically get more precise in calculating the 'worst case scenario' of VD, but I personally wouldn't bother.

The reality is that, even if one sticks to the guidance figure of 5% maximum VD, you are most unlikely to exceed that with a sockets circuit in a normal domestic property.

Kind Regards, John

 

[JohnW2]

If you are that worried then use 4 mm² between CU and the cluster of heavy loads on the shorter side of the ring. Or use 4 mm² for the whole ring and thus reduce voltage drop for all soickets.

The OP is talking about a 20A radial circuit. If there were 'heavy loads' close to the origin of such a circuit, they would have relatively little impact on VD (as observed at any socket), because of the short cable lengths.

Kind Regards, John

 

[ericmark]

I have only once had problems with volt drop. It was a three phase shrink wrap machine which needed a 20A supply so 32A plug and socket and it had been supplied using 6mm² singles in trunking around a large building. I quickly found the fault as I had already questioned the use of 6mm² and it was rewired with 16mm² as did not want to take a chance it would fail a second time. In real terms with 45 years in the trade that was the only time I found something would not work simply because of volt drop. The heaters were not getting hot enough and it was not shrink wrapping as it should as a result.

Fluorescent lamps are something which can suffer, and old pre switch mode power supply equipment which turns supply to DC inside. In a caravan site I have measured below 200 volt yet there seemed to be no real problem. In that case most of the volt drop was before the DNO head so not really the camp site wiring.

And as I said measuring with prospective fault current or loop impedance meter is so inaccurate it needs to be well out before anyone can really pick it up as a fault.

As to 26A on ring it was told to me at IET meeting I have never seen it written down either, however to get the 106 meters that was claimed as maximum cable for a ring final you need to use 26A otherwise we drop to 85 meters. This is how I came to be asking the question I could only get 85 meters and wanted to know how the 106 meter length was being quoted.