Is it common to find too-thin wire in an installation?

[berndj]

From reading some of the other threads, I'm the guy you guys love to hate: a homeowner who wants to add a socket outlet or two himself. By the way it's a single-phase installation, I remember a blurb in the standard about that being relevant.

I did, however, shell out for SANS 10142-whatever. I have my issues with some of the wording (ambiguity in places), but overall it's quite clear to me what I need to do to make my socket outlet(s) good.

In particular, wire size. By my reading, the correct wire size for an outlet on a 20A circuit has to be 4mm^2, not 2.5mm^2, because I'm using "installation method 3" (conductors in conduit in air, in ceiling space). That means I already can't use the 27A claimed for 2.5mm^2 wire on its packaging, and more importantly by the electrical goods sales staff, due to the method alone. Then I add some thermal derating, assuming a roof space temperature of 50C (I measured 10C higher than indoors temp, assuming worst-case 40C in Cape Town); I think the factor was about 0.65 or something - I don't have the spec in front of me. And then I find that the 4mm^2 wire is just *barely* adequate for such a 20A circuit.

It was actually quite funny, I'm normally quite a softly-softly type of guy, but there I was at the counter insisting that I wanted the thicker wire, and no, they can just give me what I asked for instead of the 2.5mm^2 they strongly suggested. In a room full of people who looked like they do this stuff all day.

Am I needlessly overengineering my new outlet? I don't think so. But what's with the supply shops claiming 2.5mm^2 is enough? When I looked at (some of!) my other pre-existing outlets, there seems to be a lot of the thinner wire around. Only a few with what looks to be a thicker wire - it might even be 4mm^2 wire from the quick look I gave it.

What's going on here? (And also, shouldn't supply shop staff be trying to get me to buy *more* copper than I need?

[murdock]

it all boils down to design and application...thats why i spent years in an apprentice school and at tech. and the doctor spent years at varsity.
if you are asking a saleman at the counter which wire to buy...good luck.

My son is like you he did 6 months with me then quit...got fired...what ever...now he gets all his spec from the saleman at the counter and does all his own electrical work.

why dont you just use rip cord so many diy people do....and as the saying goes i hear on a daily basis...but its worked for years

on a more serious note...it is very common for electricians to use 2.5 mm wire for plugs...depending on the age of the installation you will find in older installations they use to use 4 mm wire with a 30 amp circuit breaker...new installations normally have 20 amp circuit breakers...just remeber ohms law dictates everything in electricity...bigger is not always better.

[Sparks]

2.5mm cable may even be overkill. There are other factors to consider too. What will your end load be? That is the primary consideration. From there you can determine the size of cable you need which will then indicate what CB is required to protect that cable. Note: the CB protects the cable, it is not selected according to the requirements of your load.

[berndj]

Sorry, that doesn't really address what I was asking.

I just don't see a way to justify running a 20A circuit with 2.5mm^2 wire in South Africa - a pretty warm place, using installation methods that seem to be in common use. Let's take "installation method 2"'s ratings, which to me seems closest to the cable-in-conduit-on-roof-beams that is used in, at least, my house. That's already just 23A, at 30 degrees Celsius - not much margin. Now take thermal derating into account - I think 50 degrees up there in the roof is a reasonable assumption? That's a derating factor of 0.71.

23A * 0.71 = 16.33A

So I don't see a 20A circuit breaker adequately protecting a cable with a current-carrying capacity of only 16-odd ampere. Is there some rule that I haven't yet spotted that lets you wriggle out of this?

Note: I'm NOT asking, "what should I do?". I AM asking, "Is this apparently common practice kosher?"

[Sparks]

Like I said , look at the load. With the exeption of the geyser circuit you would seldom get a load in excess of 10A on a plug circuit. 6A less than what you consider max. Remember that the length of the circuit is also a factor. The geyser circuit usually being reasonably short it is acceptable. A socket outlet is only 16A, so before the 20A CB pops your switch will start frying

[berndj]

Hmm, gosh, am I speaking Martian or what? That wasn't my question.

Fact 1: most household DB's seem to be full of 20A breakers on plug circuits.
Fact 2: it seems to be popular to wire these circuits with 2.5mm^2 wire.
Interpretation 3: 2.5mm^2 wire (of any length) cannot safely carry 20A using typical installation methods.

Question: how to reconcile these?

Sparks, why do you suppose circuit length matters? Amps is amps is amps! Length is only an issue when considering voltage drop along the circuit. It doesn't affect the current carrying capacity of the wire.

[AndyD]

From the regs a 2.5mm 2-core cable in conduit in a ceiling has a max current of 23A so long as your ambient temp is 30C or less. Only when you reach 45C or higher ambient temp does the max allowable current fall below 20Amps which is the standard rating of a socket circuit breaker.

45C is pretty damn warm for a ceiling space. If the temp is higher that 45C then yes, you'd need to go with a 4mm cable. But riddle me this Batman; if you just install a single socket on the end of your cable, would you be covered because it would only be possible to plug in a single appliance which would have a load of 16Amps or less because it has a limitation of 16A plug on it?

[berndj]

From the regs a 2.5mm 2-core cable in conduit in a ceiling has a max current of 23A so long as your ambient temp is 30C or less. Only when you reach 45C or higher ambient temp does the max allowable current fall below 20Amps which is the standard rating of a socket circuit breaker.

45C is pretty damn warm for a ceiling space. If the temp is higher that 45C then yes, you'd need to go with a 4mm cable. But riddle me this Batman; if you just install a single socket on the end of your cable, would you be covered because it would only be possible to plug in a single appliance which would have a load of 16Amps or less because it has a limitation of 16A plug on it?

Are you looking at the table for rubber-insulated cables? Because as far as I know we're dealing with PVC-insulated cables. At 45 degrees I see the factor 0.79. 23A * 0.79 gives me 18A.

And no, Mr Quizmaster, of course the appliance doesn't save you. As Sparks pointed out, the CB is there to protect your cable. To hell with your appliance if it develops a fault! But you're hinting at another odd thing I picked up - the standard specifically allows (can't remember the chapter and verse) for 16A socket outlets on a 20A circuit. But I'm willing to believe that the 16A rating is a switch rating (i.e., can it interrupt that current), rather than a current-carrying capacity.

As for 45 degrees vs higher... I dunno hey, it doesn't seem that crazy to me to expect at least an occasional 50 degrees. I measured a 10 degree rise over ambient around noon on a sunny day. 40 degrees ambient isn't impossible - even in Cape Town. I suppose with the 150 degree rated cable you can get away with the thinner wire. I might use that for another project that'll come later.

Thanks for the engaging reply. It's exactly tables 6.2 and 6.10 that I had in mind.

[Sparks]

To counter the voltage drop you would use a thicker cable thereby falling within what your 20A query. Something you need to keep in mind is that the specs in the tables are the max acceptable under legislation. This is not the max as per tests by SABS or cable manufacturers. The minimum requirements expected by SABS from manufacturers are higher than SANS tables. As for the temperature, I am regularly in the roof and having lived in the heart of the Karoo I can guarantee you that 50deg. is not unrealistic or unusual. A salesman at the counter is not qualified to give any technical advice. We live in SA. Every time I want a 3pole+N CB or isolator, I have to order it. In PE it is not considered movable stock because " no other electricians ask for it". If you take a closer look at SANS10142 and the ammendments you are guaranteed to find a bucketload of inconsistencies and contradictions, even on the new COC. With so much unclear in the legislation you can forget about finding consistency in the field. My new DB CB ratings are according to what I am informed the load will be. I am quite happy to and quite often use 10A or 15A CB's for plug circuits. My logic for this being that the lower the breaker rating the higher the protection.

[Jacques#1]

Nobody is wrong here. It will depend on your installation, what cable you are using and your load factors.

Cable manufacturers also differ, it has got to do with the density of the copper etc. So in some instances, a cable that they sell as 2.5mm or 4mm is in fact 0.3mm smaller etc.

The lebgth of the run with affect the volt drop, ohms law, since cables have an inherent resistance. This in turn will affect the maximum output, but more importantly the actual power at the load. Your equipment will break 10 times faster since i.e. that shaft needs so much power to be turned by that motor, which no has limited torque etc. Heating up, catching fire and killing 50 people in the process.

Also consider how you determine the load factor for a house. For socket outlets you use 5kW for the first 100m2 and then 1kW for every additional 100m2. Why do you think this is? Its because you never use all 50 or whatever socket outlets at the same time all loaded to exactly 15 amps or whatever the case may be. Consider your ceiling using this knowledge, you would have to load that Socket outlet at 17 amps, exactly on the one day of the year its 40 DC outside. I'm not saying this won't happen, but I am not too sure that you get anything on a plugtop except for a welding machine that uses that type of power. Lets say you've got a plug in stove with two 1500W elements, and a fridge 1500W (which is buggered, so it starts up and runs for an hour at a time), all running at 13h00 in the afternoon on the hottest day.......then you have a problem, maybe, consider that you calculated your voltage drop and its withing spec, so you used a 2.5mm cable, but the amps is 19.XXXXXamps. Yet, like you pointed out the temperature only rates the cable at 16 amps, adding the safety factor built into the cable, and the fact that you'll be cooking for no longer than an hour on both plates at maximum with a buggered fridge in the same room, the cable may get a little hot, but thats about the worst that could potentially happen.

I always use 2.5mm cable for my socket outlets, but I did do a job this past week using 4mm cable because my cable run was around 40 metres, and the outlets would be used for space heating. Still overkill for the application, but I calculated the worst imagineable loads......20 amps be unit.

Point is electricians will use a bigger size cable, if the conditions ask for it, and if you would rather wire in 4mm, than do it if you would prefer that.

[AndyD]

Are you looking at the table for rubber-insulated cables?

No, the tables attached are for PVC but the correction factors table has values for rubber as well.

Incidentally I hate the term rubber. It's such a broad term it practically has no real meaning in engineering terms. I wish they would just state the material correctly like 'silicone rubber' or even better give the cable code like HO7RNF if it's neoprene rubber for example. There's dozens of different rubber products in use and they all have different temperature characteristics.[/RANT]

[123]

I am quite happy to and quite often use 10A or 15A CB's for plug circuits. My logic for this being that the lower the breaker rating the higher the protection.

Have to agree, i do the same where ever possible.

[murdock]

so long as you dony use 1.5 mm or smaller wire

[Dave A]

I find this thread fascinating.

I just don't see a way to justify running a 20A circuit with 2.5mm^2 wire in South Africa - a pretty warm place, using installation methods that seem to be in common use.

Has anyone found scorched 2.5mm behind 20A protection (other than right next to a bad connection, of course)? The justification is it seems to do the job and it's the way it's always been done.

But the interesting thing is based on the evidence here, it does not comply with SANS 1507. Now when you sign off an electrical COC, what are you signing off on? That the installation is coping under existing loads or that it meets regulation (effectively standards set by SANS codes)?

[AndyD]

....what are you signing off on? That the installation is coping under existing loads or that it meets regulation (effectively standards set by SANS codes)?

Very much the latter.

'Existing loads' are very organic values and can change radically depending on season or even time of day so unless you use a power analyzer to obtain a max, min and mean value it's not predictable. The only thing that is predictable is the circuit protection tripping at 20 amps on a socket circuit so this would be the figure to work with when referring to the regs.

The only thing in doubt here is the maximum ambient temperature of the ceiling space and the corrective value that should be used accordingly. If you come across a ceiling space that's hot enough to warrant sufficient derating of the max current then you should fail the installation on a COC for this reason. The owner of the premises would no doubt install a R50.00 fan unit and invite you back for the retest on a cooler day.