B-curve vs C-curve

Following on from this post, it reminded me of something I was going to bring up. I'll start another thread rather than hijack that one ...

JohnW2 said:

I would hope that they won't ("become commonplace", unnecessarily), but I fear that aptsys may well be right, and everyone knows the reason as well as I do. Not installing them requires thought and a sensible/intelligent judgement, but fitting them can be done mindlessly.

Click to expand...

I've never seen C-curve MCBs commonly used - even where they would be suitable. My assumption is that this is the same reason - "everyone uses B curve, and it needs thought to use C curve" so no-one bothers doing the thinking required to see if C curve would be appropriate.
One of the things that set me off thinking about this is that SBS are stating to replace their AC type RCBOs with A type - but at present only do a few sizes with C curve which are aimed (or at least, provided because they need them) primarily at EV charger installations.

So looking at a few sets of measurements (my house, mum's house, church) I see that most if not all circuits would take a C curve.
In fact, at mum's house the garage supply (20A to suit the existing 2.5mm² T&E cable) was swapped to C curve as the RCBO couldn't start many power tools without tripping. I'd initially assumed that this swap would be a problem, but looking at the figures on the installation certificate (it had a new CU and partial rewire) it was well within spec for a C curve breaker.

SimonH2 said:

Following on from this post, it reminded me of something I was going to bring up. I'll start another thread rather than hijack that one ...

I've never seen C-curve MCBs commonly used - even where they would be suitable. My assumption is that this is the same reason - "everyone uses B curve, and it needs thought to use C curve" so no-one bothers doing the thinking required to see if C curve would be appropriate.
One of the things that set me off thinking about this is that SBS are stating to replace their AC type RCBOs with A type - but at present only do a few sizes with C curve which are aimed (or at least, provided because they need them) primarily at EV charger installations.

So looking at a few sets of measurements (my house, mum's house, church) I see that most if not all circuits would take a C curve.
In fact, at mum's house the garage supply (20A to suit the existing 2.5mm² T&E cable) was swapped to C curve as the RCBO couldn't start many power tools without tripping. I'd initially assumed that this swap would be a problem, but looking at the figures on the installation certificate (it had a new CU and partial rewire) it was well within spec for a C curve breaker.

Click to expand...

Bit of a different scenario, but quite common for industrial power supplies to be fed from C curve MCB's, as per manufacturer spec.

SimonH2 said:

Following on from this post, it reminded me of something I was going to bring up. I'll start another thread rather than hijack that one ... I've never seen C-curve MCBs commonly used - even where they would be suitable. My assumption is that this is the same reason - "everyone uses B curve, and it needs thought to use C curve" so no-one bothers doing the thinking required to see if C curve would be appropriate.

Click to expand...

There's undoubtedly some of that - most things in life, including electrical installations, have examples of "we do it that way because that's what everyone does" or "we do it that way because we always have done", without any thought about the wisdom (or even correctness) of the practice.

However, in this case, I'm not sure that there will be all that many instances of what you suggest. It is (as below) desirable that one should use a MCB with as low a magnetic trip threshold as possible - and for most domestic (and many other) circuits a Type B is perfectly satisfactory/appropriate. In the few cases in which a Type C is needed (e.g. because of trips on start-up of something), then a Type C would, if possible, be installed.

SimonH2 said:

So looking at a few sets of measurements (my house, mum's house, church) I see that most if not all circuits would take a C curve.

Click to expand...

Well, the question which first came to mind was "why would you want to change to Type C", but then yoiu go on to say ...

SimonH2 said:

In fact, at mum's house the garage supply (20A to suit the existing 2.5mm² T&E cable) was swapped to C curve as the RCBO couldn't start many power tools without tripping.

Click to expand...

I have to say that, in my experience, that is very unusual. I have only Type B MCBs and a lot of power tools, including a 'substantial' table saw, drill press and lathe, but cannot recall ever having experienced an MCB trip on start-up of any of those tools.

As above, it is surely always best to have as low a magnetic trip threshold as possible (without getting 'nuisance' trips) and, as I've implied, I think it is probably very rare (certainly in domestic installations) for a Type B to be inadequate. Indeed, I would think that there could even be an argument for MCBs with a trip threshold lower than 5In (was there ever an A curve" one?) in many cases ...

SimonH2 said:

I'd initially assumed that this swap would be a problem, but looking at the figures on the installation certificate (it had a new CU and partial rewire) it was well within spec for a C curve breaker.

Click to expand...

In some senses, we're almost back to (the opposite of) what you said at the start (and what I had previously written about other devices", in that you are almost suggesting that one should use Type C devices "if one can", rather than because they are necessary. "If one can" implies that the Zs is likely to be fairly marginal, which means that one has to get very close to the hypothetical "zero impedance fault" to be guaranteed to achieve the expected (and 'required') disconnection time.

Anyone who has ever 'touched conductors together' in order, say, to 'zero' a low-resistance measuring device, knows that it's only too easy to get at least 'a fraction of an ohm' resistance between the conductors - and, if the Zs is fairly 'marginal', a fault which has an impedance of "a fraction of an ohm" (rather than zero) could well be enough to reduce the fault current below the magnetic trip threshold. In contrast, if one uses ('without problems') a Type B on a circuit on which one "could" use a Type C, then one can have a fault with an impedance 'appreciably' higher than zero and still get magnetic tripping.

Kind Regards, John

I use C for lights.

AndyPRK said:

I use C for lights.

Click to expand...

In the days of incandescents, particularly circuits fairly heavily loaded with incandescents, then that would certainly have made sense, since it was relatively common for a dying incandescent bulb/lamp to trip a B6. However, I have hardly ever (if ever at all) experienced a trip when a CFL or LED has died, so I'm not sure that it's any longer really necessary. Do you intend to continue the practice?

As I just wrote, to use a C-curve MCB when a B-curve one 'would be adequate' greatly increases the risk that a slightly-non-zero-impedance fault will fail to achieve the expected/'required' disconnection time.

Kind Regards, John

If I were to still have incandescents and C were easy to get, yes.

I don’t want a whole light circuit taken out.

I’m not sure if I’m as trusting as you on cfls but don’t have experance to say otherwise.

AndyPRK said:

If I were to still have incandescents and C were easy to get, yes.

Click to expand...

Sure, and I might even be tempted - but incandescents are essentially history. What very few I still have are ones in such rarely-used parts of the house that i haven't yet got around to changing them!

AndyPRK said:

I’m not sure if I’m as trusting as you on cfls but don’t have experance to say otherwise.

Click to expand...

Well, for me, CFLs are almost 'history', but I certainly don't recall any have tripped MCBs - probably because their mode of death is very rarely "going bang" in the fashion that incandescents do. I've had mainly LEDs for a few years (a few CFLs left) and, whilst they do die occasionally, I'm pretty sure than none has ever tripped a B6 - most, even the cheapest ones, seem to have pretty trigger-happy 'fusible links' in them, which seem to blow before an MCB trips in the case of an 'over-current death'.

Kind Regards, John

JohnW2 said:

I have to say that, in my experience, that is very unusual. I have only Type B MCBs and a lot of power tools, including a 'substantial' table saw, drill press and lathe, but cannot recall ever having experienced an MCB trip on start-up of any of those tools.

Click to expand...

In this case, it was a handheld circular saw and it almost consistently tripped the breaker. Didn't trip it after it was changed to a C curve.

SimonH2 said:

In this case, it was a handheld circular saw and it almost consistently tripped the breaker. Didn't trip it after it was changed to a C curve.

Click to expand...

Interesting. I've been using power tools of all sorts for decades (admittedly with fuses, rather than MCBs in the earlier years), but honestly cannot recall that ever happening to me. If it was just one tool that did it, presumably there was something 'unusual' about that particular tool?

Kind Regards, John

I guess me don't use power tools on a B20

AndyPRK said:

I guess me don't use power tools on a B20

Click to expand...

Well, I certainly never have. In fact, I've never had a circuit with a B20. All my sockets circuits (2.5mm² rings and 4mm² radials) have B32s, with just a few odd things (immersions etc.) having B16s, and B6s for lighting - but no B20s.

As I've often said, unless in 'special circumstances', I've never been much of a fan of 20A radials - since it seems a bit odd to put multiple sockets on such a circuit when it could theoretically be overloaded with what was just plugged into one double socket.

Kind Regards, John

JohnW2 said:

... I've never been much of a fan of 20A radials - since it seems a bit odd to put multiple sockets on such a circuit when it could theoretically be overloaded with what was just plugged into one double socket.

Click to expand...

Doesn't the same argument apply to most things ? We can overload an RFC using just 3 sockets (1 ½ double sockets), in theory many 32A cooker circuits could be overloaded by turning on all the oven(s) and rings (and put the kettle on plugged into the socket on the cooker switch), and our house is all run through one 60A fuse in the service head.
But in practice, we don't tend to use all that much equipment at once. The most severe these days is probably the kitchen with washing machine, tumble drier, kettle etc - but even then there's a lot of diversity in loads.

Before the RCD the earth loop impedance requirements would often stop one using a type C MCB, even with a RCD if you look at the prospective short circuit current if you want it to trip on the magnetic part then often type C is a non starter, you need 30 amp to trip a B6 but 60 amp to trip a C6 and on lighting that can mean the difference between bulb contacts welding on to the holder or not, so with lights B6 every time, I used a B16 for an outside light, mistake, had to swap whole lamp when Ikea bulb welded onto B22d bulb holder.

As to ring final 32 x 10 = 320A = 0.72Ω with 5% safety margin now required 0.68Ω minus the incoming impedance = around 0.33Ω this would mean a very short ring final. With a B32 the limit is volt drop normally considered as 106 meters, with a type C down to around 61.5 meters assuming incomer at 0.35Ω

The incomer loop impedance is also some thing to consider, it may measure 0.20Ω when you measure it, but changes to the supply could mean it raises to 0.35Ω

SimonH2 said:

Doesn't the same argument apply to most things ? We can overload an RFC using just 3 sockets (1 ½ double sockets), in theory many 32A cooker circuits could be overloaded by turning on all the oven(s) and rings (and put the kettle on plugged into the socket on the cooker switch), and our house is all run through one 60A fuse in the service head.
But in practice, we don't tend to use all that much equipment at once. The most severe these days is probably the kitchen with washing machine, tumble drier, kettle etc - but even then there's a lot of diversity in loads.

Click to expand...

Using same calculations as previous post with a C type looking at just 30 meters however working on volt drop not much better around 32.5 meters so yes unlike the ring not much in it.

SimonH2 said:

Doesn't the same argument apply to most things ? We can overload an RFC using just 3 sockets (1 ½ double sockets), in theory many 32A cooker circuits could be overloaded by turning on all the oven(s) and rings (and put the kettle on plugged into the socket on the cooker switch), and our house is all run through one 60A fuse in the service head. But in practice, we don't tend to use all that much equipment at once.

Click to expand...

That's what people always say, but it feels a bit different to me. A multi-sockets circuit which is not guaranteed to be able to cope with more than one plugged-in load seems a bit marginal to me. I agree that even 32A is not very large in comparison with 13A, but ~2.5 is a lot bigger than ~1.5 - so there is a fair bit more scope for 'diversity'.

SimonH2 said:

The most severe these days is probably the kitchen ....

Click to expand...

Indeed - but, ironically, kitchens seem to be the places in relation to which one most often hears 20A radials being suggested. I would have far fewer 'concerns' (if that's the right word) if one was talking about a circuit supplying receptions rooms, bedrooms or whatever.

However, it's just a personal thing, and I certainly agree that, in reality, one is virtually never going to have any problems with 20A radials. However, I would say exactly the same in relation to lots of things (like multiple sockets on an unfused spur from a ring final)!

Kind Regards, John