Manual Spring Can design

[RBMECH123]

I am practicing how to design a spring can with Caesar II. I know that I can ask the program to select a can using the free code, but for some reason it is forcing me to use a "rod hanger". Well since I am using an example from an old project, I find that there is something wrong. The input is identical and the example has the spring can for that location with a hot,cold, and spring rate. Can anyone help me step by step how to model a can, meaning do I put a rigid support, then get a load from the output, then look at the load tables from PTP? I am confused, please help.

Thanks in Advance!

 

[RBMECH123]

In other words, how can I properly determine the hot load and movement?

 

[ApC2Kp]

RBMECH123,

The manual selection of a spring support requires two itms of information - the load to be carried and the expected thermal movement. There are other data like allowable force variation of the spring, minimum movement for selection, available space to install spring support, and the working range of force and travel of a particular spring unit - but the load and movement are primary.

You could obtain the support load value from a weight analysis of the piping or approximate value from adding up the weights of pipe, fittings, flanges, valves, and fluid content. The thermal movement could be approximated from calculation of thermal expansion of piping and equipment connections.

The manual selection of spring unit for the load and thermal movement will be a lookup of values in the spring manufacturer's table. One commonly used table is the Anvil International Co. (successor to Grinnell Corp.)

http://www.anvilintl.com

download their 'Pipe Hanger Catalog PH-2006' and the table on pages 134-135 shows the loads and movements that their spring supports will provide, as well as more instructions.

The CaesarII spring selection parameters include a threshold minimum travel, 'Rigid Support Displacement Criteria' which can be set different from the default value of 0.1 inch, for which the spring selection reverts to 'Rigid'. Why use a spring support when the piping does not move? So it will revert to a rigid suppport.

Another spring selection outcome is 'Constant spring' if the allocated space is less than the dimensions of a selected spring hanger or spring can. Also the selection outcome of 'Constant spring' is result of the force variation (travel X spring rate) being greater than the specified allowable variation, usually 25% by default but could be reset to allow more variation or less variation to control the maximum / minimum forces from a spring unit.

Now you should appreciate using CaesarII to let the software select the spring. Instead of clicking the box for 'Restraint', you click the box for 'Hanger', enter the node number to locate and identify the spring hanger, and enter any other information. Set up the analysis load cases to include the spring hanger loads (CaesarII is able to construct and recommend the load cases for you), and run the analysis. Done.

If you want to input your manual selection values for a spring, then you provide 'Predefined Hanger Data' in the input fields at bottom after entering the hanger node number. The input of spring rate is required along with either the cold installation load or constant effort support load.

 

[RBMECH123]

Do I use a Y support or let it hang and run the analysis to take the load of the pipe from there (prior to selecting a spring can)?

 

[RBMECH123]

Its easy to choose a hanger, (two in my case) and bam! I pass with only 54% allowable, but I know its important to stay within 25% variation. I am using the PTP Spring Selection tool, but I need to know if I have the correct hot load. I have read everything from the internet.

 

[richay]

Do not put a (vertical) restraint at the hanger node, just indicate that there is a hanger there.

Richard Ay
COADE, Inc.

 

[ApC2Kp]

RBMECH123,

In a perfect world some might say the hot load would be the same as cold installed load. A constant effort spring is almost perfect with a +/- 5% variation of force over its travel range 10"+, but you don't want to use or pay for them unless necessary. The typical 25% variation limit is usually accommodated by the Fig B-268 or the Fig 98 double stack spring models. For small movements less than 0.50" a half height spring like the Fig 82 is all that is required.

The requirements for 'correct' hot load for selected springs could be to keep the pipe stress within allowables, but more often the spring supports are used to keep equipment connection loads within limits. Again, ideally the piping would put zero loads on equipment connections, but thermal growth of equipment and piping will result in weight loads being shifted between rigid supports and the equipment. That is one of the reasons for the 'Free Y' at the equipment connection's restraint conditiion of 'Anchor' or the displacements of its thermal growth. The 'Free Y' of restraint has the spring support take all the weight of piping - no weight is supported by the equipment connection, at least for the sizing of the spring force.

 

[RBMECH123]

Thank you ApC2Kp,
The pipe is 20" and it coming out of a tank, which has a valve connecting to the nozzle. The pipe turns and then loops around before entering the ground. I have two hangers, and well, the program sizes one can farthest from the tank and other is changed to a rigid support. This is the problem that im faced with. What I did was placed a vertical support on both hanger nodes ( in know richay said not to do this but my coworker suggested so). I then ran an anlysis, took the loads from the hot and cold conditions. After that, I used the PTP spring hanger tool and entered the cold and hot loads. It gave me the spring rate and I entered the values in the program as a spring hanger. I ran the analysis again and I passed. I did see that the nozzle had a high moment, not that bad though. I resized the inner can at a different location and it did lower the nozzle load. This thread was to ask really, am I doing this wrong??
Thanks again for the help!

 

[richay]

What I meant was "don't put a vertical restraint in and expect CAESAR II to design a hanger there". Your manual step by step approach (put in a rigid support, determine the load, etc) is what CAESAR II does in its hanger design load cases.

Richard Ay
COADE, Inc.

 

[ApC2Kp]

RBMECH123,

The loads on tank nozzles are usually limited, due to thin design of the wall thickness. Thermal expansion or growth of a tank would normally be less than say the growth of a boiler or distillaion column. An important consideration for tanks is the soil properties and how much settlement is expected when the tank is filled. The piping support next to the tank nozzle / valve needs to allow for that settlement. Typical settlement values could be more than 1 inch - that would change the program selection from a rigid support to spring support to carry the Nps-20 pipe + valve.

Some discussions have labeled the tank settling as similar to a thermal displacement. It could be considered as an expansion type stress, instead of a sustained stress. Did the tank nozzle have a displacement defined for analysis? Did the hanger selection allow for 'Free Y' of the tank connection? These are important for the loads on tank.