On Sat, 02 Aug 2014 10:36:24 -0400, wrote:

On Sat, 2 Aug 2014 05:51:37 -0700 (PDT), trader_4
wrote:

On Saturday, August 2, 2014 12:40:28 AM UTC-4, wrote:
On Fri, 01 Aug 2014 23:34:13 -0400, wrote:



On Fri, 01 Aug 2014 00:32:34 -0400, wrote:



On Thu, 31 Jul 2014 20:25:57 -0700 (PDT), wrote:



SMZ42H46ZOGX



When I look this up I do see where the manual says you need 6 ga wire

but it also says "Minimum circuit ampacity 26a" which would be 10

gauge copper according to the NEC. It really sounds like the chinese

book writer does not understand the electrical code.



Does the label on the outside unit say "26a" on the minimum circuit

ampacity line?

Maximum over current protection probably says "50a"

26 amps is more than 80% of 30 amps. Generally a circuit should not

excede 80% of the circuit protection rating (fuse or breaker - and I

assume wiring).



That is not what the label says or what it means.

If you look at the U/L marking guide, you will see the 80% is built

into that "minimum circuit ampacity"

+1

This is a specific piece of eqpt on a dedicated circuit. It's not
a circuit for receptacles. The rules are different. Just like a
previous poster thinks you can't have a 50A breaker on 8g wiring for
an AC, when you can, because the rules are different.


Yup, you set a guy's hair on fire if his knowledge stops at the Time
Life "electric for dummies" book. There are places where you can have
12 gauge wire on a 50 and 14 on a 40. It is a common question on the
inspector tests.
(typical was, 1HP 120v, 16 FLA motor on a 40a breaker with 14ga
copper)



I bet the design FLA is more like 19-20a and the actual max you would

ever read with an amp probe is more like 15-16 in normal operation

They say strange numbers like "26a" to keep you out of 12ga which is

rated for 25a in the 60 and 75c column.

The marking guide and the label itself is clearly saying 10 ga copper.



Also, if I remember and read correctly, the

specification is for a "stranded" wire - which "generally" should be

sized up 1 size.



Cite that.

Table 310-16 does not make that distinction

Yes, I'd like to see that too. AFAIK, solid and stranded of
the same gauge have the same current carrying capacity, unless
this AC is connected to a 1Mhz power source, or similar where
skin effect comes into play.

Yup, the stranded is actually better if you start getting up in the
higher frequencies. You can see the effect at 400hz which used to be
pretty common in computer rooms. (back when they actually had real
computers in there) ;-)
The big 400hz cables were usually fine stranded.

If you look at table 8 you will see that the DC resistance of stranded
#10 is actually 0.03 ohms more than solid per 1000 feet but that is
insignificant for this calculation and the difference is a little less
at 60hz

Skin effect doesn't even begin to have any effect in residential or
normal commercial wiring. I know code does not require upsizing for
stranded wire - I never said it did. I said, generally one should use
one size heavier flexible cord than one would use for solid permanent
wiring. I stand by that. Flexible cable is submitted to a lot of
flexing which will eventually compromize some strands, increasing the
resistance. It is a good idea to use one size heavier cable than
"required", particularly when running close to the design current
limit for the cord.
In general practice, MOST people use a cord that is AT LEAST one size
too SMALL for the load.