Power cable to workshop/garage

[JohnW2]

A waveguide carries electrical energy from A to B. Mentioning 50Hz merely reminds us that frequency is a factor to consider, and that at some frequencies cables might need to become hollow.

The skin effect only starts to have an effect at very high frequencies.

Quite so - although it is seemingly enough at 50Hz to persuade them to sometimes use groups of multiple (usually 4) smaller conductors, rather than one large one, for overhead EHV transmission (well, at least, I've always assumed that is the reason, since it's what I was taught at school!). In any event, BAS seems to be thinking of using the outer wall of a waveguide as a single conductor, rather than using the waveguide as a waveguide. As I implied, a true 50Hz waveguide would surely be ludicrously large! (IIRC, the 'lower cutoff freqency' of a waveguide is half a wavelength!). The other mistake BAS seems to be making is to imply that one has 'right angle' corners in waveguides - which, if one has them, are, I think, extremely inefficient.

Being pedantic there is no such thing as a square corner. Nearest is a right angle corner. But a square has four corners.

That's true. However, since the OP had asked whether one could have "90° corners", thought it would confuse him if I had said "yes, but you can't have a right-angled corner" - hence I wrote 'square'. I should have written 'zero radius bend'. However, I'm not convinced that your pedanticism is necessarily correct - cannot "square" be used as an adjective to mean "right-angled"?

Kind Regards, John

 

[ban-all-sheds]

In any event, BAS seems to be thinking of using the outer wall of a waveguide as a single conductor, rather than using the waveguide as a waveguide.

No I wasn't.

The other mistake BAS seems to be making is to imply that one has 'right angle' corners in waveguides

If you look at the leftmost one:

you'll see that the bend radius is unaffected by the outside edge being curved. It's a right-angled bend.

 

[EFLImpudence]

cannot "square" be used as an adjective to mean "right-angled"?

No. Bernard is correct.

Please do not encourage the population to further unnecessarily massacre/evolve the language .

 

[ban-all-sheds]

 

[JohnW2]

cannot "square" be used as an adjective to mean "right-angled"?

No. Bernard is correct.

As BAS has kindly demonstrated for me, it would appear to be I who was correct! Have you never said something like "check that the joint is square", when you were referring to the angle between two components, not to a equal-sided right-angled quadrilateral?

Please do not encourage the population to further unnecessarily massacre/evolve the language .

In view of the above, do you wish to withdraw that request, at least as far as the word under discussion is concerned?

Kind Regards, John

 

[JohnW2]

In any event, BAS seems to be thinking of using the outer wall of a waveguide as a single conductor, rather than using the waveguide as a waveguide.

No I wasn't.

OK - so, if it were actually being used as a waveguide, what would you say would be the appropriate dimensions for one that was to operate at 50Hz?

The other mistake BAS seems to be making is to imply that one has 'right angle' corners in waveguides

If you look at the leftmost one: <pic> you'll see that the bend radius is unaffected by the outside edge being curved. It's a right-angled bend.

That's interesting - I think that I've only ever seen, or used, ones like those on the right, and have therefore assumed that ones like that one on the left would be 'lossy'. Mind you, as you say, its outside edge is curved - if that were also a 'straight right angle' ('square'!), then I suspect that it really would be 'lossy' - but I may be wrong.

Kind Regards, John

 

[EFLImpudence]

I will have to withdraw as it would seem to be too late and evolution has already occurred as verified by the inclusion of "having the shape or approximate shape of a square".

So, a trapezium could literally be a square.

 

[JohnW2]

I will have to withdraw as it would seem to be too late and evolution has already occurred as verified by the inclusion of "having the shape or approximate shape of a square".

I think that in the sense we are talking about, the evolution probably took place a very long time, maybe centuries, ago. I'm sure that my grandfather, who was initially trained (over 100 years ago) as a carpenter, talked about "making sure that joints were square".

So, a trapezium could literally be a square.

That would obviously be silly, but the definition doesn't actually allow that. I think you're missing the point that is says that about things which are "the approximate shape of a square" in relation to the use of the word as an adjective, not when it's being used as a noun. They are therefore suggesting that one could say that "a rectangle is square" (or "a rectangle is a square shape") but not "a rectangle is A square". The definition of use of the word as a noun (e.g. "A square") only allows it to be used to refer to a 'true square'. I don't think it applies to a trapezium since the examples they give (in relation to use as an adjective) of things which have "the approximate shape of a square" implies that 'square' (as an adjective) refers to having right-angled corners, which a trapezium or rhombus obviously doesn't have.

Kind Regards, John

 

[bernardgreen]

The right angled bend on the left may have a more complex curved face on the inside of the curve.

I gave up with wave guide ( as in marine radar ) technology many decades ago. Only one rule :- DO not have any part of your body near an open ended one in case some idiot ( ship's captain ) on the bridge switches the set from Standby to Run

 

[JohnW2]

The right angled bend on the left may have a more complex curved face on the inside of the curve.

Yes, that's a possibility.

Kind Regards, John

 

[ban-all-sheds]

OK - so, if it were actually being used as a waveguide, what would you say would be the appropriate dimensions for one that was to operate at 50Hz?

1) IHNI.

2) I struggle to understand what that has to do with the price of fish:

you obviously cannot have a 'square corner' in any cable!

If you consider that a waveguide is a cable for carrying microwaves, you can.

If you consider a length of copper with a rectangular cross section a cable, then you can (although you would have to make it with a bend in it).

 

[ban-all-sheds]

My point was that the regs allow a ring final for which the CCC of the cable is only 62.5% of the In of the 32A OPD

No they don't.

They allow the use of a cable with a CCC of 20A and an OPD of 32A. Not the same thing.

No other design where the CCC of the cable happens to be 62.5% of the rating of the OPD is allowed.

so a hypothetical 40A ring in which the CCC of the cable was 67.5% of the In of the OPD would be 'at least as safe, wouldn't it?

Arguably yes, but for that to be of any practical relevance you'd need an electrician prepared to design and install a non-compliant circuit, prepared to state on the EIC that he had not complied with BS 7671, and prepared to argue that his non-compliant installation was at least as safe as one which complied. Bear in mind that one of the two regulations which mention departures does so to allow the use of new materials and inventions. There is also a strong circumstantial case against your proposal given that they changed the regulations to explicitly state 20A, and that therefore they probably intended that the old requirement of

“minimum current carrying capacity Iz of the cable be not less than 0.67 x the
rated current setting In of the protective device”

no longer be used - it would have been simpler to change 0.67 to 0.625.

 

[JohnW2]

OK - so, if it were actually being used as a waveguide, what would you say would be the appropriate dimensions for one that was to operate at 50Hz?

1) IHNI. ... 2) I struggle to understand what that has to do with the price of fish:

As I said, IIRC, the 'lower cutoff frequency' of a waveguide corresponds to a waveguide width of about half a wavelength - i.e. about 3,000,000m (3,000km) at 50Hz - which doesn't sound 'practicable' to me!

you obviously cannot have a 'square corner' in any cable!

If you consider that a waveguide is a cable for carrying microwaves, you can. If you consider a length of copper with a rectangular cross section a cable, then you can (although you would have to make it with a bend in it).

You are producing semantic/terminological arguments just for the sake of it. You know fully well, as does everyone else, what I meant when I wrote that one could not have a "square corner" in any cable.

Kind Regards, John

 

[JohnW2]

My point was that the regs allow a ring final for which the CCC of the cable is only 62.5% of the In of the 32A OPD

No they don't. ... They allow the use of a cable with a CCC of 20A and an OPD of 32A. Not the same thing.

You're quibbling again. 20A is "62.5% of the In of the 32A OPD". I agree that the regs to do not say anything about allowing a ring final when the CCC of the cable is 62.5% of the In of an OPD with any In other than 32A - but I didn't say or suggest that they did.

so a hypothetical 40A ring in which the CCC of the cable was 67.5% of the In of the OPD would be 'at least as safe, wouldn't it?

Arguably yes, but for that to be of any practical relevance you'd need an electrician prepared to design and install a non-compliant circuit, prepared to state on the EIC that he had not complied with BS 7671, and prepared to argue that his non-compliant installation was at least as safe as one which complied.

Indeed. That was precisely what I was suggesting. BS7671 clearly does not allow such a design - so, as you say, any designer who wanted to use it would have to be able to argue that it was at least as safe as what BS7671 does allow.

There is also a strong circumstantial case against your proposal given that they changed the regulations to explicitly state 20A, and that therefore they probably intended that the old requirement of “minimum current carrying capacity Iz of the cable be not less than 0.67 x the rated current setting n of the protective device” no longer be used - it would have been simpler to change 0.67 to 0.625.

That is an interesting point, and I agree that it suggests that, for some reason, they maybe did not intend that the previous 'formula method' should still be used. However, if, as above, a designer can successfully argue that his/her design is at least as safe as a design which the regs allow, then I don't see that the speculated 'intent' (whatever it was) of those who wrote the current regs would be of any relevance, would it?

As I said before, if we are going to speculate about intent/reasons, I would suspect that they were not very confident that CCC always properly took into account installation methods and de-rating factors throughout the length of the cable run, and therefore were being fairly cautious by writing the reg in terms a cable of minimum CSA 2.5mm² which was allowed to have a CCC as low as 20A . However, if a real-world designer is confident that the CCC of his/her cable really is, say, 27A, then (s)he could well argue that a ring final made of such cable, and protected by a 40A OPD, would be at least as safe as something which the regs do allow.

The one argument I was expecting some people to present, but which hasn't yet been wheeled out, is that accessories to BS1363 may not be 'rated' for use in a circuit >32A. I'm not sure about the specification.

Kind Regards, John

 

[ban-all-sheds]

As I said, IIRC, the 'lower cutoff frequency' of a waveguide corresponds to a waveguide width of about half a wavelength - i.e. about 3,000,000m (3,000km) at 50Hz - which doesn't sound 'practicable' to me!

OK - that has addressed my point #1.

But since nowhere have I, or anybody else, suggested using a waveguide to carry EMR at a frequency of 50Hz I still don't see what it has to do with anything.

You are producing semantic/terminological arguments just for the sake of it. You know fully well, as does everyone else, what I meant when I wrote that one could not have a "square corner" in any cable.

Shame on me.