I've got a tower with a 10" vapor line coming out the top of it and running the side of the vessel. Near the top of the vessel is a rigid support, halfway down is a guide with a +Y support, and near the bottom is another rigid support. The pipe has a small horizontal run a little ways down the pipe for flexibility. See picture.

My company's standard vessel pipe supports can support a maximum of 2500 lb each. During the hydro case, my top support is carrying 6400 lb, and the middle support is for some reason carrying none. My understanding is that the middle +y support should be carrying some of the weight during the hydro case, so I need help determining why it is not.

Pipe design temp is 275F and vessel average oper temp is 235F. Pipe normal operating temp is 225F, although this is irrelevant I suppose.
[img:left]https://www.sugarsync.com/piv/D7819491_65874299_618591[/img]

A. Upgrade the top pipe support to carry the full load of the pipe under Hydro-test condition.

B. Keep the "half-way" Guide as guide but place it half way between the top support and the lower guide's new position (see C.).

C. Change the bottom "support" to a "Guide" and move it up twenty diameters from the bottom of the vertical run.

I do not want to simply beef up the top support and hang the entire piping load from it because my company's standard support is limited to only 2500 lb. I've seen plenty of towers out there with piping expansion loops so I know this is not how it is always done.

The intent of my original post was to seek help in determining why my static results would not indicate that the middle support is carrying some of the weight.

An additional point of confusion is that when I add up the +Y reaction at all the supports, the total is roughly 150% that of the actual weight of the pipe + water during the hydro case. Also during the operating case, the sum of the +Y reactions is roughly 150% of the weight of the pipe. Maybe one or more of my cnodes are not entered correctly?

Here is a link to my CAESAR II file:
My Caesar II file

Thanks in advance for any help.

Why are you using full anchors for your supports at nodes 150 and 210? With the rigid connections you have from those supports to the vessel, I can't say I'm surprised your node 160 has essentially no load on it.

Also, does your client intend to hydro the overhead line in place? Overhead lines are typically vapor service and it is often impractical to try to design the line for a liquid full condition for a one time hydro on the tower.

Finally, though not directly related to your hydrotest question, you have a very strange temperature profile on your column. Distillation columns are typically hottest at the bottom and coolest at the top. What's causing you to start at 245°F at the bottom, drop to 235°F for the bulk of the rise, and then climb to 275°F for the top portion and overhead line?

OK, really finally this time - 2500 lbs is not a very useful max load capacity for a vessel support. What's limiting it?

Edward,
The intent for using full anchors at node 150 and 210 is to use two supports to carry the weight of the pipe instead of one. This allows us to use our standard vessel pipe support which has a max load of 2500 lb. I thought that with 3 supports and no thermal expansion, the weight of the pipe would be more or less equally divided among the three supports. I think Caesar should be able to solve a statically indeterminate case such as this.

Our client does intend to hydro the overhead line in place. This line is in vapor service. This is a 10" line, so yes it could be impractical to hydro in place. I could make the call to consider it impractical and decide to have it pneumatically tested.

This column is an aqua ammonia stripper, and apparently it doesn't follow the typical rules of a hydrocarbon tower. This profile is an approximation given to me by our process engineer. Flow goes from the top of this tower to a condenser.

2500 lb is the limit stated on our standard support drawing. These supports are pre-designed, so I don't know which piece is limiting it. I was hoping to not have to design a one-off support for this project, although this would be possible. Support drawing is attached.

I found this in the KLM technology standards, indicating that a 10" overhead line would normally be hydrotested:

4. If for any reason it is not practical to carry out a hydraulic test, a pneumatic or partially pneumatic test may be substituted subject to prior agreement with the Company. Full details, including proposed safety precautions, will berequired. The following are usually excluded from hydrostatic testing, and are usually tested with compressed air and soap suds:
a. Instrument air lines (test with dry air only, if possible).
b. Air lines to air-operated valves (test with dry air only).
c. Very large (usually over DN 600 or 24 inches) gas or steam overhead lines.
d. Pressure parts of instruments in gas or vapor service.

Interesting about the temperature profile. I guess there must be some kind of reaction going up up at the top.

The reason you aren't getting any load for the hydro (or SUS for that matter) at support 160 is that it is becoming disengaged from the CNODE point 162. In both cases, the downward movement of node 162 is slightly more than the downward movement of node 160. It is not readily apparent to me what it causing this.

I suspect using two anchors on the overhead line is a factor. It is in practice difficult to get away with such an arrangement as the anchors will end up fighting each other due to the difference in temperature between the overhead line and the column itself. Your support down at 150 is picking up 3000 lbs going from SUS to OPE, 1500 of which comes from it unloading the support at node 210 and another 900 coming from lifting off the off column support at node 100.

If you are going to be limited to 2500 lbs per bracket, you are going to need to size springs at the other support points and manually size them for the loads you want them to carry. (We typically call those "helper springs") The hydrotest case makes that a complicated proposition, since the line is large enough that the liquid weight is going to about double the weight of the whole piping system.

Now, as to why Caesar isn't giving you a distribution on all three supports for the hydro case - in part, it's due to that weird difference in deflection you are getting between nodes 160 and 162. The other part though is that you have your two hard supports at 150 and 160 directly in line with one another and I expect that is giving Caesar some trouble resolving the load distribution.

Although I still don't understand why support at 160 is disengaging from its CNODE, I think it is time to move on. I will be adding a few larger standard supports to our company's library. I agree with you that in this case it would be best to hang the entire pipe's weight from the top support and change the lower supports to guides, and eliminate the horizontal run of pipe halfway down the column. I spoke with the vessel vendor today and he said this is what he was expecting, and he thought it was strange that I wanted to distribute the load to several supports.

Thanks for pointing out that more than one set of helper springs in the same vertical run would need to be manually sized. This might save some frustration on a future project. Although I think that during the hydrotest the springs could remain locked, right?

Thanks for helping with this one.

Trevor:
You do NOT need Springs at all if you support and guide this pipe drop as I suggested.

Jop:
I will be following your suggestion. My question about springs was only to help my general understanding.

Trevor,

For the spring part: Springs are normally kept locked (i.e rigid) with lock nuts in place during hydro test.

It may happen that due to system imbalance, spring may see more load during hydrotest than generally allowed (generally its twice the rated load of the spring). If such a situation occurs , you may go with either of the understated options:

A)Put a note on Isos stating "replace spring with temporary rigid support during hydrotest".

B)Ask the spring vendor to provide specially designed "hydrostatic test stops" which can carry the excessive load.

Hope this helps.

& I guess you must be happy by now by putting your stress calc to Rest in Peace!!!

Trevor,

Sorry for going back to your first question. The reason for support at node 160 not getting any load in cold condition (Sus and Hydro) is Flexibility of tall column. Since you have modeled column as pipe , due to most of pipe acting weight acts towards – Z direction , it is causing top portion of column to sway in –Z direction by around 6mm . This is causing node 162 to disengage with node 160 as node 150 is again a anchor. If you model column as rigid element, support at node 160 take fairly good load in both sustained as well as hydro condition.
This is just a reason for support at node 160 not taking load , your support arrangement needs to be changed as advised by others.

-------------------
Khalid

Khalidmf:
I still think it's interesting the the CNODE is disengaging, but at least we have found the source of the problem. Next time I will try modeling the vessel as a rigid if I have a weird arrangement like this again.

And now we can put this one to rest. Thanks for your help all four of you.

Here are the files, in case I remove them from the other hosting service, and people want to reference them in the future.