I want to install 1 - 2000W and 2 - 1500w heaters on one circuit, with one t-stat.
Is there something specific in the cec that says I can't run a 10/2 to the t-stat, from there a 10/2 to 2000W heater and then downsize to 12/2 for the next, and again to 14/2 for the last one?

Pool, your question suggests you need to learn the trade.

We're not plumbers. We don't downsize our wire after each load because the rest of the circuit won't "need" all that electricity.

Instead, we -usually- size the wire according to the size of the breaker (or fuse) we have protecting the circuit. A 20-amp breaker needs a circuit with '20 amp' wire for the ENTIRE circuit.

This is one of the basic principles of electric work. The code won't directly spell it out - but the code will tell you how to size your wire.

Nice!

62-108(1)(b) and 62-114(4). To summarize, The ampacity of the wire must be greater than the load and at least 1/3 the rating of the overcurrent device; and, the smaller wire must not be more than 7.5 meters long. Or, it must have an ampacity not less than the load and at least 80% of the overcurrent device; in which case the length isn't limited.

The code specifically allows it.

Reno see twh response but I will add another limiter. Bond wire size. This reduction works from a 20 amp breaker because the bonding wire is #14 if loomex is in use. You cannot put larger than a 25 amp breaker or the bond wire is too small for a 30 amp breaker.
You can start out in #12 and at 240 volts use #14 for up to 3600 watts or 15 X 240 volts. (Not 12 X 240) the 20 amp breaker may only supply 16 amps so the #12 may only have 1 amp more on a 20 amp breaker or the full 20 amps on the #12 with a 25 amp breaker for a total load of 4800 watts.
Those 2 rules in section 62 create inspection chaos.

The size of the bond conductor didn't hit my radar but now that I look at it, I'm going to challenge your rejection of my work based on ground conductor size.

Maybe #14 loomex was a bad example because 10-814(1) says the bond doesn't need to be larger than the largest ungrounded conductor. (That actually makes sense)

Appendix B 10-814(1) says "In any case, the bonding conductor incorporated into a cable assembly is deemed to be of adequate size for the purposes of this rule." So, I'm not changing out my 12 loomex, either.

In the American version of the CEC (2012) #14, 90C is good for 25 amps, so it can be tied onto a 70 amp circuit subject to the distance limitation.

Are you using Table 16? According to that table, a 14 bond is too small for 14 loomex at 25 amps.

I don't see how this can create inspection chaos. Just pretend there are no rules and you'll be pretty close.

Thanks, guys ,for the corrections. I had failed to note the OP was from Canada, where men are men, and money has a pretty scent

US practice would not allow such downsizing between loads.

I can't say I've ever heard of an "American" version of the Canadian Code. Though I will admit you guys speak a dialect that's a lot closer to what you hear in Minnesota - than what you hear in Yorkshire!

Personally, I like our (USA) way! 14= 15 amps, 12=20 amps, etc.

That was our way before 2012. Now we size our breaker like that but derate from the higher ampacity. I thought I read that was to make our rules consistent with yours. Maybe I'm wrong.

I was wrong once. I said I was wrong and everyone said I was right. I was wrong because I was right.

Depends where you live on your interpretation of CEC

CEC = Canadian Electrical Code
CEC = California Electrical Code.

CEC, eh because it's the CEC eh forum.

Good grief ... I never dreamed that Canada would allow wires to be sized using a different set of rules.

Next thing you know, someone will say you can mount your panels sideways!

And upside down darn that rule where the highest breaker handle cant be more than 67 inches from the finished grade.

Electricians are too stuck on whatever they saw the first day at work. The first panel had the main on top so that must be right-side-up. The first plug had the ground up, so that must be right. We're stuck on convention down to the direction of the slot on the cover screw.

I probably wouldn't downsize wires between heaters, but I'll defend my co-workers' (in C eh N eh D eh) right to do it - even if it pisses off a few old-fogey inspectors who didn't do it that way when they had tools.

Code changes that relax rules bug me, too. Not because I'm against change, but it isn't right to criticize someone's work one day and the next day start doing it that way yourself.

Thanks for your responses!

Down sizing in this fashion is electrically safe, as the heaters are fixed loads in a parallel circuit, only drawing what they require... not exactly plumbing

Bond wire was what I was looking for... thanks!!

I worked for someone who had this installation rejected and I couldn't remember why.
And now I also remember why that rejection caused a lot of eye-rolling.

So I need 14/2 and 12/2 nmd cable with a #10 bond wire in them... Awesome thanks!!
I'll just fish it into the cables, if the supplier can't have it made for me.

You re not going to find a #14 or 12 with a #10 bond wire. The bond wires are #14 in those cables.

How did I fail to see this thread was on the "Canadian"forum? Simple- I was grabbing a thread from the 'recent post' sidebar.

Now, pool, I'd like a shot at changing your opinion of plumbers!

In your house, it's common for the plumbing to be laid out like a tree ... a big line at the start, getting smaller as you near the final appliance. This is so the first pipe can supply the shower, the toilet, and the kitchen sink all at once. By the time you get to the final run to any of those things, the line is much smaller- as it needs only to supply that one.

I submit that's exactly what you're doing with your wire sizing. What can go wrong?

Well, there's a big difference with electrical work, in that we have the possibility of dead faults, and circuit breakers.

Break a pipe,and only a certain amount of water will flow. There's no sort of break that will draw more water than the pipe can pass. Not so with electricity.

The wires have to be able to handle hundreds - even thousands- of amps until the breaker trips. That dead short has the same maximum fault current, pretty much regardless of the wire size. Your only hope of saving a smaller wire is to have a smaller breaker on it.

The same matter comes up when someone suggests that we use a smaller wire for the switch leg. After all, it's only a 60 watt bulb - why not use bell wire?

That's why "down South" we don't do as you propose. One might be able to do so,if there is a fuse in the heater (tap rules).

It's probably why Mike reminds you that you still need a full-size ground wire. It's all about the fault current.

Sometimes I feel like I'm the only one who cares what the rules say. 10-814(1) says the bond doesn't need to be larger than the largest ungrounded conductor. After all, if there is a fault, it gets its current from the ungrounded conductor. Why would it have to be larger?

You are perhaps one of the gentlemen in the audience who is not laughing.
If I personally read the post TO YOU...
Apparently the voice in your head isn't as funny sounding as mine...

I jotted down a few more, that I'll save for later.

I'm going to need to test this puppy out in The Lab (the bedroom where the panel is) for my own personal knowledge.
My instincts tell me it'll be just fine.
I'll wire it up and introduce some shorts in the circuit in various places, and get back to you.

If tesla was still alive the book would be thrown at him by his own fans.

If I torch the rug, I'll blame him too.

I can tell you from my own experience that you can trip a 60 amp breaker with short on a #14 wire.

I guess the issue would be an overload to ground that puts the maximum on the wire without tripping the breaker. You could get there if you disconnect the first heaters and re-connect them on the #14 wire and tie them to ground, but you'll need to add a bunch of heaters to make up for the lower voltage across the heater.

The likelihood of this accidentally happening is pretty remote.

Isn't this similar to the rule that allows us to run from a splitter to a fused disconnect?

wire size does limit fault current as does anything that increases resistance and impedance. in fact I have solved more than a few fault current problems by adding a meter or two of wire. Yes I have seen a job with a 24 KA avaialable fault current that would have needed a 42 KA main breaker but by making the service conductors 1 or 2 meters longer wire we got the fault current below 22 ka. This turned out to be very simple as the original fault current calculation was made with shorter wire than was actually needed.

Reno
the tree parallel is more accurate than you think for fault current too.

One of the things we try to do is ensure the fault current is high enough to trip breakers in their instantaneous range. This can require at least 60 amp fault current for a 15 amp breaker or 6 times the breaker rating.
Every wire size has a damage curve and it is very important to open the O/C device before damage occurs. Without some research i think a #14 can carry 300 amps for .01 sconds but I doubt the wire would pass that much current for even that long.