Opinions on selecting AC wire

[Syberdog]

Sorry, this is a long post- My posts tend to be lengthy- so if you want to get right to my questions, skip down to the ###############

The concrete for the ballast blocks is coming Monday so....
While the County inspector was here for the pre-pour setback inspection I took the opportunity to get his opinion on a few things.
I don't recall how it came up but he mentioned that my wire in conduit from the inverter to the utility panel was considered a wet location and the #10 THHN I was planning on using wasn't rated for wet locations in the 2011 code book he uses, but it was totally up to me if I wanted to use it or not.

So this got me reviewing the wire...

It appears that the Southwire brand THHN is dual rated for THWN wet locations, but at a reduced temperature (from 90C to 75C).
Further research shows that THWN-2 is rated to 90C in wet locations, but try to find it in #10...

Looking at various spec sheets many of them claim their wire is dual rated THHN and THWN-2 but reading the fine print shows that THWN-2 rating is ONLY on #8 and above. S
ince I need 80', up-sizing to #8 could add as much as $200+ bigger conduit to my project.

Initial calculations seemed to indicate that #10 THWN would work fine, although it could be borderline a couple days of the year when temps get up to the official range of 104F (40C), but I know we have seen higher on occasion, up to 111F.

Where I got concerned was when I did the the wire sizing ampacity calcs at this link....

http://www.thesolarplanner.com/steps_page9b.html

Skipping the array side calcs and going to the inverter AC wiring...

SolarABC data for my location shows a 2% avg. of only 37C (98F -even the .40% is 39C) but like I said I know we get hotter so I used 40C and the occasional 111F would put me into the 41C group....

So, calculating the AC wire from the inverter output ...
Inverter spec output= 21A x1.25 (ampicity multiplier) = 26.25A (nominal ampicity).
The inverter is outside,on a pole but in the summer will get direct sun.

Then I did comparison calcs using both 40C and 41C as the temp, for both 75C THWN & 90C THWN-2.
NEC table 310.16 shows temp correction factor for 75C conductors of 0.88 at 40C while at 41-45C the correction is 0.82
90C conductors have a correction factor of .91 at 40c and .87 at 41C
In conduit, THWN is rated for 35A & THWN-2 is rated at 40A and my planned breaker is 30A
With only 2 'hot' conductors in the conduit it has a factor of 1 so is omitted from the formula.

Using the simplified formula:
Inverter Derated Ampacity = Nominal Ampacity / ( Temp Correction Factors)
Derated Wire Ampicity= Rated Amps x Derating factor

I come up with:
nominial @ 40C= 26.25/.88= 29.8A
nominial @ 40C= 26.25/.91=28.8A
nominial @ 41C= 26.25/.82= 32A
nominial @ 41C= 26.25/.87= 30A
Checking the #10 wire ampicity capacity in conduit and applying the derating factors:
THWN @ 40C= 35x.88= 30.8A - wire can just handle the load
THWN-2 @40C= 40/.91=36.4A - wire easily carries the load
THWN @ 41C=35x.82= 28.7A - wire cannot handle the load- will burn before the breaker trips
THWN-2 @41C= 40/.87= 34.8A - wire will work fine

So there is a potential problem during very hot days - but isn't this offset but output inefficiency of the array as temps rise?

Searching the internet I find THWN-2 doesn't exist in #10 - have to jump to #8 to get the rating...
However XHHW-2 is also rated 90C in wet conditions and does come in #10 (and 12 and other sizes) and it is priced within a couple cents of THHN, comein colors and is overall rated superior. Most places sell it in minimum 500' rolls but I did find a couple places that sell it by the foot and shipping is about $20 for 400'

So I think I have it solved but hold off ordering to double check local supply houses.
One old timer tells me that all the solar installers he knows use #8 for the hot wires and #10 for the neutral and ground. and that he sells #8 for .31/ft - so now this info is real comparative and may even beat using XHHW-2 cost-wise, plus the #8 will definitely give me a safety factor BUT I may have to return the 3/4" conduit for larger.

Finally my questions for discussion....
################
First question I have about the wire temp derating factor is that... when it is dry out and 104-110F, even though the location is considered "wet" it is in fact dry, so wouldn't the wire perform at the 90C (THHN) rating? and then during Monsoon season when it rains every day for a month, the temps are much cooler (80-90's) it would perform at the 75C THWN rating?
If this is the case then I shouldn't have any problems as the temp factors should change with the season and conditions.

2. Which option would a better choice-
a. using XHHW-2 #10 for all 4 wires,
b. XHHW-2 for only the two Hot wires and THHN for the ground and neutral or
c. Two #8 for the hot and THHN for the other two
d. Just use common THHN/THWN for all the wires as it won't be a problem?


Thanks
Al

[sensij]

Thwn-2 absolutely does exist in #10. I bought a 500 ft spool for my DIY installation off of eBay for a fraction of the price per foot I would have paid at a big box stores, and then sold the rest of what I didn't need to someone else.

*Southwire* doesn't make #10 thwn-2, but other brands do.

[sensij]

Here is one option. There are others.

http://www.ebay.com/itm/NEW-COLONIAL-THHN-MTW-THWN-2-WIRE-10-19AWG-325FT-D573838-/382204258226?hash=item58fd2627b2:g:Ya4AAOSwhMpZnIHI

[Syberdog]

That's interesting info and good to know.
I should have said "essentially doesn't exist" or " practically doesn't exist", or something not as absolute, as I went thru a dozen or more oem and big reseller sites, I searched the net extensively and didn't come across that animal in any listing and whenever I traced a page claiming THWN-2 #10, it always ended up at some brand of THHN, and the local IEC guy said he didn't know anyone who made it....so Its good to find out it really does exist.

However, at this point and after reading the comparison specs, if I have to order wire by mail I want XHHW-2 and I want color in 100' roll or by the foot.I don't want a 500' spool to mark up myself.

My decision point is whether to use the locally available #8 for the conductors or stay with #10 and then choose between THWN or order XHHN based on the data presented. The other thing I'd like to verify is that if I go up to #8, I can still use a #10 for the green and white wires.

[sensij]

However, at this point and after reading the comparison specs, if I have to order wire by mail I want XHHW-2 and I want color in 100' roll or by the foot.I don't want a 500' spool to mark up myself.

Again, ebay is a great source.

My decision point is whether to use the locally available #8 for the conductors or stay with #10 and then choose between THWN or order XHHN based on the data presented. The other thing I'd like to verify is that if I go up to #8, I can still use a #10 for the green and white wires.

I don't think you've done your ampacity calculations correctly. I'd suggest you take another pass at 690.8(B)(1) and (2).

Size of the green wire depends a bit on how you are doing grounding. If it is serving as the DC GEC per 690.47(D)(3), it is usually sized by 250.122, assuming an ungrounded system.

Neutral wire sizing is determined by 705.95(B), which means it must also follow 250.122.

[tyab]

On your wet/dry issue - code is really clear and makes a lot of sense. All conduit outside/underground is considered a wet location. All wire inside must be wet rated and junction boxes must be wet rated and all wire connections must be wet rated. Even if you think it is dry - it can be wet due to condensation. Equipment like sub panels in a 3R enclosure are considered wet below the backplane. Listed equipment in enclosures (like the inverter) in damp/wet locations will have in their listing at what point in the enclosure the wet rating is (usually the space below the lowest bus type connection or backplane).

Above ground conduit subject to direct sun has itself additional temperature adjustments. You need to design for worst case that is likely to happen. Historical highs/lows temps should be used (lows are important if your considering using PVC above ground to compute temperature expansion.)

Running a quick voltage loss check, the #10 wire is getting close to 2% in losses. Simply from a piece of mind so the inverter has as much wiggle room on grid voltage the #8 gives you around 0.7% more room. That itself may be worth the $200 to you and removes all issues about temperature adjustments and the SouthWire sold at the big box stores is Simpul - super easy to pull. And getting it in the proper colors is nice.

The down side to moving to something like #8 is if you have to do things like splices or combining - wire nuts can be a real pain with those sizes. Doable and listed ones exist - just a pain to work with. Maybe its my old fingers but I tried with the power tool adapter and gave up and used Polaris Grey's with my #8 and #6 even thou I had listed wire nuts for those sizes. If you are doing a continuous long run right the the OCPD - then that is not an issue the OCPD's at those sizes can take #8 directly.

[Syberdog]

OK, I've gone over the tutorial sample and code sections again and am having a hard time seeing where I went astray...

-Maybe it was the higher temperature factor I used..
As I mentioned SolarABC uses 37C for my area, which would seem to allow #10 wire to work... as far as temp derating goes... but I know it gets hotter than that...

Checking local weather data for May thru Sept I only found 2 days this year where temps were 104-106F but they held that temp range for 4 hrs each day.
Looking again at table 310.16 the 40C range is actually 96-104F... revisiting weather data for June I found 13 days when the temp was 96 or higher and looking at those days I frequently found up to 7 1/2 hours straight when the temp was 96 or higher - mostly from 11:30Am until 7pm. Based on this I think the 40C (96-104F) temp correction is more appropriate for my area than the ABC data.

Even though it is outside under the north side of the array, it will be in the sun during summer until I build a shade for it, and even that may not reduce the temp into the next lower range.
The wire conduit is mostly 18" underground but it does come up into air 4' at the inverter and 5' at the PV Meter (which faces east but might get some early afternoon hi-temps). These are the worst case sections that the temp adjustments would be applicable to.

Dealing with the Inverter Output AC side...
o- 690.8(A)(3) nominal Inverter max output current (Imax)= 21A = inverter nominal ampicity
o- 690.8(B)(1a) requires OCPD greater than inverter output circuit 21A x1.25= 26.25A = 30A
o- 690.8(B)(2a) Wire sized to carry larger of 26.25A or 690.8(A) after conditions of use are applied (is this temp factor 310.16 below [29.8A to 28.8A]) or (310.15(B)(2a) [30.8A and 36.4A]) or ?

o- using Table 310.16 temp correction (TC) for above inverter output (using TC .88 for 75C wire and TC .91 for 90C wire), 26.25/TC, gives 29.8 & 28.8 Amp potential output from the inverter (although I had thought the current would go down with higher temps due to efficiency losses).

o-Using Table 310.15(B)(16) #10 75C wire is good for 35A while 90C #10 wire can handle 40A (before temp derating) -
o-Applying Table 310.15(B)(2a) temp correction factors to wire ratings... 35A x.88=30.8A and 40A x.91=36.4A.respectively.
>>> Table 310.15(B)(2b) is for 40C and gives a rating of 1 (ie: no derating) - I cant find a code section, nor do i understand, why I would use this instead of 2a?

As long as the temp stays under 105F either of these wires should work but the 75C #10 has very little room for error as it is only 1 amp above the inverter output and .8A higher than the breaker. Then there are those two days where the temp was in the 41C category fo 4 hours- Is that enough to fry the 75C wire?

If for some reason I have to go to a higher OCPD breaker (35 or 40A) only #10 rated 90C will work. If I stay with #10 wire then I need to get either THWN-2 or XHHW-2 for the 90C rating. As i mentioned, finding 100' lengths of #10 rated 90C is not as easy as it seems.

It's starting to look like the safest and easiest choice is to use #8 for the current conductors.
Table C.10 (Max # conductors in rigid PVC allows up to 5 #8 in 3/4" conduit, so I seem to be good there.
I think I can stay with #10 for the green wire and maybe the white also... don't recall the sections allowing that

What am I missing?
please be specific cuz I can be dense at times.... trees, forest and all that...


Thanks Tyab for your input - I found #8 at the local IES for .31/ft so it is going to be almost a wash. The #8 will go inverter->PVmeter->Disco->SES Breaker, so looks like 4 connection points.

[sensij]

Dealing with the Inverter Output AC side...
o- 690.8(A)(3) nominal Inverter max output current (Imax)= 21A = inverter nominal ampicity
o- 690.8(B)(1a) requires OCPD greater than inverter output circuit 21A x1.25= 26.25A = 30A
o- 690.8(B)(2a) Wire sized to carry larger of 26.25A or 690.8(A) after conditions of use are applied (is this temp factor 310.16 below [29.8A to 28.8A]) or (310.15(B)(2a) [30.8A and 36.4A]) or ?

690.8(B) isn't talking about OCPD, just conductor sizing. We'll get to breaker sizing later. Here are the two conditions:

(1) One hundred and twenty-five percent of the maximum currents calculated in 690.8(A) before the application of adjustment and correction factors.

125% of 21 = 26.25. 10 AWG THWN is rated for 35 A in table 310.15(B)(16). Apply no adjustment or correction factors here. 35 > 26.25, so OK.

(2) The maximum currents calculated in 690.8(A) after the application of adjustment and correction factors

This means that after adjustment and correction factors are applied, the conductor must be able to carry 21 A, which is the current calculated in 690.8(A).

Using the 310.15(B)(2)(a) 0.82 correction factor (and no adjustment, since you won't have more than 3 current carrying conductors) gives 35 * 0.82 = 28.7 A. This is greater than 21 A, so you pass this check too.

o- using Table 310.16 temp correction (TC) for above inverter output (using TC .88 for 75C wire and TC .91 for 90C wire), 26.25/TC, gives 29.8 & 28.8 Amp potential output from the inverter (although I had thought the current would go down with higher temps due to efficiency losses).

You don't apply temperature corrections to the inverter output. Only to the conductor ampacity in 690.8(B)(2), as just shown.

If for some reason I have to go to a higher OCPD breaker (35 or 40A) only #10 rated 90C will work. If I stay with #10 wire then I need to get either THWN-2 or XHHW-2 for the 90C rating. As i mentioned, finding 100' lengths of #10 rated 90C is not as easy as it seems.

690.9 specifies that OCPD for the inverter output circuit be determined by article 240. Using the 0.82 adjustment factor, conductor ampacity is 28.7 A. (also worth checking... the 60 deg rating for terminations is 30 A, higher ampacity than the temp corrected conductor so not a factor in this case) 240.4(B) says you can use the next higher standard overcurrent device (30 A) if three conditions are met:

(1) Circuit is not a branch circuit with more than one receptacle for cord and plug portable loads. [OK]
(2) The ampacity does not correspond with the standard amp rating... [OK]
(3) The next higher standard rating does not exceed 800 A [OK]

Also, 240.4(D)(7) says in no case shall 10 AWG be protected by greater than 30 A, so you are ok there too.

So, 10 AWG THWN w/ 30 A breaker is OK for the 41 deg C temp range you are considering.



You also asked when 310.15(B)(2)(b) applies. That would be the temp correction if the uncorrected ampacity was determined at 40 deg ambient, as in tables 310.15(B)(18)-(21)

[Syberdog]

Thank You Sensij
I will go over it again in the new light you shed so that I understand it better how the correct answer was derived and where I misunderstood.
The code language is sometimes hard to understand what they actually mean, especially without working knowledge.

It's good to know I would have been ok as is, but better to know why...

Concrete is coming this AM so I see a busy day ahead...
...
... OMG- 6 1/2 hours later I can barely move- but my brain said I did this before (so what if it was 35 years ago). Moved 4 yards today with the help of my wife running the finish trowel. Gonna sleep good tonight!