Expansion factor

[dbday]

Can anyone comment on this;

I have a typical surge tank, out in the open, half filled with water and half with air.

The air volume is required to expand / contract in-line with the surge pressures that arise in the connected pipeline.

Does the expansion / contraction of the air follow an isothermal process or an adiabatic process or is it, in reality, somewhere in between ?

Would insulating the tank make any difference to this ?

Comments appreciated.

 

[racookpe1978]

An exposed tank (in outside air especially - but even if indoors) would have much greater change in temperatures from ambient air temperature changing than it would have from a small compression change of the fluid (water) pushing up into the tank and compressing the air trapped in the top of the tank. Water flow in/out of the tank will be fairly slow - this isn't a diesel piston after all - so heat transfer out of the air to the tank walls to the room/ambient will keep heat rises (falls) minimized to those of the ambient air.

Heat exchange between the lowest air level (where it touches the cooler water) would cool down the trapped air in any event over a period of an hour.

(If the expansion tank were vented, then there is no change in pressure of the air volume above the water.)

 

[Unotec]

Most surge tanks are either vented or have some sort of pressure relief and gas make-up if a blanket is needed.

In any event, the only significant pressure build up will be from fluid entering the tank from the pipeline.

In rare ocassions, and when the tank is very large, the ambient temperature and sun radiation, might expand the gas and cause an overpressure. But this is very rarely an issue for the vessel itself. However, depending on the process line sensibility to pressures, it might be an issue.

Answering your question, it would be an adiabatic expansion or compression

<<A good friend will bail you out of jail, but a true friend
will be sitting beside you saying ” Damn that was fun!” - Unknown>>

 

[MortenA]

adiabatic isentropic (like a bicyclepum / piston).

Best regards

Morten

 

[rmw]

When you compress air you heat it. How much you compress it determines whether or not it is detectable with normal instrumentation. Where the heat goes depends on whether or not your tank is insulated. If not insulated, it goes out through the walls to the surroundings as well as into the liquid. If insulated, principally into the liquid.

rmw

 

[GuyFromDenver]

I don;t think you have much to worry about because air is compressible. As with and pressure vessel you should, by code, have a relief valve on it just in case. It would be safest to vent it to ambient and thereby solve the problem. If you cannot vent it then calculate the pressure in the vessel were it full of air with no fluid fraction at your system's ambient pressure and then compressed to whatever level is considered to be full.

PS: It is adiabatic compression

 

[dbday]

Hi GuyfromDenver,
It's not a code issue that I have, it is one of modelling the behaviour of the air charge correctly. Within the surge analysis package I have there is an input box for the expansion factor. The software provider offers no real advice on what input to use other than to say to use a factor between 1.2 to 1.3 on the basis that in their view the process falls between a true isothermal one (factor=1) and a fully adiabatic one (factor=1.4).
I guess it is a safe option if you are not sure either way as the results obviously fall in the middle.

I was just trying to get a better handle on things and be more correct.

 

[FredRosse]

To answer your question:

Does the expansion / contraction of the air follow an isothermal process or an adiabatic process or is it, in reality, somewhere in between ?

ANSWER: It is virtually always somewhere in between.

Closer to adiabatic if the compression/expansion is fast, closer to isothermal if the process is slow. Of course, the influence of exterior heating or cooling of the expansion tank (as mentioned previously) could change the results. The mass (and specific heat) of the tank walls, and the heat transfer coefficient between the interior gas and the tank walls also play here.

Would insulating the tank make any difference to this ?

Insulating the tank would make a difference. However the mass of the tank itself (and the liquid inside the tank) may be more relevant for many configurations.

In theory, if you could have an expansion tank with zero mass, or zero specific heat, perfectly insulated, and perfect insulation between the expansion tank gas and the liquid inside the tank, then you would have a purely adiabatic expansion/compression.