Cable calc confirmation required!

[IAmSparkytus!]

Think I may have been oversizing cables for some three phase circuits...

For example, when ive calculated a 11.5V volt drop on a general circuit I've took this as being the 5% allowed but I've just come to realise that maybes this is actually a 2.875% volt drop at 400V

Can someone confirm please? Could save me money moving forward!

 

[davesparks]

Is the circuit supplying 230V or 400V loads?

 

[ipf]

As above. Are you calculating volt drop on a single phase or three phase circuit?

 

[telectrix]

Use 115/ phase as the max. However that will only accurately if phases are ba 4laced balanced 9

 

[IAmSparkytus!]

As above. Are you calculating volt drop on a single phase or three phase circuit?

Is the circuit supplying 230V or 400V loads?

Well I calculate both for obvious reasons but have been calculating them both in the same way and ending at the voltage and not calculating the percentage.

For example (mVAm x L x Ib) ÷ 1000 = X volts

Then as long as its been below 11.5V, all OK

But now figured that 11.5V as a percentage of 230 is different to a percentage of 400V

Hope this makes sense???

 

[davesparks]

Well I calculate both for obvious reasons but have been calculating them both in the same way and ending at the voltage and not calculating the percentage.

For example (mVAm x L x Ib) ÷ 1000 = X volts

Then as long as its been below 11.5V, all OK

But now figured that 11.5V as a percentage of 230 is different to a percentage of 400V

Hope this makes sense???

You've missed me point, which vokatge you consider depends on the load.

If this 3 phase circuit is going to feed 230V loads, such as a distribution circuit, then you need to assess the voltage drop at 230V

However if the 3 phase circuit is feeding a 3 phase 400V load, a heater for example, then you assess the voltage drop at 400V.

 

[IAmSparkytus!]

You've missed me point, which vokatge you consider depends on the load.

If this 3 phase circuit is going to feed 230V loads, such as a distribution circuit, then you need to assess the voltage drop at 230V

However if the 3 phase circuit is feeding a 3 phase 400V load, a heater for example, then you assess the voltage drop at 400V.

Yeah got ya. I was just meaning for a circuit feeding three phase equipment, not a sub mains or anything

 

[davesparks]

Yeah got ya. I was just meaning for a circuit feeding three phase equipment, not a sub mains or anything

If it is for specific equipment then you assess voltage drop based on the rated voltage of the equipment.

 

[IAmSparkytus!]

Yeah so for 400V circuits supplying 400V equipment, I've been taking the voltage drop and calculating the volt drop percentage at 230V rather than 400V meaning percentage of volt drop is actually lower and could have potentially used smaller cables saving money

 

[pc1966]

The cable figures have this included if you use them in the "obvious" way.

• If you have a 3-phase system then you use the L-L voltage and then apply the cable drop as a percentage of that. For example 10A on 50m of 1.5mm at 27mV/A/m = 27E-3 * 10 * 50 = 13.5V drop = 13.5/400 = 3.38%
• If the same cable type was single phase then 31mV/A/m so 31E-3 * 10 * 50 = 15.5V = 15.5/230 = 6.74% so basically twice the percentage drop.

Why twice? Well in the 3-phase case where we assume it is balanced there in no neutral drop (either 4C or 3C delta case) because in the balanced case there is no current in N and hence no voltage drop associated with it, i.e. you are seeing the drop on just the L conductor. This is the same reason why in the 3P case you typically measure L-N PFC and double it for the worst-case "bolted 3-phase fault" condition.

So a given cable and a given star-equivalent load current you get half the voltage drop when balanced, but if 2 of the L were open the drop would double, back to the single-phase calculation case.

TL;DR use 3-phase cable VD drop numbers with L-L voltage, and single phase VD with L-N voltage.

 

[IAmSparkytus!]

The cable figures have this included if you use them in the "obvious" way.

• If you have a 3-phase system then you use the L-L voltage and then apply the cable drop as a percentage of that. For example 10A on 50m of 1.5mm at 27mV/A/m = 27E-3 * 10 * 50 = 13.5V drop = 13.5/400 = 3.38%
• If the same cable type was single phase then 31mV/A/m so 31E-3 * 10 * 50 = 15.5V = 15.5/230 = 6.74% so basically twice the percentage drop.

Why twice? Well in the 3-phase case where we assume it is balanced there in no neutral drop (either 4C or 3C delta case) because in the balanced case there is no current in N and hence no voltage drop associated with it, i.e. you are seeing the drop on just the L conductor. This is the same reason why in the 3P case you typically measure L-N PFC and double it for the worst-case "bolted 3-phase fault" condition.

So a given cable and a given star-equivalent load current you get half the voltage drop when balanced, but if 2 of the L were open the drop would double, back to the single-phase calculation case.

TL;DR use 3-phase cable VD drop numbers with L-L voltage, and single phase VD with L-N voltage.

Thanks for the reply.

Yes, that's basically what I've came to realise today. Three phase loads have a lower voltage drop and therefore I could have potentially been using cables with a smaller conductor size.

Another thing I do aswell which I'm wondering if it's common - when Im calculating voltage drop to choose a cable size, I always use the breaker size as part of my calculation rather than the design current.

Reason for this being, if say I've installed a 16amp MCB to cater for a 13amp appliance and allowed for volt drop based on 13amps, whats to say someone won't add to the circuit in future and then the cable becomes undersized?

 

[davesparks]

Reason for this being, if say I've installed a 16amp MCB to cater for a 13amp appliance and allowed for volt drop based on 13amps, whats to say someone won't add to the circuit in future and then the cable becomes undersized?

If the circuit is installed for a specific appliance then that is what you design it for. If someone is going to replace that appliance with another then it is up to them to confirm suitability if the supply.

You can't design or install for a 'what if' situation, you can always invent another 'what if' scenario.
What if someone changes a 16A mcb for a 20A mcb?
What if someone changes a 13A appliance for a 16.5A appliance?
What if someone adds insulation over the cable?
What if the DNO make a change to tbe network that changes the Ze, PFC, vokatbe etc etc

 

[Aaron b]

The volt drop calculations are sometimes impractical. Working on industrial sites the final circuit might be connected to a sub-sub-sub-submain 400m away. The design data is normally lost or a water damaged scroll that only one operative knows of and the addition and alterations somewhere else or not recorded. I doubt theres any % left on a lot of these sites.

 

[IAmSparkytus!]

If the circuit is installed for a specific appliance then that is what you design it for. If someone is going to replace that appliance with another then it is up to them to confirm suitability if the supply.

You can't design or install for a 'what if' situation, you can always invent another 'what if' scenario.
What if someone changes a 16A mcb for a 20A mcb?
What if someone changes a 13A appliance for a 16.5A appliance?
What if someone adds insulation over the cable?
What if the DNO make a change to tbe network that changes the Ze, PFC, vokatbe etc etc

All valid points. Whats your method for say a 32A radial to be used for general sockets which you don't know what appliances will be used? This needs designed for breaker size surely?

 

[pc1966]

Another thing I do aswell which I'm wondering if it's common - when Im calculating voltage drop to choose a cable size, I always use the breaker size as part of my calculation rather than the design current.

Reason for this being, if say I've installed a 16amp MCB to cater for a 13amp appliance and allowed for volt drop based on 13amps, whats to say someone won't add to the circuit in future and then the cable becomes undersized?

Generally no, but my answer is (as usual) a lot more complicated than that.

All you need to meet is the VD on the normal maximum operating currents. For most fixed loads that can be found from the name plate or instructions, or calculated from power & nominal voltage, etc.

If you have several sockets on a single circuit then it could well be the "maximum operating current" is actually the breaker rating as it would (we hope) have been sized for typical use-case of the sockets as well as to protect the cable from overload.

But you can consider pathological cases of circuit design where VD on the nominal load is met, but all is not not quite right. Consider, say, a security camera at some estate's gate that is 0.5km away. You decided to run in fibre optic for the data and 1.5mm SWA for the power. The electronics (camera, PoE switch, etc) only take, say 50W = 0.22A. Voltage drop is 31E-3 * 0.22 * 500 = 3.41V = 1.5% so no worries. You put it on a 10A RCBO as well above the CCC for the cable and the RCD side can disconnect on any L-E faults up to 1.6kOhm.

You then get a L-N short, so the RCD side ignores it. The fault current is then 14.8A and the 10A RCBO just sits there potentially for ever. They are now running up an electricity bill of £833/month until someone figures out just WTF has gone wrong!

Is it unsafe? Well no as the cable CCC is not exceeded and the regs don't actually demand L-N fault disconnection times to be met (just L-E for ADS).

Is it good or even acceptable design? Clearly no, as under fault conditions no action was taken leading to financial pain if nothing else!

A 3A FCU as supply end switch would have saved the day...

 

[pc1966]

The volt drop calculations are sometimes impractical. Working on industrial sites the final circuit might be connected to a sub-sub-sub-submain 400m away. The design data is normally lost or a water damaged scroll that only one operative knows of and the addition and alterations somewhere else or not recorded. I doubt theres any % left on a lot of these sites.

Very true.

If you measure the end of circuit PSCC and divide by 20 then you get the current for which it meets 5% VD. E.g. Measure 350A PSCC then 5% drop overall met for 17.5A.

Why? Well the PSCC is what flows at 100% voltage drop!

You don't have to meet overall 5% drop in the regs as such, as the DNO side is supposed to size the supply and voltage so under load you are within the limits, and on high current circuits it may not be practical at all as the DNO side drop could well exceed 5%, but the above is a quirk sanity check of the system when other aspects are hidden, undocumented, or incorrectly documented.

 

[IAmSparkytus!]

Just recalculated a recently installed cable and looks like I could have installed a 4mm 4core instead of a 10mm 4core to supply a 25amp TP fixed equipment!

4mm 4core
70m x 25A x 9.5mVAm = 16.625V
(16.625 ÷ 400) x 100 = 4.15%

but because I've calculated it at 230V I've ended up at the below...

10mm 4core
70m x 25A x 4.4mVAm = 7.7V
(7.7 ÷ 230) x 230 = 3.3%

Idiot.