Clevis hanger support with rod 1

[StrLamp]

Hi
We are designing the pipe support system for gravity and seismic loads, while another party is responsible for the thermal load analysis. Our standard approach is when a pipe is supported from a concrete ceiling or frame using a clevis hanger and threaded rod, and unless it is braced, we consider it a gravity-only support.

However, the thermal analysis engineer recommends using a minimum rod length of 24 inches, which is not always possible due to pipe alignment and the location of existing structures Otherwise, he treats the support as rigid and applies thermal loads accordingly. In such cases, we would then need to restrain them. I believe he is using CAESAR II for the thermal stress analysis.


Sometimes, we install a small strut frame, and the pipe hangs from a clevis hanger connected to it. Even then, if the rod length is less than 24 inches, he assumes it’s rigid and includes thermal forces. For seismic design, though, we treat these supports as not taking any seismic loads because dedicated seismic restraints are installed at regular spacing.


In my view, the clevis hanger itself provides rotational flexibility, and the rod, even if shorter, still allows some lateral movement. Therefore, these supports should not be treated as rigid unless specifically braced. I’m also aware that shorter rods are stiffer and can attract unintended lateral loads. However, we have seismic restraints installed at regular intervals to address that.


What do you think? Have you had any experience with similar situations?


Thank you.

 

[LittleInch]

To avoid doubt can you post a picture or drawing of this hanger.

Also how is the hanger attached to the pipe?

If you really want to make the hanger effectively friction less you might need to say it is a roller hanger.

Like this.

 

[StrLamp]

To avoid doubt can you post a picture or drawing of this hanger.

Also how is the hanger attached to the pipe?

If you really want to make the hanger effectively friction less you might need to say it is a roller hanger.

Like this.

Hi Little, Thank you for the reply. Please see the attached file.
It has about 3" of magnesia insulation, and the pipe will have a saddle plate at the bottom.

 

[LittleInch]

I can see why the stress engineer thinks that way and with the insualtion, that's a pretty stiff support.

A lot will depend on the amount of expansion being forecast and what in reality is the maximum "swing" or flex you get from that support with minimal force.

Your only way around it as far as I can see that is easy is a roller support and possibly a bottom half shell or third of a shell to avoid digging into the insulation with the point load from the roller.

I would do the analysis assuming rigid and then looking at what happens if you made them simple sliding supports in terms of movement.

Reality will be somewhere in between, but only you can decide whether the claimed expansion on the worst support can be accommodated or not.

 

[RVAmeche]

I think there's a terminology issue here. Non-spring supports in Caesar II are called rigids (clevis, rollers, shoes, etc all are rigids) while spring supports are hangers. A rigid +Y representing a clevis will still move laterally (typically check for 4 deg rod angle if its tight) and show lift off if the piping movement does so. Rigid in this context doesn't mean an unmoving anchor.

 

[Snickster]

I think that if I remember correctly a hanger is considered to be able to swing a certain amount of degrees before it locks. I have an old Grinell manual hard copy that goes into a discussion on this situation. I need to look for it and make a copy and will post later.

 

[Snickster]

I found a digital copy of the Grinnell manual on their website attached. It discusses the design of hanger systems including how to handle horizontal movements in supported piping.

 

[StrLamp]

I can see why the stress engineer thinks that way and with the insualtion, that's a pretty stiff support.

A lot will depend on the amount of expansion being forecast and what in reality is the maximum "swing" or flex you get from that support with minimal force.

Your only way around it as far as I can see that is easy is a roller support and possibly a bottom half shell or third of a shell to avoid digging into the insulation with the point load from the roller.

I would do the analysis assuming rigid and then looking at what happens if you made them simple sliding supports in terms of movement.

Reality will be somewhere in between, but only you can decide whether the claimed expansion on the worst support can be accommodated or not.

Thank you for the input. They provided us with the thermal loads in both the lateral and longitudinal directions, even for the trapeze supports. I believe the trapeze only resists load through the friction between the pipe and the strut. However, I don’t think the friction is very high between the pipe insulation and the strut. Once the frictional resistance is exceeded, the pipe begins to move and so does the support. That is my understanding. Sorry, thermal stress is not my area of expertise, which is why I’m having some difficulty visualizing it.

 

[StrLamp]

I think there's a terminology issue here. Non-spring supports in Caesar II are called rigids (clevis, rollers, shoes, etc all are rigids) while spring supports are hangers. A rigid +Y representing a clevis will still move laterally (typically check for 4 deg rod angle if its tight) and show lift off if the piping movement does so. Rigid in this context doesn't mean an unmoving anchor.

Thank you, RVA.
I think the same. When we say "rigidly connected" structurally, we understand it to mean that it does not allow any rotation.
Do you mean that if the hanger can deflect/swing by 4 degrees from the point of connection, then you consider that support flexible and not engaged in resisting thermal loads?

 

[StrLamp]

I found a digital copy of the Grinnell manual on their website attached. It discusses the design of hanger systems including how to handle horizontal movements in supported piping.

Thank you for the attachment. I will take a look. Thank you

 

[LittleInch]

It's just the way you've shown it on the drawing it looks like the pipe is almost clamped into the hanger.

How the pipe actually responds to axial or lateral movement or forces is not easy to see given it is so tight into the hanger and could easily lock up to me resulting in large forces going into the support.

Hence why I suggested a roller hanger to let the pipe move with something like friction factor 0.1.