Sealtight as a Sleeve?

mpcarnahan

Member
Location
Pittsburgh PA
So on many a commercial flat roofs, we use short lengths of rigid conduit and grounding bushings on b-line chock blocks as sleeves for PV(USE-2) wire between sub-arrays or inverters. Nothing water-tight, no heyco connector, just 'protection from damage' and such.
Can I do the same kinda thing on a residential roof with sealtight? It's a Crazy, multiple surface pitched asphalt shingle roof with multiple sub-arrays. The USE-2 doesn't need waterproofing; just protected between sub-arrays. (and supported by quickmount flashings)
Picture a 10' length of sealtight between 2 sub-arrays. With either a standard sealtight connector and a grounding bushing or an insulated throat connector w/integral lug... No junction boxes. EGC would stay continuous since ST is longer than 6'.
(not pretty... but hidden under modules anyways.) Permitted?
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
So on many a commercial flat roofs, we use short lengths of rigid conduit and grounding bushings on b-line chock blocks as sleeves for PV(USE-2) wire between sub-arrays or inverters. Nothing water-tight, no heyco connector, just 'protection from damage' and such.
Can I do the same kinda thing on a residential roof with sealtight? It's a Crazy, multiple surface pitched asphalt shingle roof with multiple sub-arrays. The USE-2 doesn't need waterproofing; just protected between sub-arrays. (and supported by quickmount flashings)
Picture a 10' length of sealtight between 2 sub-arrays. With either a standard sealtight connector and a grounding bushing or an insulated throat connector w/integral lug... No junction boxes. EGC would stay continuous since ST is longer than 6'.
(not pretty... but hidden under modules anyways.) Permitted?
If you are looking for a cheap sleeve why not just use some liquidtight non-metallic? no reason to worry about whether it is bonded or not. I am not even sure I would install a fitting if it is not being used as a raceway.

Just a thought from someone who knows nothing to speak of about PV work.
 

jaggedben

Senior Member
Maybe we're not as innovative as we should be, but we just use EMT with connectors and bonding bushings.

Keep in mind that a lot of rooftop wire sizing calculations assume a minimum 1/2" distance off the roof temperature adder for derating. If you use a flexible method that you can't keep from drooping on the roof then that could be a problem.

Also, make sure that if you use a non-metallic method that it has a temperature rating suitable for the roof. Most PVC does not which is one reason we don't use that.

I'm interested in your reference to an "insulated throat connector w/integral lug", I would love to have better hardware for bonding these chases.
 

iwire

Moderator
Staff member
Location
Massachusetts
Also, make sure that if you use a non-metallic method that it has a temperature rating suitable for the roof. Most PVC does not which is one reason we don't use that.
Can you elaborate on that? I have never heard this before.

PVC exposed outside is pretty common around here.
 

Carultch

Senior Member
Location
Massachusetts
Can you elaborate on that? I have never heard this before.

PVC exposed outside is pretty common around here.
Most standard rigid PVC has a 90C rating. LFNC by contrast, and its LFMC counterpart, have much lower temperature ratings, which means your derated cable ampacity must start with the lower of either the wire temperature or the raceway temperature.
 

iwire

Moderator
Staff member
Location
Massachusetts
Most standard rigid PVC has a 90C rating. LFNC by contrast, and its LFMC counterpart, have much lower temperature ratings, which means your derated cable ampacity must start with the lower of either the wire temperature or the raceway temperature.
I was asking about his comment about PVC not LFMC.

Most PVC does not which is one reason we don't use that.
If the roof is hotter than 90C I think the building might be on fire. ;)
 

jaggedben

Senior Member
See 352.12 (D). Most rooftops are going to see temperatures above 50C, especially if you're choosing locations ideal for solar production. It's really a different category from merely being 'exposed outside'. Table 310.15 (B)(C)(3) reflects that.

It may be that most PVC is suitable for over 50C but I've also heard stories of PVC cracking on rooftops. Note that the LMFC and LNFC sections have language about temperature limitations as well, whereas the various metallic methods do not.
 

iwire

Moderator
Staff member
Location
Massachusetts
See 352.12 (D). Most rooftops are going to see temperatures above 50C, especially if you're choosing locations ideal for solar production. It's really a different category from merely being 'exposed outside'.
I have never seen anything but 90 rated PVC, might be a regional thing.

Table 310.15 (B)(C)(3) reflects that.
It may reflect that but it does not apply to raceways and my understanding is that it is going away.

We run PVC on roofs at times, of course we are not exactly in the tropics here in NE either. :)
 

shortcircuit2

Senior Member
Location
South of Bawstin
The 90 degree rating on the PVC pertains to use of conductors within the PVC.

The National Electrical Code Section 352.10(I) reads as follows:
"(I) Insulation Temperature Limitations. Conductors or cables rated at a temperature higher than the listed temperature rating of the PVC conduit shall be permitted to be installed in PVC conduit, provided the conductors or cables are not operated at a temperature higher than the listed temperature rating of the PVC conduit."

One example of an application for this exception is the use of 105°C rated medium voltage cables, Type MV, where the cable ampacity at the 105°C rating is reduced to the cable ampacity at 75°C or 90°C and, thus, matches the listed operating temperature rating of the nonmetallic conduit (75°C or 90°C.)
 

iwire

Moderator
Staff member
Location
Massachusetts
Many rooftops in my area can reach temperatures in excess of 140 degrees F, therefore raceways with 122 degree F limitation used on rooftops would be prohibited by the listing.
Yet Carlon says their product is used successfully on FL roofs and I see it installed and inspected on roofs here in MA.

Oh well, tradition vs code. :D
 

Zee

Senior Member
Location
CA
PVC is absolutely terrible on roofs. I have experience with this back in 2001.

After all, it sounded great:
cheap
non conductive
lightweight
water proof! Perfect, right?

1. UV (sunlite) makes it brittle to the point where it can shatter from touch within a couple years. This is a known effect of UV on PVC.
2. Thermal expansion coefficient. It expands and shrinks at triple the rate of metal conduit. Triple. No matter how you attach it........ it will sag and also buckle...... and even rip out straps.
Straight runs will look like wavy lines in a small time.
Those aesthetics aren't even the biggest issue: it tore itself right out of glued LB's and connectors: leaving the THHN inside to fray and crack in the sun.....
Since then i have never used PVC on a roof, or seen it on solar.

I cannot speak to Liquidtite or sealtite. The only thing that concerns me there is it is going to be ugly. sagging. you just cannot support it often enough for it to not sag...like every couple inches.
So if it's visible IMHO it serves us well to keep solar looking good so people like what they see and want more. :)
 

iwire

Moderator
Staff member
Location
Massachusetts
1. UV (sunlite) makes it brittle to the point where it can shatter from touch within a couple years. This is a known effect of UV on PVC.
I think you are overstating it.

2. Thermal expansion coefficient. It expands and shrinks at triple the rate of metal conduit. Triple.
This is well known and why the NEC requires expansion joints

No matter how you attach it........ it will sag and also buckle...... and even rip out straps.
Straight runs will look like wavy lines in a small time.
Those aesthetics aren't even the biggest issue: it tore itself right out of glued LB's and connectors: leaving the THHN inside to fray and crack in the sun.....
If any of that is happening you are not instilling it per the code requirements.

The PVC is required to be able to slid through the supports, it also has to be supported at much closer intervals than 10'.
 
Top