Is it ever acceptable to exceed voltage drop limits if Zs is still within limits of OCPD?

[EFLImpudence]

Sure the current will drop as the voltage drops for pure resistive loads due to Ohms Law, and thus the current and time required to reach the required limit to operate the OCPD will go up.

Yes.

But as I make it and likely incorrect as well, for equal percentage in drop in voltage and current in regards to resistive loads, the impedance will stay the same. 240V ÷ 1A = 240Ω / 120V ÷ 0.5A = 240Ω

Yes, but you have done that the wrong way round. You don't know the current until you know the voltage and impedance.

So if you get Zs from Ze + (R1 + R2) and it is within the limits of the OCPD for the 0.4s disconnection time, would it still not operate within the 0.4s disconnection time providing voltage drop for the design load did not exceed 25.3V

That depends on the limit, doesn't it - different for different ratings of OPD.

If you had a circuit right on the maximum Zs limit - when calculated at 218.5V (Cmin) and allowing for the actual temperature of the conductors - then any drop in voltage below 218.5V will reduce the current below that which is necessary to operate the OPD, although the temperature and the impedance will change a little.

 

[bernardgreen]

The MCB protects against faults in the cable. So the type of load is not part of the equation.

So if you get Zs from Ze + (R1 + R2) and it is within the limits of the OCPD for the 0.4s disconnection time, would it still not operate within the 0.4s

I do wonder if specifying a maximum acceptable voltage drop has anything at all to do with the calculations for sizing cables and MCBs in over current fault protection. With a dead shirt fault the voltage at the fault is zero. 100% drop and not the 10% or 3% used in cable size calculations.

 

[Nozzle]

But in Sizewell 'B' the tunnels were lit with reduced low voltage and there was no emergency lighting or a requirement to carry even a touch.

Things have changed. I'll take a guess you were there during build/commissioning? In those heady days, you could walk around in your jeans and short sleeved shirt if you so chose to (I've only seen it on the photos...) Now, to go across the "red line" you need boiler suit, steel toe caps, hard hat with LED torch, gloves, ear defenders and LEP.

Mind you, the LED thing is not compulsory but they are widespread, and the gloves & ear defenders are to be worn as appropriate (local signage etc).

3.3kV is more closer to 3.4kV so all the other fixed ratio transformers are proportionally higher too.. There's no catering for aged sodium lamps though

Nozzle

 

[eveares]

If you had a circuit right on the maximum Zs limit - when calculated at 218.5V (Cmin) and allowing for the actual temperature of the conductors - then any drop in voltage below 218.5V will reduce the current below that which is necessary to operate the OPD, although the temperature and the impedance will change a little.

But what happens when you design your load with voltage drop on 230V nominal, but you have 216V at the incoming DB from the grid. I would have thought the 5%/3% voltage drop allowance would have been designed so that the OCPD will still operate safely in the required time within the 230V +10% -6% voltage range.

 

[ban-all-sheds]

Why would low voltage (high voltage drop) increase Zs?

It won't.

I=V/R. If the problem is insufficient I, then either a lower V or a higher R will create that problem.

 

[ban-all-sheds]

But what happens when you design your load with voltage drop on 230V nominal, but you have 216V at the incoming DB from the grid. I would have thought the 5%/3% voltage drop allowance would have been designed so that the OCPD will still operate safely in the required time within the 230V +10% -6% voltage range.

http://electrical.theiet.org/wiring-matters/54/max-zs-values/index.cfm

 

[eveares]

If the problem is insufficient I

I get that, in the sense that a OCPD may not operate at a lower voltage due to insufficient current (Ohms Law) where at a higher voltage and thus higher current it would operate, but below....?

i.e. A 6A Type B MCB (BS EN 60898) with a max permitted Zs of 6.13Ω (80% of value in BS7671) operating at a very low voltage of 205V is still going to pass 33A when shorted (what is greater 5×6A, and thus should trip) on a circuit with a Zs of 6.13Ω or less. This example is dangerous because.....

http://electrical.theiet.org/wiring-matters/54/max-zs-values/index.cfm

Doh! Why did I forget about the <1 voltage factor multiplier!

 

[EFLImpudence]

eveares

Strictly speaking your question would be answered with a 'yes, that will happen'.

However, you are forgetting about all the safety margins.

Firstly, as you are talking about a lighting circuit (presumably 6A OPD and 1mm² cable) with a maximum allowed Zs of 7.66Ω (230/30)
Then multiplying by 80% for temperature correction.
The actual temperature correction is to divide Zs by 1.2 - the reciprocal of 1.2 is 0.833 (1/1.2) so there is 3.33% over correction for a start - 6.38Ω not 6.13Ω (an extra 0.25Ω to play with).

Also, a 1mm² cable carrying 6A will never get to 70°C - (even if it were carrying 16A, this is probably not the actual maximum current of 1mm² allowing for safety margin - so more safety tolerance). Therefore if the cable doesn't even get warm the maximum Zs can be 7.66Ω not 6.38Ω - another 1.28Ω to play with.

Then multiplying by 94% for a voltage of 218.5V (a reduction of 6%) - perhaps I should have done this bit first but it is the same difference.
This reduces the maximum Zs from 7.66Ω to 7.2Ω - minus 0.46Ω, far less than the previous excessive tolerances.

So, your 5% volt drop worry is likely covered by all the previous 'errors'.

Obviously there may be more fully loaded circuits than a lighting circuit but as I said earlier, a further 5% volt drop will only matter to OPDs of circuits which are right on the limit of actual maximum Zs.

 

[ericmark]

I think one has to look at a real situation and decide if with that installation is volt drop a problem. To try and come up with a one size fits all will not work. The question was "Is it ever acceptable" so the answer must be yes, there are situations where one can accept a volt drop well above the guide lines.

I had a problem with reduced low voltage, the overload was on the low voltage side of the transformer a 10A push button type thermal only over load and a 13A fuse. If something like a scaffold pole crushed the reduced low voltage cable close to the transformer it would trip the thermal overload and in some cases blow the 13A fuse.

However if as so often found on a building site there was cable plugged into cable plugged into cable then the volt drop to even zero volts would not draw enough current to operate the trip. 10 x 230 volt = 2300W at 55 volt to earth that's 2300 / 55 = 41.8 amp which is well over what a 1.5mm² attic flex will take and it would melt, what was the problem is fire, if the cores shorted near to the transformer due to heat melting the insulation then it may trip, but there was a high chance it would set fire to wooden scaffolding planks first.

However this was because the rules were flaunted with 110 volt (63 - 0 - 63 or 55 - 0 - 55 volt) and the overload was the wrong side of the transformer. With the overload on the output side there was no problem. Even with a 16A trip which is really too big for 1.5mm² flex there was no problem, in fact even with a 20 amp MCB for two 16A outlets there was no problem.

I can't see any way that having the protective device on the incoming has ever been acceptable with a split phase output? Even if you could get a 7.65 amp fuse the current to earth would still be 32A not 16A. Only way is really with a twin or triple pole MCB on the output.

With so many accidents it is not one fault, it is a series of faults which added together causes the accident. The problem with switch mode devices is as the volts drop the current goes up, which clearly causes the volts to drop more and current goes up even more. So looking at power supply for my lap top at 230 volt around 0.4A at 100 volt around 0.92A working it out from output with no losses, but the input states 2.0A over double the output and yes it gets hot, but not that hot. However with a 3A fuse in the plug (the IEC 60320 clover leaf C5 connected is rated at 2.5A) then it is unlikely any problem will be found even if the volt drop is extreme.

But how many times when PAT testing do you really check the fuse size? and this is where the problem lies, human nature is human nature and it does not matter how many times you tell some one fuse sizes should always be checked after the first 50 and no incorrect fuse found then one gives up. I was given packets of 3A fuses and expected to fill the draw with 13A fuses I had removed.

This

cable reel has a thermal trip so although rated 5A which seems strange with 3 sockets over load will likely be detected by the built in thermal trip and since in the centre of cable reel if used coiled it will warm up thermal trip so it should trip with less current. But although one should fit a 5A fuse, 5A fuses are a non preferred size, so I for one often did not have 5A fuses in my tool box. So can't use a 3A so it gets a 13A which is wrong, but understandable. So three laptops plugged in and due to volt drop they are using more than 1.6A and as a result there is an overload and cable becomes soft so simply standing on the cable causes a short.

So the fault is caused because of what? Overload, Volt drop, Wrong fuse, or incorrect location? All were the cause. And that's my point, it is rarely one fault which causes an accident.

 

[Nozzle]

Nicely written piece that,

Nozzle

 

[Spark123]

The actual temperature correction is to divide Zs by 1.2 - the reciprocal of 1.2 is 0.833 (1/1.2) so there is 3.33% over correction for a start - 6.38Ω not 6.13Ω (an extra 0.25Ω to play with).

I thought it was 1.24? From a design temp of 70°C down to measured at 10°C is a difference of 60°C. This multiplied by 0.004Ω/°C gives a result of 0.24. Add 1 (so we take into account it's own starting figure) and we end up with a dividing factor of 1.24. The reciprocal of this equates to around 80%

 

[EFLImpudence]

Yes, that's true for 10°C but aren't most values given at 20°C?

Appendix 4 uses values at 30°C.
Appendix 14 uses the 0.8 figure but does not state an ambient temperature used.

 

[Spark123]

On site guide and GN3 are at 10°C iirc.

For example a B32 MCB requires (5x32A) 160A to trip within 0.4s.

(230V/160A)= 1.4375Ω

1.4375Ω/1.24 (factor from above post)= 1.1593Ω

1.1593 x 0.95 (volt drop) = 1.1Ω which is the figure for a B32 in the OSG.

 

[EFLImpudence]

On site guide and GN3 are at 10°C iirc.

Ok. OSG would be. I don't have one.
The book I use for resistance values of conductors uses 20°.
As does TLC
https://www.tlc-direct.co.uk/Book/5.3.6.htm

For example a B32 MCB requires (5x32A) 160A to trip within 0.4s.
(230V/160A)= 1.4375Ω
1.4375Ω/1.24 (factor from above post)= 1.1593Ω
1.1593 x 0.95 (volt drop) = 1.1Ω which is the figure for a B32 in the OSG.

Yes, that's correct -
but if you are in a house at 20° you would use 1.2 - or the value for whatever temperature it was.

I have honestly never heard or seen the correction factor of 1.24 in regard to Zs correction.

 

[Spark123]

I'd just use the value given in the OSG, it errs on the side of caution even with a conductor temperature of 20°C.

 

[EFLImpudence]

I'd just use the value given in the OSG, it errs on the side of caution even with a conductor temperature of 20°C.

So, in the context of my explanation to eaveres, even more safety tolerance for his volt drop question.