Maximum voltage drop on SWA runs to outbuildings, am I over engineering the solution

Whilst it could be argued that modern LED lights are more tolerant of reduced supply voltages than are incandescent lamps, that does not tell the whole story.

These days we should be considering the energy cost of voltage volt drop, especially for large and long hour loads such as EV chargers. Consider a typical 7 kw charger used for 1000 hours a year. A voltage drop of 5% represents the loss of 350 watts. 350 watts for 1000 hours a year is 350 kwh. At current prices of about 30 pence a unit, that is over £1,000 a year in wasted energy.

Doubling the cable size to halve the loss to about £500 a year is likely to be worthwhile. Cable of twice the size costs about twice as much, but the trench or other installation overheads are about the same.

I would aim for a voltage drop of no more than 2.5% in a submain to an outbuilding that includes an EV charger. That allows 0.5% for the presumably short conductors between the service cut out and the origin of the sub main, and between the load end of the submain and the EV charger. If these connections are not short, then calculate properly for all parts of the install.

So whilst the light fitting should be fine at say 200 volts, the 3% LIMIT IS SENSIBLE for energy saving rather than for lighting effectiveness.

Broadgage has said exactly what I was going to say, for things like an EV charger, which is using a lot of current for a long time, cable losses could be a significant cost, so going up one or 2 cable sizes could actually save money over the life of the charger. The only downside is trying to fit, say, a 50mm cable into typical domestic style consumer units and switch fuses. You need to size up the enclosures to get those cables in.

Or run it in T & E or singles to an adjacent metal box and gland off the SWA there, leaving the consumer unit untouched. Provided you have space to mount such a box. 

Yes the volt drop is something you have to watch and the way fuel prices have gone recently then more so than ever. If we had a crystal ball and could reasonably predict the  expected prices of KWhrs in future then we might rethink our allowable losses too

Has a factor of 10 slipped in there Broadgage? Even so £100 is worth saving if possible. Unless it's massive runs I try to go at least one size up to future proof against unexpected demand in the future.

I agree I think Broadage has slipped up with decimal places but I agree cost of the losses should be a consideration.

I am talking about situations where using armoured cable is the only practical solution as outbuildings are a long way from the house or getting the cable through the house is a significant challenge. In most cases voltage drop with just the charger running is within 3% but then there are other loads on a socket circuit, tumble dryers, power tools and worst case hot tubs. In most cases these are running for less time than the charger and cost of losses are less of an issue.

I agree with short runs on dedicated charger circuits moving from say 4mm to 6mm cable makes good sense but if we are looking at a 50m cable run and deciding between 10mm or 16mm cable with over £100 cost difference the decision is less clear.

With the attached calculation if someone drives 12000 miles per year they will spend 417 hrs charging, 5% cable loss equates to about £46 worth of electricity per year. If the a larger cable was used to bring loss down to 3% they would save £18.

From talking to customers at the moment most are probably closer to 6000 miles per year, although this may change as people get back in to more normal travel patterns post Covid. Also I suspect that high millage drivers are still viewing EV's as impractical, which I assume will change as range increases.

XLSX

So in conclusion if my calculations are correct 

Moving from 4 to 6mm for shorter cable runs probably makes sense (especially if looking at none SWA cables)

Beyond that from a cost of voltage drop loss point of view selecting a larger cable doesn't make sense in my opinion, although having a 10mm earth cable run out to outbuildings has some advantages if there is a desire to export a TNCS earth now or in the future.

I also agree that it's important t stay within the 5% and to make allowance for future loads, especially as open pen detectors can be upset by low supply voltages.

Yes, 350 kWh @ £0.30 costs £105.

I am not sure that the EVCP would be running at 7kW for 1000 hours per year. At 3 - 4 mi per kWh, that is 21000 - 28000 miles per year and assumes no charging at work, en route, or elsewhere. So 250 hours per year may be a better estimate and now the 50% saving of doubling the CSA is only £13.

Getting 25 mm² cable into domestic stuff is bad enough, but you'd only be going up from 4 mm² to 10 mm² or perhaps even 6 mm² to 16 mm² if there are other significant loads.

alanblaby said:

The only downside is trying to fit, say, a 50mm cable into typical domestic style consumer units and switch fuses.

Have you been sold an undersized cable to supply your electric vehicle charger?

Contact us at www.wherethere’sblamethere’saclaim.com to claim your refund for excessive electric bills from your electrician on a no win no fee basis.

Note- there is 99% fee on successful claims.

Then coming back to my original question 

What drove the original 3% voltage drop limit for lighting circuits and does it still apply to modern led lighting. 

How much voltage drop can an LED accept before significant changes in brightness occurs

Do LED fittings include power supplies that provide a fixed output for a range of input voltage and how big is that range in practice.

Flicker on LEDS is a very variable thing - unlike hot filaments that have a natural time constant of 10-50msec depending on the filament weight, or gas tubes that have some afterglow and do not completely die away between mains half cycles, the LEDs themselves are very fast on-off indeed, tens to hundreds of nanoseconds in fact and can be used for high speed signalling.. (e.g. LiFi  as used in places that do not want to radiate their data outside the room)

So it falls to the power supply and fitting designers to determine how flicker sensitive or not.  There are those that contain complex switching supplies and can track up and down mains variations and maintain perfect output.

Some designs are less complex, but clearly well smoothed and can ride over voltage fluctuations that would be objectionable on filament lamps bit slowly dim in response to a failing supply.  Others are much simpler, being little more than a series capacitor, a bridge rectifier and some LEDS as a load. This sort are cheap and popular, but very vulnerable to supply flicker - in some case more so than the old filament lamps.

So there is no simple answer - apart perhaps from using 12V LED strip and a power supply you smooth yourself,

Mike.