PVC underground thermal expansion

[Carultch]

PVC expands so much, that just a single 10 ft piece expands more than 1/4" in typical outdoor ambient temperature ranges.

When used underground, I've rarely seen expansion fittings installed in the long runs below grade. And I'd expect sand to clog the piston, rendering it ineffective. A handhole may allow some thermal expansion accommodation, but other than that, I don't see expansion accommodation installed underground in practice.

I'm aware that the temperature swing is typically a lot less underground, maybe 10F instead of 100F. However, if 10 ft requires expansion accomodation above ground, then you would expect that 100 ft would require expansion below ground.

So my question is, how do you (if at all) accommodate thermal expansion in underground PVC?

 

[gadfly56]

PVC expands so much, that just a single 10 ft piece expands more than 1/4" in typical outdoor ambient temperature ranges.

When used underground, I've rarely seen expansion fittings installed in the long runs below grade. And I'd expect sand to clog the piston, rendering it ineffective. A handhole may allow some thermal expansion accommodation, but other than that, I don't see expansion accommodation installed underground in practice.

I'm aware that the temperature swing is typically a lot less underground, maybe 10F instead of 100F. However, if 10 ft requires expansion accomodation above ground, then you would expect that 100 ft would require expansion below ground.

So my question is, how do you (if at all) accommodate thermal expansion in underground PVC?

Even with the relatively high coefficient I'm guessing you wouldn't see much movement. Don't forget that the surrounding soil has a pretty fierce grip on the pipe. It may try to expand and simply build up stress in the pipe wall. The vertical legs are another issue entirely. Everyone knows how destructive ground heave can be, whether it's from frost, settling, or clay soil getting wet and drying out.

 

[Carultch]

Even with the relatively high coefficient I'm guessing you wouldn't see much movement. Don't forget that the surrounding soil has a pretty fierce grip on the pipe. It may try to expand and simply build up stress in the pipe wall. The vertical legs are another issue entirely. Everyone knows how destructive ground heave can be, whether it's from frost, settling, or clay soil getting wet and drying out.

I'm aware of how to accommodate the vertical frost heave movement with expansion fittings in the stub-ups. It's the horizontal movement for which these don't account.

 

[wwhitney]

I'm aware that the temperature swing is typically a lot less underground, maybe 10F instead of 100F. However, if 10 ft requires expansion accomodation above ground, then you would expect that 100 ft would require expansion below ground.

There are two strategies to dealing with thermal/expansion contraction: let the material move as it wants to and accommodate that movement, or constrain it well enough to handle the stresses induced by the thermal strain, so the material doesn't move.

Below grade, solid PVC conduit can use the continuous constraint method, via friction with the surrounding soil. Similarly, in a PVC coated steel conduit, the PVC is continuously constrained by the bond to the steel.

Above grade solid PVC conduit usually uses the movement accommodation strategy: the straps allow longitudinal movement, and the there are expansion fittings to take up the accumulated deformation. One could fully constrain PVC conduit above grade so it doesn't expand/contract thermally by using enough tight clamps, but it would take a lot of clamps. An engineer could calculate it, maybe every few inches would do it.

Cheers, Wayne

 

[iwire]

I don't think anyone worries about this underground.

I also am curious how much temp swing you expect? Soil temps are pretty steady.

 

[jumper]

One could fully constrain PVC conduit above grade so it doesn't expand/contract thermally by using enough tight clamps, but it would take a lot of clamps. An engineer could calculate it, maybe every few inches would do it.

Cheers, Wayne

Technically not legal.

352.30 Securing and Supporting.
PVC conduit shall be installed as a
complete system as provided in 300.18 and shall
be fastened so that movement from thermal expansion or con-
traction is permitted. PVC conduit shall be securely fastened
and supported in accordance with 352.30(A) and (B)

 

[jumper]

I don't think anyone worries about this underground.

:thumbsup:

My 85 year old great aunt in Ogden Utah might, I don't.

 

[wwhitney]

I also am curious how much temp swing you expect? Soil temps are pretty steady.

In New England, at a depth of 18 to 24 inches, it looks like the temperature swing could be 20 to 30 degrees F. At least, that is my quick take-away from glancing at the graphs at this site:

http://nesoil.com/climate/

Cheers, Wayne

 

[wwhitney]

Technically not legal.

Oh, I agree the NEC requires using the movement accommodation method. I was just trying to say that the reason the NEC takes that approach is that the frequency of clamps required to sufficiently constrain the PVC would be way too high.

Cheers, Wayne

 

[iwire]

In New England, at a depth of 18 to 24 inches, it looks like the temperature swing could be 20 to 30 degrees F. At least, that is my quick take-away from glancing at the graphs at this site:

http://nesoil.com/climate/

Cheers, Wayne

I would not argue with that, a darn site less than the easy 110F+ change we can have in ambient.

Take that lesser movement per unit of length with your 'continuous constraint method' and we end up with a perfectly functioning raceway.

 

[jumper]

In New England, at a depth of 18 to 24 inches, it looks like the temperature swing could be 20 to 30 degrees F. At least, that is my quick take-away from glancing at the graphs at this site:

http://nesoil.com/climate/

Cheers, Wayne

For long runs that could be an issue, 1" over 100', but for short runs between boxes and such, not really.

352.44 Expansion Fittings.
Expansion fittings for PVC conduit shall be provided to compensate
for thermal expansion and contraction where the length change, in accordance
with Table 352.44, is expected to be 6 mm (1⁄4 in.) or greater in
a straight run between securely mounted items such as boxes,
cabinets, elbows, or other conduit terminations

 

[jumper]

Oh, I agree the NEC requires using the movement accommodation method. I was just trying to say that the reason the NEC takes that approach is that the frequency of clamps required to sufficiently constrain the PVC would be way too high.

Cheers, Wayne

Fair enough.

 

[Carultch]

I would not argue with that, a darn site less than the easy 110F+ change we can have in ambient.

Take that lesser movement per unit of length with your 'continuous constraint method' and we end up with a perfectly functioning raceway.

Right. I'm just wondering how we still don't have to think about thermal expansion, even when the run is as long as 100 ft, 200 ft, or 300ft. Just as I mentioned, Wayne's data source confirms my estimate that it is a 10F temperature swing.

If 7 ft or greater is the expansion limit for above ground applications, then I would expect that 70 ft or greater would be the expansion limit for below grade applications. And I've rarely seen a handhole placed every 70 ft, or a horizontal expansion piston underground.

 

[wwhitney]

Right. I'm just wondering how we still don't have to think about thermal expansion, even when the run is as long as 100 ft, 200 ft, or 300ft.

The reason it works is that the soil continuously constrains the PVC. If you ran your underground PVC conduit through a larger underground chase, then it wouldn't be constrained and you should provide expansion fittings to accommodate the change in length.

Of course, 352.44 as jumper quoted doesn't exempt underground PVC runs from the requirement for expansion fittings. Seems like an oversight in the NEC.

Just as I mentioned, Wayne's data source confirms my estimate that it is a 10F temperature swing.

Not sure which graphs you were looking at, the ones I checked were showing 20F or 10C temperature swings, and I think one of the later ones shows a 30F swing.

Cheers, Wayne

 

[Carultch]

The reason it works is that the soil continuously constrains the PVC. If you ran your underground PVC conduit through a larger underground chase, then it wouldn't be constrained and you should provide expansion fittings to accommodate the change in length.

Of course, 352.44 as jumper quoted doesn't exempt underground PVC runs from the requirement for expansion fittings. Seems like an oversight in the NEC.


Not sure which graphs you were looking at, the ones I checked were showing 20F or 10C temperature swings, and I think one of the later ones shows a 30F swing.

Cheers, Wayne

Ok, so you are saying the failure mode the rule is trying to prevent is buckling, and direct tension/compression isn't the issue. And the continuous support from the soil or sand, is what stops the buckling.


I misread what you said about temperature swing data. I thought you said the temperature underground ranged from 20C to 30C.

 

[RICK NAPIER]

I don't think PVC can be sufficiently restrained. When I was young I had repair at the beach where sodium fixtures were mounted on a white brick wall about 20' up. These fixtures were about 40' apart with 3/4" PVC between the fixtures. The fixtures where secured with four 1/4-20 lead anchors. The PVC was strapped with steel straps with 1/4-20 lead anchors every 2'. The installation was about 6 or 7 years old when we were contracted to make repairs. There were fixtures pulled off the wall, fittings pulled open and where the PVC couldn't move to well a horizontal shear. We removed half the straps, put in straps that allowed movement and a lot of expansion fittings. This worked.

 

[gadfly56]

I don't think PVC can be sufficiently restrained. When I was young I had repair at the beach where sodium fixtures were mounted on a white brick wall about 20' up. These fixtures were about 40' apart with 3/4" PVC between the fixtures. The fixtures where secured with four 1/4-20 lead anchors. The PVC was strapped with steel straps with 1/4-20 lead anchors every 2'. The installation was about 6 or 7 years old when we were contracted to make repairs. There were fixtures pulled off the wall, fittings pulled open and where the PVC couldn't move to well a horizontal shear. We removed half the straps, put in straps that allowed movement and a lot of expansion fittings. This worked.

But that's in free air. Buried in the ground is another story, no?

 

[Fulthrotl]

Right. I'm just wondering how we still don't have to think about thermal expansion, even when the run is as long as 100 ft, 200 ft, or 300ft. Just as I mentioned, Wayne's data source confirms my estimate that it is a 10F temperature swing.

If 7 ft or greater is the expansion limit for above ground applications, then I would expect that 70 ft or greater would be the expansion limit for below grade applications. And I've rarely seen a handhole placed every 70 ft, or a horizontal expansion piston underground.

3' down, temperature swings aren't very great, or very fast.

it's more important how you install it. most folks glue it up and
throw it in the ditch. where you have problems is gluing it up in
the sun in july, throwing it nice and hot into a ditch 4' deep, and
pouring nice cold concrete right over it. 500' of pipe can easily
separate in more that one place from the temp. differential.

glue it up, let it dry a while so nothing pulls apart, kick it into the
ditch, leaving the end up above ground, with a pipe to hold it up,
out of the ditch. let it sit there an hour. then cut the end, glue a
90 onto it, and stub it up. this lets the pipe match the ground temp
in the bottom of the ditch.

backfill it with sand, or slurry, as you like. it'll never pull apart, ever.

this applies to large feeders. small crap in the slab, 1" and under,
just glue and go.