Ovens, plugs, kitchen rings and diversity

[JohnW2]

I'm a bit lost here. What is the difference (other than the obvious!) between two elements cycling independently

They won't be cycling independantly. The thermostat will switch them both on and off together.

If they're both in the same oven/cavity, yes, but ....

Indeed, the (OSG) diversity formula everyone seems to use would be applicable to a single 4.68 kW element which was time-cycling, wouldn't it?

Would it? I am saying "No".

This appears to be the problem. You don't seem to accept that diversity can apply across time (even if there is only a single, time-cycling, load) as well as 'space' (multiple loads, independently cycling). If you continue not to accept this, I fear that the discussion probably cannot progress ...

... consider some simple hypothetical situations (using the previous example numbers). Firstly, two 2.34 kW (at 240V) elements (say hobs), each drawing about 9.34 (so 18.68A when both were on), each independently time-cycled by a thermostat such that they had an approximately 50/50 on/off ratio. Over any moderate period of time, the time-averaged current would be approximately 9.34A (although the maximum instantaneous current will still be 18.68A, at some points in time). That is why diversity is allowed, and you would presumably accept that.

Secondly, consider a single 4.68 kW (18.68A at 230V) element, again time-cycled by a thermostat such that it had an approximately 50/50 on/off ratio. Again, over any moderate period of time, the time-averaged current would be approximately 9.34A (and the instantaneous current either 18.68A or zero). However, if I understand what you're saying, in this situation you would not accept the application of diversity, despite the time-averaged current being exactly the same as in the first scenario. If I'm understanding you correctly, can you perhaps explain the basis of your reasoning?

Well, yes, the regs themselves are obviously not retrospective - but manufacturers can (and often do) 'go beyond the regs'!

Intentionally?

Seemingly. Where, for example, in the regs will you find a requirement for fans fed from 6A lighting circuits to have 3A FCUs or 3-pole isolators?

However, the shower itself has an all plastic exterior.

Which is presumably not totally immune from being broken.

KindRegards, John

 

[EFLImpudence]

This appears to be the problem. You don't seem to accept that diversity can apply across time (even if there is only a single, time-cycling, load) as well as 'space' (multiple loads, independently cycling). If you continue not to accept this, I fear that the discussion probably cannot progress ...

You are saying I'm wrong so I will try and learn.

... consider some simple hypothetical situations (using the previous example numbers). Firstly, two 2.34 kW (at 240V) elements (say hobs), each drawing about 9.34 (so 18.68A when both were on), each independently time-cycled by a thermostat such that they had an approximately 50/50 on/off ratio. Over any moderate period of time, the time-averaged current would be approximately 9.34A (although the maximum instantaneous current will still be 18.68A, at some points in time). That is why diversity is allowed, and you would presumably accept that.

If those numbers are correct and acceptable, then I would but I am unsure whether that is the case.
As you have probably looked into it I will accept it.

Secondly, consider a single 4.68 kW (18.68A at 230V) element, again time-cycled by a thermostat such that it had an approximately 50/50 on/off ratio. Again, over any moderate period of time, the time-averaged current would be approximately 9.34A (and the instantaneous current either 18.68A or zero). However, if I understand what you're saying, in this situation you would not accept the application of diversity, despite the time-averaged current being exactly the same as in the first scenario.

That is my position.
The reverse diversity of 13A would be a 20A load.
Are you saying it would be alright to put a 20A single load oven on a 13A plug?

If so, then there would seem to be little need to recommend that >2kW loads should not be on a ring circuit.

If I'm understanding you correctly, can you perhaps explain the basis of your reasoning?

Probably not. It just seems sensible.

I can understand the reasoning but I still do not consider it the same.

Would you consider it acceptable, as with a 12kW cooker being on a 22A opd and 22A cable, to have the same for a 12kW kiln, for example, which was either on or off with cycling?

Seemingly. Where, for example, in the regs will you find a requirement for fans fed from 6A lighting circuits to have 3A FCUs or 3-pole isolators?

Is it not commonly accepted that that is either unnecessary or because of poor manufacturers omitting integral fuses?

Which is presumably not totally immune from being broken

I have not seen that mentioned in any regulation.

 

[JohnW2]

This appears to be the problem. You don't seem to accept that diversity can apply across time (even if there is only a single, time-cycling, load) as well as 'space' (multiple loads, independently cycling). If you continue not to accept this, I fear that the discussion probably cannot progress ...

You are saying I'm wrong so I will try and learn.

It's not necessarily that simple - it could be me who is wrong! I can tell you as much as you want about statistical averaging processes, but I can't tell you for sure what approach is considered appropriate for diversity calculations in electrical installations. As you are aware, BS7671 is totally silent on details of diversity calculations/allowances, whilst the OSG (upon which everyone seems to rely) gives no indication of the thinking behind the numerical details of the diversity allowances it suggests.

... consider some simple hypothetical situations (using the previous example numbers). Firstly, two 2.34 kW (at 240V) elements (say hobs), each drawing about 9.34 (so 18.68A when both were on), each independently time-cycled by a thermostat such that they had an approximately 50/50 on/off ratio. Over any moderate period of time, the time-averaged current would be approximately 9.34A (although the maximum instantaneous current will still be 18.68A, at some points in time). That is why diversity is allowed, and you would presumably accept that.

If those numbers are correct and acceptable, then I would but I am unsure whether that is the case. As you have probably looked into it I will accept it.

The current figures are certainly correct, as you can work out as well as I can. If each element is on for 50% of the time, then the average current over a period of time will be 4.67A (9.34A / 2), so the average for two of them over a period of time will be 9.34A (4.67A x 2). I wasn't trying to make a numerical point, since (as above) we haven't a clue as to how the 'standard diversity formula' was arrived at. I was merely pointing out that the fact that, conceptually, it is considered appropriate to apply diversity becasue the average current over a period of time is less that the 'full current' of the elements (even though the current at some points in time will be as high as the 'full current' of both elements (18.68A in above example).

Secondly, consider a single 4.68 kW (18.68A at 230V) element, again time-cycled by a thermostat such that it had an approximately 50/50 on/off ratio. Again, over any moderate period of time, the time-averaged current would be approximately 9.34A (and the instantaneous current either 18.68A or zero). However, if I understand what you're saying, in this situation you would not accept the application of diversity, despite the time-averaged current being exactly the same as in the first scenario.

That is my position.

Fair enough, but I'm not sure that it's necessarily logical/rational. Of one thing we can be sure (as BAS has been saying), namely that the OSG makes no distinction between different types of 'cooking appliance' load - so someone following the word of the OSG would apply the same diversity calculation for any cooking load, whether it had one element or 8 'independently cycling' ones. ... I'll try to produce some illustrations for you. The only difference between the two scenarios I've described is that, on average, the current will be at the maximum (without diversity) figure for different proportions of the time, even though the average current over time will be the same.

The reverse diversity of 13A would be a 20A load. Are you saying it would be alright to put a 20A single load oven on a 13A plug?

I'm suggesting/asking, rather than "saying" - but IF what one is concerned about is the average current over time, then putting a 20A "single load" on a 13A plug would be no different from putting 4 x 5A independently cycling (cooking) loads on a 13A plug (which I think you would find acceptable?).

If so, then there would seem to be little need to recommend that >2kW loads should not be on a ring circuit.

That's not really got much to do with diversity - it is one of the measures which Appendix 15 suggests as a means of minimising the risk of overloading any part of a ring final for long periods of time.

Would you consider it acceptable, as with a 12kW cooker being on a 22A opd and 22A cable, to have the same for a 12kW kiln, for example, which was either on or off with cycling?

I don't think the OSG has anything to say about what diversity (if any) can be applied to a kiln, but if you changed that to "12 kW mega-oven" then that certainly is what someone following the OSG would say and, as above, IF what one is concerned about is the average current over time, then it would seemingly be logical.

Kind Regards, John

 

[JohnW2]

Seemingly. Where, for example, in the regs will you find a requirement for fans fed from 6A lighting circuits to have 3A FCUs or 3-pole isolators?

Is it not commonly accepted that that is either unnecessary or because of poor manufacturers omitting integral fuses?

Quite so - but, whatever the thinking behind it, it's an example of MIs "intentionally" requiring something which the regs don't, isn't it? (and that was the question I was responding to).

Which is presumably not totally immune from being broken

I have not seen that mentioned in any regulation.

I'm not saying it is, but the fact that the insulating casing of a shower could get/be damaged, in the presence of electricity, water, wet naked people and earthed metal, perhaps explains why it 'feels right' to have RCD protection!

Kind Regards, John

 

[EFLImpudence]

whilst the OSG (upon which everyone seems to rely)

Nooooo, it just happens in this case to give the same information as other publications and has been satisfactory for decades.

... consider some simple hypothetical situations (using the previous example numbers). Firstly, two 2.34 kW (at 240V) elements (say hobs), each drawing about 9.34 (so 18.68A when both were on), each independently time-cycled by a thermostat such that they had an approximately 50/50 on/off ratio. Over any moderate period of time, the time-averaged current would be approximately 9.34A (although the maximum instantaneous current will still be 18.68A, at some points in time). That is why diversity is allowed, and you would presumably accept that.

Yes, I can do the maths but where does the 50/50 come from?

I'm suggesting/asking, rather than "saying" - but IF what one is concerned about is the average current over time, then putting a 20A "single load" on a 13A plug would be no different from putting 4 x 5A independently cycling (cooking) loads on a 13A plug (which I think you would find acceptable?).

Mathematically true but it is unlikely that 4 x 5A are all on at the same time. I thought that was what diversity was all about.
I do not know how to work out the 'time' diversity (I don't mean the maths but the formula).
10A + 30% has nothing to do with time, does it? It is no good for working out how long it would be before the MCB was damaged, surely.


I have seen elsewhere it is thought acceptable for a 44A shower to have a 40A MCB because it is not on very long.
Short period overcurrent is not defined, is it?

IF what one is concerned about is the average current over time, then it would seemingly be logical.

It would, but averages ...


Why is diversity not allowed below 10A?

 

[JohnW2]

whilst the OSG (upon which everyone seems to rely)

Nooooo, it just happens in this case to give the same information as other publications and has been satisfactory for decades.

Fair enough - you're clearly one up on me. I'm not aware of these 'other publications' - the only definitive guidance/advice on diversity calcs I'm aware of is that in the OSG (or other documents which are citing/using the OSG figures)

Yes, I can do the maths but where does the 50/50 come from?

It was totally arbitrary, for illustration. I could have used any other duty ratio I wanted (30/70, 20/80 or whatever) to illustrate that the average current over time would be the same whether there was 'one load' or 'N independently cycling loads'.

Mathematically true but it is unlikely that 4 x 5A are all on at the same time. I thought that was what diversity was all about. I do not know how to work out the 'time' diversity (I don't mean the maths but the formula).

[I'm not sure what distinction you are making between "maths" and "formula" ] As I keep implying (with my emboldened and capitalised IFs), I don't know for sure what aspects of the pattern of current are considered most important - time-averaged current or what. As you say, the more independently cycling loads there are, the lower the proportion of time will they all be 'on' simultaneously - but I don't know to what extent that 'matters' if the average current over time is the same.

10A + 30% has nothing to do with time, does it? It is no good for working out how long it would be before the MCB was damaged, surely.

I'm not sure that overcurrent necessarily "damages" MCBs - it causes them to operate, doesn't it. As for where 10A+30% comes from, that's the $64,000 question. I can but presume that it started life as a result of empirical observation of how cookers behaved in normal use. However, as I've often said, those figures seem to have been unchanged for a very long time, and therefore are based on cooking appliances and 'cooking practices' of many decades ago, hence not necessarily still totally appropriate.

I have seen elsewhere it is thought acceptable for a 44A shower to have a 40A MCB because it is not on very long. Short period overcurrent is not defined, is it?

I think that's just a case of exploiting the built-in safety margins, in violation of the regs. It's different for cookers, since, whilst the regs still require that In≥Ib, that Ib is allowed to be the after diversity design current. Mind you, having said all that, the spec of a B40 requires that I1 is 1.33In - so, if it's required to allow 45.2A (40A x 1.13) to flow 'indefinitely', one can but assume that it is regarded as 'safe' for it to carry that current 'indefinitely'!

Why is diversity not allowed below 10A?

As above, "10A + 30%" probably resulted from mathematical modelling of the observed behaviour of cookers in service - so that's just 'how it is'. However, it's probably easier to conceptualise if you turn your question/statement on it's head - and say that diversity is only allowed when one gets to higher total loads, which probably/presumably happens far less often (many meals will involve only one or two hob elements). Also, mathematically speaking, the averaging will not work so well when there are less loads (hence lower total current), so it's not unreasonable that there should be an initial chunk of current immune from diversity.

Kind Regards, John

 

[ban-all-sheds]

... consider some simple hypothetical situations (using the previous example numbers). Firstly, two 2.34 kW (at 240V) elements (say hobs), each drawing about 9.34 (so 18.68A when both were on), each independently time-cycled by a thermostat such that they had an approximately 50/50 on/off ratio. Over any moderate period of time, the time-averaged current would be approximately 9.34A (although the maximum instantaneous current will still be 18.68A, at some points in time). That is why diversity is allowed, and you would presumably accept that.

Secondly, consider a single 4.68 kW (18.68A at 230V) element, again time-cycled by a thermostat such that it had an approximately 50/50 on/off ratio. Again, over any moderate period of time, the time-averaged current would be approximately 9.34A (and the instantaneous current either 18.68A or zero).

We could also consider the IET (or IEE as was) when they produced that glorious fudge to allow ring finals to continue to be installed.

Although they didn't use the term, they did use the concept of diversity over time in conjunction with the requirement that under the
intended conditions of use the load current in any part of the ring circuit should be unlikely to exceed for long periods the current carrying capacity of the cable.

And then there's 533.2.1....

 

[ban-all-sheds]

The reverse diversity of 13A would be a 20A load.
Are you saying it would be alright to put a 20A single load oven on a 13A plug?

If so, then there would seem to be little need to recommend that >2kW loads should not be on a ring circuit.

That's a separate issue. I thought it was the use of a plug which you were challenging? Would you be quite happy with such an oven plugged into a radial?

 

[ban-all-sheds]

I have seen elsewhere it is thought acceptable for a 44A shower to have a 40A MCB because it is not on very long.
Short period overcurrent is not defined, is it?

Is not the implication of what 533.2.1 says about thermally equivalent constant loads and I2, and what the Wiring Matters article above says, that "short period" means "less than the conventional time for the protective device"?

 

[JohnW2]

We could also consider the IET (or IEE as was) when they produced that glorious fudge to allow ring finals to continue to be installed. ... Although they didn't use the term, they did use the concept of diversity over time in conjunction with the requirement that under the intended conditions of use the load current in any part of the ring circuit should be unlikely to exceed for long periods the current carrying capacity of the cable.

Indeed - but, of course, the one type of kitchen load they did not consider is the one we are discussing - namely cooking appliances.

Kind Regards, John

 

[EFLImpudence]

I thought it was the use of a plug which you were challenging?

Well, that was the subject of the thread but as usual we have wandered.

Would you be quite happy with such an oven plugged into a radial?

No, I don't see a difference in this respect.

 

[JohnW2]

I have seen elsewhere it is thought acceptable for a 44A shower to have a 40A MCB because it is not on very long. Short period overcurrent is not defined, is it?

Is not the implication of what 533.2.1 says about thermally equivalent constant loads and I2, and what the Wiring Matters article above says, that "short period" means "less than the conventional time for the protective device"?

That bit of 533.2.1 relates to 'cyclic loads' - do showers actually qualify as that? If they do, then this is,in effect, diversity, even though our dear OSG does not appear to allow diversity to be be applied to showers.

Kind Regards, John

 

[EFLImpudence]

Is not the implication of what 533.2.1 says about thermally equivalent constant loads and I2, and what the Wiring Matters article above says, that "short period" means "less than the conventional time for the protective device"?

Fair enough.

I presume the conventional time is per cycle period of the appliance or could it be a total of the cycled periods?
I would be unsure how to calculate the figures.

 

[ban-all-sheds]

Indeed - but, of course, the one type of kitchen load they did not consider is the one we are discussing - namely cooking appliances.

I know - I posted that extract from Wiring Matters to show the use of the concept of diversity over time.

 

[winston1]

If one assumes that (as usual) the 2.34 kW in your (my) example relates to 240V, then at 230V, each of them will consume about 9.34A - i.e. 18.68A for two of them. Applying the usual calculation, that amounts to an after-diversity current of about 12.6A ... which, rightly or wrongly, is regarded as the maximum average over a period of time that the appliance (the two elements) would be expected to draw.

Why have we got this 230/240v nonsense again?
2.34 Kw relates to 240v. Nobody has a 230v supply, it is 240v (measure it). At 240v each will consume 9.75A, 19.5 for two of them. After diversity 13.15A so too much for a 13A plug.