thread774-312901 is it necessary t

[Praveen Davis]

thread774-312901
is it necessary to test boil off at completely filled condition. we are doing this test at 60% liquid level. denominator is same total volume. is it a wrong practice?

 

[iainuts]

Hi PD. Regarding heat leak into a VJ vessel, let's talk about the heat leak first. Consider breaking it down like this:
1) Convective and radiation heat transfer from the environment to the outer jacket.
2) Thermal conductivity through the outer tank wall.
3) Thermal conductivity through the inner vessel supports and insulation, radiation heat transfer through the vacuum space, and some almost negligible convective heat transfer (essentially, conductive heat transfer) through the vacuum space.
4) Thermal conductivity through the inner tank wall.
5) Convective and radiation heat transfer from the inner wall of the inner vessel to the fluid.

The 5 heat transfer mechanisms above can be seen as being in series from atmosphere to the cryogenic fluid. Of those 5, only #3 matters regarding the overall heat transfer. The other 4 are so slight that they contribute essentially nothing to the insulating value of the cryogenic tank. The overall heat transfer is dictated almost exclusively by #3.

Given heat transfer is dictated almost exclusively by #3 above, then as long as the atmosphere and cryogenic temperatures are held constant, the heat transfer is held constant. But here's where it can get a bit tricky.

In a vessel that's 60% full, there's some amount of the inner vessel that's covered by liquid and some amount that's covered by gas. Assuming the liquid is boiling (saturated), the liquid is always at 1 temperature. That's a simplified description of the liquid, but as long as it's close to saturated, it's generally a valid assumption. The gas is also going to be close to saturation although here we have some variation. Actual gas temperature in a vessel can be upwards of 20 degrees C above saturation. The gas (and liquid) tends to have some thermal gradient between the liquid/gas interface and top of the vessel. There are also convection currents in both the liquid and gas so you can imagine the gas having some amount of circulation. It is not all the same temperature and often the gas that's escaping the vessel is significantly warmer than saturated temperature. So, 2 observations regarding a vessel that's 60% full:
1) The liquid is boiling at saturation temperature (assuming it's come to equilibrium) across the entire inner vessel surface where there is contact.
2) The gas is warming and is slightly superheated across the entire inner vessel surface where there is contact.

The slightly warmer portion of the vessel in contact with the gas results in a slightly reduced heat transfer across that area. Consider heat transfer mode #3 above - heat transfer is a function of inner vessel temperature and that might be very slightly warmer than saturated temperature for the 60% full vessel.

The other consideration is that the temperature of the gas escaping the vessel for the 60% full case is going to be slightly warmer when compared to the 100% full tank, so part of the energy that gets into the inner vessel from ambient goes into superheating the gas (increases enthalpy). However, this added heat is very small compared to the heat of boiling.

These two factors, (1) lower heat leak from outer tank to inner tank due to lower dT and (2) more heat per unit mass removed by higher temperature gas escaping, tend to reduce the apparent heat leak for the 60% full tank when compared to the 100% full tank.

In practice, this difference is very slight because of the difference in dT between inner and outer tanks is very small and the sensible heat absorbed by the gas is small compared to the heat absorbed during boiling.

Is the 60% test valid? Honestly, it's not much different than a full vessel, though heat leak measured may be slightly reduced. Variation will be worse for a vertical vessel when compared to a horizontal vessel but regardless, I tend to think the 60% full vessel is going to be relatively accurate. Measurement inaccuracies could exceed the inaccuracy due to the liquid level conditions.

 

[EdStainless]

I'll take a shorter answer, if you always use 60% then you will be able to compare tanks.
If you find one with a high boil off rate at 60% rest assured that it will still be high at 100%.
It is nice to use 60% so that when you find a bad tank there is less gas to offload before you can warm it and pump it down.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube