Pressure Increase Due To Temperature Increase 10

[lumpyrugger]

Anyone know how to calculate the pressure increase due to temperature increase for a liquid in a closed, sealed system? I have the formula: CoTE x temperature increase x copmpressibility, but I can't get the compressibility factor for the liquid I'm using.

 

[btrueblood]

Well, what fluid are you using? Maybe somebody else has a number you could use.

Also, to be complete you should make sure you account for the expansion (both thermal expansion and pressure-induced expansion) of whatever the walls of your system are made of.

 

[dcasto]

hate use this but, just try a little of this

http://en.wikipedia.org/wiki/Coefficient_of_thermal_expansion

 

[zdas04]

For water the number conveniently works out to about 100 psi per degree F. Other fluids work out to other numbers.

David

 

[lumpyrugger]

The fluid is Shell Thermia Oil C, which is a heat transfer mineral oil.

 

[zerosum]

I would contact the oil supplier, they should have that data.

 

[lumpyrugger]

Already tried...The Chemists at Shell have no clue. Thanks anyway.

 

[zerosum]

Well doesn't that suck.

 

[zerosum]

You trying to do calcs for thermal relieving?

 

[iainuts]

Does the container have no air or other gas in it at all? Does the oil completely fill the container with no voids?

What do you know about the initial conditions? Pressure? Temp? Volume?

What do you know about the final conditions? Heat flux? Total heat input? Final temperature?

Maybe you could also give some description of how this container is used, and why you're interested in determining the pressure increase. Is this to size a thermal relief valve? Are you concerned with overpressurizing a sealed container? Tell us what you can about the situation.

 

[lumpyrugger]

I have not done any thermal relieving calculations.

I did some calcs based on the expansion of the oil to determine how much to fill the vessel to allow it to expand to fill the void.

The initial conditions are ambient or atmospheric pressure and ambient temperature (70F).

I'm heating to 400F with a Firerod cartridge heater. Trying to transfer the heat evenly and quickly to the outside of the vessel.

The vessel is a welded stainless steel cylinder with at least 1/2" wall thickness.

I found some discussion online that concerned me. A guy was doing something similar with a petrol based oil and stated that his pressure increase was roughly 60-70 psi per F degree. If this holds true for my application, I'm approaching 22,000 psi. The last thing I want is a stainless steel bomb in my R&D lab.

A colleague suggested filling the vessel and heating to 400F in an oven, then sealing it off, thus relieving the pressure and filling the void.

 

[zerosum]

A couple of others possibilities:

Dow and Solutia make competing products. They may have a database which would help.

Last resort: Perry's Handbook, Section on Prediction and Correlation of Physical Properties, Sub Section "Coefficient of Expansion" in mine it's p. 3-227

 

[lizking]

API 521 has a whole section on hydraulic expansion with formulae and expansion coefficient estimations.

 

[Compositepro]

This is why all fluid heat transfer systems contain an expansion tank.

 

[vpl]

Or a relief valve.

Patricia Lougheed

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

 

[iainuts]

I'd agree the best thing to do is simply design the system to accomodate the expansion whether that means an expansion tank or a relief device, but that's not always practical in a lab I suppose.

A quick Google search gives typical values for thermal expansion for various types of mineral oil from .00035 to .00054 /F.
Ref:

http://www.multitherm.com/multitherm-ig-1.html

http://www.therminol.com/pages/products/xp.asp

http://www.pantherlubes.com/tl_6.html

If it were me, I'd add a safety factor on top of that, but for the worst case (.00054), a dT from 70 to 400 F will increase volume 17.8%. Round that up to 20% - 25% if you want to be sure.

The remaining volume will be your gas (air?). Assume ideal gas behavior if air and use the change in gas volume and increase in temperature to determine final pressure. That will be your increase in pressure as long as the calculated volume isn't completely displaced.

 

[Latexman]

You don't really need to know what pressure it will go up to. Your vessel should have an MAWP. A PSV with set pressure <= MAWP will be required per Code. CoTE can be calculated from density vs. temperature data:

CoTE = rho1 x (1/rho1-1/rho2) / (T2-T1)

or an EOS. Then you can calculate W (flow requirement) for this scenario using CoTE, Q(heat input), and Cp:

W = CoTE x Q / Cp

As Compositepro said "This is why all fluid heat transfer systems contain an expansion tank." You do not want to build a system expecting it to pop the PSV, but the PSV needs to be there when someone or something goes wrong.

Good luck,
Latexman

 

[btrueblood]

Um. Some more things to consider, all of the above is good advice so far.

1. Add on the vapor pressure of the fluid at the max. temperature to whatever you calculated for thermal-expansion pressure rise.

2. And then ask what happens if the temp. sensor fails, and the heater goes open loop? Ask the heater supplier for a resistance-vs-temp curve, do some heat transfer calcs, and see if there is a peak temperature the device will reach if the operator falls asleep. If you can't guarantee fail-safe operation with this, you may want to consider a relief system like vpl notes - and then you must worry about where you are going to vent a hot hydrocarbon fluid/gas mixture to (hopefully nowhere near my cubical).