When are expansion fittings required per 352.44?

[marmathsen]

Hypothetically...

If I have an outside run of PVC that runs up a wall ~10' then 90° along the wall ~40' then 90° back down the wall ~10' (10' up, 40' over, 10' down) and the temperature range (let's say in Seattle) is ~90°, am I required to have 3 expansion fittings, 1 for each of the straight runs?

Table 352.44 states PVC expands 3.65"/100ft for 90°F temp range, which means any run longer than 6.85' requires an expansion fitting! Wow! Then on top of that, if it were a South facing wall that gets direct sunlight, it's even more exaggerated.

Also, should I be determining the temp difference based on all historical data? That would be 108°F range for Seattle requiring an expansion fitting on runs longer than 5.6'!

 

[tom baker]

If possible, avoid running pvc outside on a wall, on that south facing wall it will droop and sag between supports.
EMT is better and only has to be supported every ten feet on a run

 

[infinity]

I alsways found this kind of puzzling, if your temperature range is 108° but it was 54° when you installed it could you use 1/2 expansion/contraction for your calculation?

 

[marmathsen]

I alsways found this kind of puzzling, if your temperature range is 108° but it was 54° when you installed it could you use 1/2 expansion/contraction for your calculation?

This seems only partially relevant.

Yes, you are right that if installed when the temp is in the middle of the range, then the PVC will only expand 50% and contract 50% of the length listed in Table 352.44.

But the fact of the mater is still that the temperature range is 108°F between the hottest day and the coldest day on record, thus the PVC will expand or shrink by 4.46"/100ft between the hottest and coldest day. Right?

 

[marmathsen]

If possible, avoid running pvc outside on a wall, on that south facing wall it will droop and sag between supports.
EMT is better and only has to be supported every ten feet on a run

Yes, good argument for EMT.

But if I WERE to use PVC, do you feel am I correct that I need an expansion fitting for every straight section of a run of PVC where it would expand more than 1/4"?

 

[infinity]

This seems only partially relevant.

Yes, you are right that if installed when the temp is in the middle of the range, then the PVC will only expand 50% and contract 50% of the length listed in Table 352.44.

But the fact of the mater is still that the temperature range is 108°F between the hottest day and the coldest day on record, thus the PVC will expand or shrink by 4.46"/100ft between the hottest and coldest day. Right?

I'm just thinking out loud. We calculate for the maximum expansion yet if the maximum temperature only increases by 50% of the temperature on the day of installation then the expansion will only be 50%. It will never expand to the 100% used in the calculation unless it is installed on the coldest day of the year. I know that none of this is relevant because it will still expand based on the entire temperature range.

 

[retirede]

I'm just thinking out loud. We calculate for the maximum expansion yet if the maximum temperature only increases by 50% of the temperature on the day of installation then the expansion will only be 50%. It will never expand to the 100% used in the calculation unless it is installed on the coldest day of the year. I know that none of this is relevant because it will still expand based on the entire temperature range.

The temperature at installation should be considered when figuring out how much engagement (position) of the slip coupling as installed.
For example, if the temperature was at the minimum of range, you would fully collapse the coupling, if at max temp, you would fully extend it. Most likely, you’re somewhere in the middle.
But the installation temperature does not affect the total travel required.

 

[infinity]

But the installation temperature does not affect the total travel required.

But it does affect the amount of maximum overall expansion. For example I mount two boxes with the PVC in between and lets say that the total expansion for 100° temperature change is 3/8". It exceeds 1/4" so I need an expansion fitting. But if I install that same PVC when it's 50° then it will only expand 3/16" so it can never exceed the 1/4" maximum expansion which may damage the boxes. The other 3/16" will be contraction from when it's 50 degrees lower than the air temperature of the day it was installed. Just for the record I'm not saying that we calculate it this way.

 

[tom baker]

Yes, good argument for EMT.

But if I WERE to use PVC, do you feel am I correct that I need an expansion fitting for every straight section of a run of PVC where it would expand more than 1/4"?

Maybe. If you have a horizontal run from an LB, that run turns and goes down via a 90, then that short horizontal run if it expanded would push towards the 90. Think of steam pipes outside that have expansion offsets.

 

[tom baker]

Yes, good argument for EMT.

But if I WERE to use PVC, do you feel am I correct that I need an expansion fitting for every straight section of a run of PVC where it would expand more than 1/4"?

And I find EMT is faster to install than PVC- where the PVC is outside it needs some time on installation so it looks good.
If you end up using PVC I would recommend Arlington quick latch straps. These are nice as you can install the straps first and snap the PVC into them
I have used lots of these inside where we ran pvc in corrosive areas, They are not super tight on the PVC and that will allow it to slide. Keep in mind on 1/2-1" pvc you have to support every 3 ft. that can be a lot of straps, which is why I prefer EMT outside

 

[electrofelon]

Hypothetically...

If I have an outside run of PVC that runs up a wall ~10' then 90° along the wall ~40' then 90° back down the wall ~10' (10' up, 40' over, 10' down) and the temperature range (let's say in Seattle) is ~90°, am I required to have 3 expansion fittings, 1 for each of the straight runs?

Table 352.44 states PVC expands 3.65"/100ft for 90°F temp range, which means any run longer than 6.85' requires an expansion fitting! Wow! Then on top of that, if it were a South facing wall that gets direct sunlight, it's even more exaggerated.

Also, should I be determining the temp difference based on all historical data? That would be 108°F range for Seattle requiring an expansion fitting on runs longer than 5.6'!

IMO you do not need any expansion couplings. There is no straight run between "securely mounted items".

 

[marmathsen]

But it does affect the amount of maximum overall expansion. For example I mount two boxes with the PVC in between and lets say that the total expansion for 100° temperature change is 3/8". It exceeds 1/4" so I need an expansion fitting. But if I install that same PVC when it's 50° then it will only expand 3/16" so it can never exceed the 1/4" maximum expansion which may damage the boxes. The other 3/16" will be contraction from when it's 50 degrees lower than the air temperature of the day it was installed. Just for the record I'm not saying that we calculate it this way.

But the code says to account for expansion AND contraction.

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.

 

[infinity]

IMO you do not need any expansion couplings. There is no straight run between "securely mounted items".

What about the elbows?

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.

But the code says to account for expansion AND contraction.

That's correct and that's why I said that's not the way it's calculated but in my example in post #8 you can see the flaw in that.

 

[electrofelon]

What about the elbows?

I know they list "elbows" in 352.44, but I am not sure why. Who puts a strap on elbows? IF you just put your strap 3 or so feet away it will have enough flex IMO.

 

[don_resqcapt19]

But it does affect the amount of maximum overall expansion. For example I mount two boxes with the PVC in between and lets say that the total expansion for 100° temperature change is 3/8". It exceeds 1/4" so I need an expansion fitting. But if I install that same PVC when it's 50° then it will only expand 3/16" so it can never exceed the 1/4" maximum expansion which may damage the boxes. The other 3/16" will be contraction from when it's 50 degrees lower than the air temperature of the day it was installed. Just for the record I'm not saying that we calculate it this way.

The is why the instructions for the expansion fitting tell you to set it based on the current temperature and the maximum temperature. If the current temperature is at the mid point of the range you set it at 50% of travel. If you are at the maximum temperature you set it in 3/4s of the total travel and at the low temperature you set it at 1/4 of the total travel. This is per the instructions from Carlon.

 

[wwhitney]

I know they list "elbows" in 352.44, but I am not sure why. Who puts a strap on elbows? IF you just put your strap 3 or so feet away it will have enough flex IMO.

That's obviously not true in the limit, that an arbitrarily long straight section can expand and contract when a point 3 feet past the elbow is fixed with respect to the direction of expansion and contraction.

The 3 feet from the elbow could plausibly give you some certain fixed amount of expansion/contraction takeup. E.g. if the support is allowed to rotate through a range of 1 degree, 36" * tan(1deg) = 5/8". So I can see an argument that you should be allowed to subtract 5/8" from the expansion/contraction range, and if the remainder is less than 1/4", you'd be OK. And that's for one elbow with the strap 3 ft away; in a U-configuration as in the OP, there are two elbows, so they could accommodate twice as much.

But I just picked 1 degree out of a hat, not sure what a reasonable allowance would be.

Cheers, Wayne

 

[wwhitney]

But it does affect the amount of maximum overall expansion. For example I mount two boxes with the PVC in between and lets say that the total expansion for 100° temperature change is 3/8". It exceeds 1/4" so I need an expansion fitting. But if I install that same PVC when it's 50° then it will only expand 3/16" so it can never exceed the 1/4" maximum expansion which may damage the boxes. The other 3/16" will be contraction from when it's 50 degrees lower than the air temperature of the day it was installed.

But the tension when the PVC contracts could also damage the boxes. Counting the total range as is done at least accounts for that to some extent. Admittedly, -3/16" to 3/16" deformation is going to be less stressful than 0 to 3/8" deformation.

Cheers, Wayne

 

[electrofelon]

That's obviously not true in the limit, that an arbitrarily long straight section can expand and contract when a point 3 feet past the elbow is fixed with respect to the direction of expansion and contraction.

The 3 feet from the elbow could plausibly give you some certain fixed amount of expansion/contraction takeup. E.g. if the support is allowed to rotate through a range of 1 degree, 36" * tan(1deg) = 5/8". So I can see an argument that you should be allowed to subtract 5/8" from the expansion/contraction range, and if the remainder is less than 1/4", you'd be OK. And that's for one elbow with the strap 3 ft away; in a U-configuration as in the OP, there are two elbows, so they could accommodate twice as much.

But I just picked 1 degree out of a hat, not sure what a reasonable allowance would be.

Cheers, Wayne

Wayne, Of course the 3 feet I threw out is just arbitrary and the code doesnt get that specific, but I dont see an elbow as being necessarily "securely mounted" (and the way I typically secure runs they never are). So by my reading, I dont see that an EC is required.

 

[wwhitney]

Wayne, Of course the 3 feet I threw out is just arbitrary

Is it? I thought it was related to the frequency of needing to support PVC. Obviously you need to have some support "after" the elbow on a long run. And typically that support would not allow transverse movement.

but I dont see an elbow as being necessarily "securely mounted"

But in terms of the linear expansion of the long straight run, the first support before and after the straight run (when it has elbows at each end) is going to "securely mount" the resulting U shape.

Now I agree the code language only applies to straight runs, but the physics certainly applies to a straight run plus a 3' or 5' or whatever return leg on each end, going to the next support point, to make a U shape. The linear expansion of the straight run is converted to rotational stress at the next support around the corner of the elbow.

Cheers, Wayne

 

[electrofelon]

Is it? I thought it was related to the frequency of needing to support PVC. Obviously you need to have some support "after" the elbow on a long run. And typically that support would not allow transverse movement.


But in terms of the linear expansion of the long straight run, the first support before and after the straight run (when it has elbows at each end) is going to "securely mount" the resulting U shape.

Now I agree the code language only applies to straight runs, but the physics certainly applies to a straight run plus a 3' or 5' or whatever return leg on each end, going to the next support point, to make a U shape. The linear expansion of the straight run is converted to rotational stress at the next support around the corner of the elbow.

Cheers, Wayne

I guess based on years of experience working with PVC, I feel like there would be plenty of flex if it were to be strapped 3-4 feet after the elbow. All that said, I rarely use PVC for any substantial runs above ground and hate it for that use. If I had to use it above ground, I would strap it frequently and probably not skimp on the expansion couplings, just because the stuff sucks so much.