Relying on loads not being able to overload

[JohnW2]

I try to avoid the OSG as it makes generalisations. I think it bases its assertions on standard circuits and that OPD are providing overload and fault current protection. As such, the circuits in the OSG should be inherently safe. With regards your postulate as to whether the subject is known abroad - I think not. I am no expert and am learning continually.

I agree with all you say about the OSG, but the impression I get is that a lot of people are (too?) heavily reliant on it. FWIW, IM(H)O it would be far better if a lot more people were familiar with (the contents of):

Untitled

• JohnW2
• 3 Apr 2014
• 1

Seems a lot more pleasant around here these days. Still contend that you would be better off posing questions of this type on the IET forum.

It is certainly different, and in some senses more pleasant, but there is also a potentially valuable resource missing; we don't know whether this is a temporary or permanent state of affairs.

What you say about the IET forum is, in some senses, very true, but it is not a place where I do (or probably should) participate; I just spectate a bit

Kind Regards, John

 

[THRIPSTER]

I would recommend Electrical Installation Calculations by Coates and Jenkins and Commentary on IEE Wiring Regulations by Paul Cook as well as the title you endorse.

To obviate the problems associated with unskilled design, I think that the powers that be designed the OSG as a 'coverall'. That's why I think you'd be better off raising questions of this ilk on the IET forum as you would get answers from people who actively contribute to formulation of the Regulations.

It is only a valuable resource if the contents are delivered with a sense of humility and without intimidation of others. Investigation would show that that had not been the case, here and in the other place.

Regards

 

[JohnW2]

I would recommend Electrical Installation Calculations by Coates and Jenkins and Commentary on IEE Wiring Regulations by Paul Cook as well as the title you endorse.

I agree. As I'm sure you understand, my main point was that those whose understanding is based on these books would be able to determine whether an 'undersized' cable had adequate fault protection, whereas those relying on the OSG probably could not.

To obviate the problems associated with unskilled design, I think that the powers that be designed the OSG as a 'coverall'. That's why I think you'd be better off raising questions of this ilk on the IET forum as you would get answers from people who actively contribute to formulation of the Regulations.

I'm sure what you say about the OSG is correct - although the various inconsistencies with BS7671 are perhaps unfortunate/confusing. I take your point about the IET forum but, as I've said, I don't really have any issue with the regulations (unless one feels that they should be made 'idiot proof'-which I think would be unnecessarily restrictive for competent electricians). For the regs to allow that (with conditions) overload protection may be omitted if the risk of overload is very small is, IMO, perfectly reasonable. What I was interested in was the interpretation that electricians put on that - i.e. what situations they feel present only a very low risk of overload.

On reflection, I think one reason I wanted to raise the issue here is that one occasionally sees the possibility of omitting overload protection being mentioned here, in the context of a question posed by a DIYer - something that I think probably shouldn't happen.

It is only a valuable resource if the contents are delivered with a sense of humility and without intimidation of others. Investigation would show that that had not been the case, here and in the other place.

That's all very true (I did say "potentially valuable" ). However, surprising (to say the least) though it is, if you look in the Building forum (and some related forums/fora), you may come to the conclusion that the most unexpected of leopards can sometimes change their spots if they want to (or are forced to)!

Kind Regards, John

 

[JohnW2]

Yes, you are right John - you wouldn't 'bat an eyelid'. But you wouldn't contemplate hanging 1mm off a 50A MCB, as you know, because it would fall foul of other design criteria......so not entirely sure what you are saying Old Bean.

I thought it was probably time to do some sums, and, if I've done them right (please tell me if you think I haven't - which is quite possible!), assuming a disconnection time of 0.1 secs and a 'k' value of 115, I get:

Untitled

• JohnW2
• 5 Apr 2014
• 1

As you can see (assuming my figures are right), and as one would expect, the situation would be pretty tight if one were hanging a cable with a 1mm² CPC (i.e. either 1mm² or 1.5mm² T+E) off a B50 circuit. However, it would seemingly not be impossible. If the Zs were between 0.63Ω and 0.74Ω, and if the PSSC were ≤364A for 1mm² T+E, or ≤545A for 1.5mm² T+E, then it would appear that both the adequacy of conductor CSA (adiabatic calculation) and disconnection times under fault conditions would meet the requirements.

As can also be seen, needless to say, as the extent of the 'undersize' of the cable gets less extreme (in relation to the In of the OPD), the scope for it being acceptable under (L-E or L-N) fault conditions becomes greater.

Have I got this right?

Kind Regards, John

 

[THRIPSTER]

Hi John,

This looks very interesting. Will have a look in a week or so, as am away with the boys.



Regards

 

[THRIPSTER]

Though at first glance, do we need to use 230V as the nominal voltage?

Regards

 

[JohnW2]

Hi John, This looks very interesting. Will have a look in a week or so, as am away with the boys.

Fair enough. Have a good time. I look forward to hearing your thoughts in due course.

Kind Regards, John

 

[JohnW2]

Though at first glance, do we need to use 230V as the nominal voltage?

Voltage obvioulsy only comes into what I did in relation to calculation of Zs (and L-N loop resistance) from PEFC/PSCC - and, yes, I used 230V as the nominal voltage, which I thought was the conventional thing to do. Is that not the case? However, if you wanted to use some other voltage (e.g. the actual supply voltage that happened to be present at the time of inspection), that would not have a major impact on the figures I presented, and my conceptual bottom-line conclusions would remain unchanged.

To my mind, the most iffy bit about adiabatic calculations when using MCBs is deciding what disconnection time to use, give that the operational curve is essentially vertical, over a wide range of times, in the region of interest. It seems fairly convention to use 0.1 sec in that situation, which is what I did. Given that assumption, everything else is obviously simple arithmetic.

If one uses disconnection times greater than 0.1 sec for the calculations, the adequacy of fault protection obviously diminishes. If I've got my sums right, in terms of the example we were discussing, using 1mm² or 1.5mm² T+E on a B50 would cease to be possible (in terms of CPC fault protection) if disconnection time rose to above about 0.137 secs.

Kind Regards, John

 

[EFLImpudence]

I would say that your calculations are what would normally be done to ensure the required information for what would be a 'worst case' situation.

The disconnection times could be (a lot) less than 0.1 seconds and the maximum Zs may not be the 80% value depending on the load.

If desperate then the let-through information from the manufacturer could be obtained.

230V should be used because that is what is used to work out the maximum Zs values.
If the actual voltage were used with the corresponding values, would not the result be the same?

 

[JohnW2]

I would say that your calculations are what would normally be done to ensure the required information for what would be a 'worst case' situation. ... The disconnection times could be (a lot) less than 0.1 seconds and the maximum Zs may not be the 80% value depending on the load.

Thanks. Indeed - which is why I believe I have undertaken the calculations in the conventional 'worst-case' fashion.

If desperate then the let-through information from the manufacturer could be obtained.

True, but I doubt many people are going to do that. I'm a little surprised that more useful information is not 'published', so one doesn't have to go asking for it.

230V should be used because that is what is used to work out the maximum Zs values. If the actual voltage were used with the corresponding values, would not the result be the same? Am I missing something?

As I said, I believe use of the nominal voltage is conventional for such calculations. However, I'm not sure that I agree with what you say - the 'maximum Zs values' are determined by dividing voltage by the minimum current required to achieve the required disconnection time. If you used a different voltage for that calculation, you would surely get a different answer? Am I missing something?

[as you will be aware, the proposal for next year's Amd3 of BS7671 involves very extensive changes to all the 'maximum Zs' figures to take into account the fact that the actual supply voltage may be less than the nominal 230V]

In any event, in the figures I presented, I did not calculate the 'maximum Zs figures' (for ADS), which I just took from the OSG. It was only when calculating the minimum Zs to provide adequate fault protection of conductors (adiabatic calculation) that I used a voltage, and for that I used the nominal 230V.

Kind Regards, John

 

[EFLImpudence]

230V should be used because that is what is used to work out the maximum Zs values. If the actual voltage were used with the corresponding values, would not the result be the same? Am I missing something?

As I said, I believe use of the nominal voltage is conventional for such calculations. However, I'm not sure that I agree with what you say - the 'maximum Zs values' are determined by dividing voltage by the minimum current required to achieve the required disconnection time. If you used a different voltage for that calculation, you would surely get a different answer? Am I missing something?

Not sure.
My point was that the maximum Zs is proportional to the voltage.
So, with our meters, at least, calculating Zs from the actual voltage but we allowing only the 230V maximum then this will include some tolerance.

Although the R1+R2 of the circuit in question will not alter, with a higher voltage a higher Zs would technically be allowed.
So, we should use 230V for calculations or alter all of the parameters.

With a higher voltage and same R1+R2, PEFC will be higher causing the OPD to operate quicker and result in the CPC acceptability, or not, being unaffected.

as you will be aware, the proposal for next year's Amd3 of BS7671 involves very extensive changes to all the 'maximum Zs' figures to take into account the fact that the actual supply voltage may be less than the nominal 230V

I was not aware.
That doesn't seem logical.

 

[JohnW2]

Not sure. My point was that the maximum Zs is proportional to the voltage. So, with our meters, at least, calculating Zs from the actual voltage but we allowing only the 230V maximum then this will include some tolerance. Although the R1+R2 of the circuit in question will not alter, with a higher voltage a higher Zs would technically be allowed. So, we should use 230V for calculations or alter all of the parameters.

Exactly. If one calculated Zs using the actual voltage, if that voltage happened to be high at the time of measurement (e.g. 240V or 250V) and the Zs calculated on that basis was low enough for satisfactory ADS at that voltage, if supply voltage were to fall (say to the nominal 230V, or even lower), PEFC may then become inadequate to satisfy ADS requirements. That is, I presume, why the regs plan to change the 'maximum Zs figures' to take into account the possibility that the supply voltage might be up to 6% below the nominal 230V.

That our meters determine Zs by dividing actual voltage (at the time) by actual current (at the time, at that voltage) is correct, and results in a true figure for the actual loop impedance, regardless of the voltage at the time. You would still get a correct figure for Zs even if (hypothetically) supply voltage were 100V or 500V at the time!

With a higher voltage and same R1+R2, PEFC will be higher causing the OPD to operate quicker and result in the CPC acceptability, or not, being unaffected.

Are you now talking about my adiabatic calculations? If so,you are essentially right - but, as discussed, in practice we are not able to determine (from readily available data) the extent of that reduction in disconnection time - but, rather, are having to 'guess' a 0.1 sec figure.

as you will be aware, the proposal for next year's Amd3 of BS7671 involves very extensive changes to all the 'maximum Zs' figures to take into account the fact that the actual supply voltage may be less than the nominal 230V

I was not aware. That doesn't seem logical.

See above - it seems pretty logical to me. As I understand it, they are seeking to ensure that the 'maximum Zs' figures they quote are low enough to result in satisfactory disconnection times even if supply voltage is only 230V-6%. Is that not logical?

Kind Regards, John

 

[JohnW2]

as you will be aware, the proposal for next year's Amd3 of BS7671 involves very extensive changes to all the 'maximum Zs' figures to take into account the fact that the actual supply voltage may be less than the nominal 230V

I was not aware. ...

FYI, an example of the proposed changes:

Untitled

• JohnW2
• 6 Apr 2014
• 1

Kind Regards, John

 

[EFLImpudence]

With a higher voltage and same R1+R2, PEFC will be higher causing the OPD to operate quicker and result in the CPC acceptability, or not, being unaffected.

Are you now talking about my adiabatic calculations? If so,you are essentially right - but, as discussed, in practice we are not able to determine (from readily available data) the extent of that reduction in disconnection time - but, rather, are having to 'guess' a 0.1 sec figure.

In a way, I'm just saying that if 230V and its 80% Zs and 0.1s are used then the result will be well on the conservative side.

as you will be aware, the proposal for next year's Amd3 of BS7671 involves very extensive changes to all the 'maximum Zs' figures to take into account the fact that the actual supply voltage may be less than the nominal 230V

I was not aware. That doesn't seem logical.

See above - it seems pretty logical to me. As I understand it, they are seeking to ensure that the 'maximum Zs' figures they quote are low enough to result in satisfactory disconnection times even if supply voltage is only 230V-6%. Is that not logical?

Perhaps illogical was not the right word.

However the situation was not exactly unforeseeable and is less worrisome since the reduction in nominal voltage.

Is the nominal voltage now 220V + 14.5% - 1.8%


Would it be too much to expect that if the actual voltage were less than 230V an adjustment could be made rather than a blanket alteration that may never be actually required or can someone who knows convert the actual figures obtained?

 

[JohnW2]

In a way, I'm just saying that if 230V and its 80% Zs and 0.1s are used then the result will be well on the conservative side.

Agreed.

Perhaps illogical was not the right word. However the situation was not exactly unforeseeable .... Would it be too much to expect that if the actual voltage were less than 230V an adjustment could be made rather than a blanket alteration that may never be actually required or can someone who knows convert the actual figures obtained?

Presumably they (but not you) are considering the fact that (although rare), supply voltage can theoretically vary widely from time to time - i.e. in the range 216.2V to 253V. If you designed an installation to have satisfactory disconnection times on the basis of the voltage at the time you measured it, or even (as they now seem to have realised) on the basis of the nominal voltage of 230V, then if (as is 'permitted') the voltage on some subsequent day fell to (a 'permitted') 216.2V, the disconnection time requirements might then not be satisfied. The new proposals mean that disconnection times will be satisfied at any 'permitted' supply voltage.

.... and is less worrisome since the reduction in nominal voltage. ... Is the nominal voltage now 220V + 14.5% - 1.8%

Wot's that all about?! Have I missed something? Whatever, it makes sense to me that the 'maximum Zs' figures quoted should be such that acceptable disconnection times are achieved even at the 'lowest permitted' supply voltage, regardless of how the permitted range of supply voltages is expressed!

Kind Regards, John