Estimating Thermal Movement of a Vessel Nozzle

[KernOily]

Hi guys. I'm trying to estimate the thermal movement of a nozzle on the bottom of a large butane sphere for use in my pipe stress work. We are having a discussion as to whether the nozzle moves up or down upon thermal heating of the sphere, such heating arising from ambient solar.

I know that thermal expansion of a body occurs in all directions relative to the center of mass. The issue here is the sphere has its support legs attached to the outside of the shell at a point about 40% or so up the diameter. My first thought is to assume the sphere is fixed at this point (the point of attachment of the legs) and therefore the nozzle moves DOWN when the sphere is heated. This is probably not 100% accurate but I believe it will give me a reasonable approximation to use for determining the nozzle movements and subsequent loadings.

What say ye?

Thanks! Pete

 

[nickelkid]

So the assumption is that the legs are nominally ambient and only the Sphere is rising in temperature? If that is the case that would be reasonable.

 

[KernOily]

No, legs and sphere would all be the same temperature since the heating is coming from ambient solar. The process runs at ambient.

 

[NBrink]

The legs will expand some as well, could very well go up.

It's going to be a very small amount though, either way. I can't imagine it being an issue.

 

[3DDave]

It's a transient problem. At night the entire structure will be pretty uniform, but come the sunshine the leg(s) on the sunny side will start expanding because it doesn't have butane to keep the temp. As the air warms the other legs will also start to expand, and eventually, if the mass of butane in the sphere warms, so will the sphere. As the sun moves through the day different legs will get different heat buildup and expand or contract. At night the sphere may slowly shrink, but the legs will certainly be shrinking faster as they are spindly and loose heat faster.

It's basically a slow motion jacking and unjacking operation, so a stress analysis taking into account that there may be a 10 degree or more difference in temp that cycles through the day should feed into a fatigue analysis.

By way of comparison, one of the operations done to large aircraft, like the B-52, involves aligning to the center line. From some acquaintances I know, the crews that did this had to do so at night. During the day first one side would warm and the fuselage would go banana shape by several inches, followed by sun movement and continuously changing deformation which meant no alignment could be verified. Hence doing this at night. I presume it wasn't the deformation that was a problem, but that it would be a constant offset on top of normal deformation in operation.

 

[itdepends]

Disclaimer- I'm a process chemist- not structural
I would expect movement due to volume of (liquid?) butane in the sphere to have a much greater impact on the nozzle vertical movement than the solar heating impact. As noted above- the sphere itself will have a relatively constant temperature due to the presence of the butane- however will change whenever the unit is taken offline and is allowed to warm.

Regardless- the downward growth due to thermal expansion when the sphere is empty should be much less than the downward deflection caused by filling the sphere.

As a chem eng/metallurgist the first part of any answer I give starts with "It Depends"

 

[BigInch]

The nozzle movement should be calculated from the fixed reference point of the base plates of the legs. The temperatures to be used to calculate the piping differential excursion temperature change should be both the coldest and the warmest ambient temperature. Any insulation would only be considered to delay reaching max or min temperatures. Excursion differential temperatures should be taken from the pipe's installation temperature. The maximum excursion temperatures will represent the outside extreme range of temperature differentials and resultant stresses that will be developed, so any lesser transient temperatures theoretically should become irrelavent. Daily solar heating temperature, or nighttime cooling calculations, if significant. Maximum pipe stress may occur when butane levels are minimal, as there would be no heat used to heat the butane and all heat would go directly towards increasing the tank and pipe temperature alone. That would allow the best chance for the system to actually reach its maximum ambient design temperature.
Minimum process temperature (as opposed to ambient) might be reached if the tank required rapid blow down and JT effects reduced temperature below ambient. In that case, take the legs at lowest ambient and the tank at min process temperature.

Personally I wouldn't bother with shading effects, unless you have some really good reason to believe that nozzle rotation, or sway, etc. from unequal leg temperatures could result in significant additional stresses. Assume all legs and tank reach the same maximum ambient design temperature all at the same time and the tank nozzle(s) moves uniformly upwards from the leg base plate elevation. For a sphere with nozzle below leg attachment point, only the height of the nozzle above the baseplate would actually be relevant as that equals the distance from base to attachement - distance to nozzle. Note that if the tank was a horizontal bullet, one end support would normally be fixed and the other designed to slide. In that case, calculate horizontal movements in a similar manner, using the fixed saddle support as the reference.

 

[BigInch]

If nozzle is not at the vertical centerline of the sphere, horizontal movements should be calculated from the vertical centerline of the sphere.

 

[KernOily]

Guys thanks for the replies!