I have some literature that states that when selecting a spring hanger in the table that if all else fails you can divide the cold load by two and use that value to find an acceptable spring hanger (Hot load being DW/2+delta*k). Although you will have to use two spring hangers instead of just one. Does that mean it would be two springs in parallel? And would that be something like an Anvil Type G hanger?

The literature is a little vague after it's first statement that you can divide it by two, and I've been trying to find another source but to no avail.

Any insight on this would be appreciated.
Thanks

Hi,

If you want to use two springs in parallel, then you will have to divide by 2 the cold load (setting load). You will need to divide by two the spring rate as well, because the system displacement remains the same.

If you have one spring with a rate = 100 N/mm and the system displacement expected = -10 mm your spring is compressing and the resulting load is increased +1000 N.

Now, if you want an equivalent system using two springs instead of one, you will need to install two springs with the half cold load and also with a rate equal to 50 N/mm. The result will be the same since your system displacement is still = -10 mm.
So. 50N/mm * 10 mm = 500 N, this times two (springs) = 1000 N. Hope this will answer your concern

Thanks Santiago,

With the cold load (3875lb) and deflection this hanger is seeing (5.75 in), we've selected a triple spring, size 15. I'm not comfortable with this because the load variation is 53%.

But if I divide the cold load by two and then look up a suitable spring I end up selecting a Fig 98, Type G, Size 13 hanger. The literature I mentioned before did not really go into an example using this method, it only suggested it. I only assumed that once you've divided the cold load by two that you continue with the selection process the same way you would if you were only using one spring. Is this correct? And do you calculate the load variation per spring or by using the effective spring constant?

My boss isn't comfortable with this method and I'm trying to determine if it's a valid method or not.

Thanks


Ps. He doesn't want to use a constant effort hanger due to their size in the horizontal plane. We have lots of room vertically.

Pss. I'm not looking for an "OK" on the hangers mentioned above. I'm only looking for any more literature on this method or advice.

Anybody?

Why don't you tell CAESAR II to put 2 springs there, and let CAESAR II design them for you?

Matt,

Richard is right. And if your system still fails adjust your load or spring rate to balance your pipes. Then check your load variation again.
Goodluck!

You imply that the two new spring should have half the stiffness. But stiffness is what the designer is selecting. The catalog will offer different stiffnesses that can carry half the load with the same amount of travel. This lower (effective) stiffness willl reduce the load variation - the problem here.

I agree with you, I was trying to explain the analogy between one and two springs system, but you are right, the problem here is to reduce the variation and that can be reached only by reducing the rate or by using a constant hanger.

This is really a simple algebra problem. If you want to deal with 3875# cold load and 5.75" movement, and limit the load variation to 25%, you need to have a maximum spring rate of 168# per inch. If you use two Fig. 11 quad springs, their total spring rate is 170# per inch. If you are fussy, this is not an adequate solution. If you are not so fussy, this will be OK.

You might be able to special order a variable spring hanger from the usual vendors where you can satisfy the letter of the rule of thumb.

Other avenues include altering the overall support scheme for your piping to reduce the design load at this point. For instance, if you are trying to support your system a certain horizontal distance from the bottom of a riser, you may find that putting a spring support with a riserr clamp some distance up tfrom the bottom of the riser will carry a lot of load with reduced thermal movements.

You may also be able to move farther from the riser for your support of the horizontal run and pick up load in the riser in other ways.

Finally, if you have this kind of thermal growth, do you have any drainage issues with your process fluid that may require you to slope the horizontal run?

Thanks everyone, this issue has been resolved. We got CAESAR II to design them as variable springs and not constant. My question as to whether or not that methadology is correct was answered by Craig. Thanks Craig, I was putting way too much thought into all of it only because my boss wasn't comfortable with that method. Whether or not he thought it was wrong I don't know. But it made me second guess myself again and again. I just needed to here it from someone that it is a valid method.

Thanks to all that replied.