LN2 Heat Exchanger Design Questions for quickly cooling JP-8

[GFS666]

Hi Everyone,
For a test setup, I need to quickly cool JP-8 in a heat exchanger. Approximate Specifications are as follows:

Fluid: JP-8
Flowrate: 25.7 L/min (=6.8 GPM, 46 lb/min)
Inlet Temp of JP-8: 17.8°C (=64°F)
Outlet Temp of JP-8: 4.4°C (40°F)
Heat Transfer: 9 kW

I have calculated that I can use LN2 to do this. Assuming a 90% transfer efficiency, assuming the LN2 boils off from liquid and the ensuing gas is raised to a temp of 77K to 173K, I get an LN2 flow rate = 0.033 kg/s = 2 kg/min.

Does anyone have experience in the design of LN2 heat exchangers, know where I could look for design guidance or know of companies that could build an LN2 heat exchanger for this type of application? (note: I do know of Oswald Consulting Engineering in England)(further note: yes, this is a h*ll of a lot of N2 gas being outgassed into the atmosphere. I and my company are very aware of the dangers)(note 3: Yes, they really want me to do this)

Thank You for your responses,

Gordon

 

[EdStainless]

Have you looked at doing this with chilled water (or brine at -5C)? The heat capacity is high which should help.
If you want this to happen fast you will need a lot of pressure so that you can run high flow velocities (very turbulent) to improve heat transfer (on both flows).
The problem with LN is that as it boils it is forming a layer of gas which reduces the heat transfer a lot.

I'll add that I have seen a hydraulic oil cooler running 60ft/sec on the water side, the noise was incredible. It as high alloy SS in order to resist erosion.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[GFS666]

Hi Ed,
Thank you very much for the reply. Chilled water is not an option sorry to say. And your totally right about the LN gas layer. I'm reading up on pool boiling from NASA tech reports and the heat transfer coefficient depends a great deal on that gas layer.

Thank You, Gordon

 

[georgeverghese]

You wont get an LN2 HX to do this duty by a long stretch. The fuel is just going to turn into a frozen brick in the HX. Find a coolant with a an inlet temp which is above the freezepoint / cloudpoint of the fuel. And there may be cases when the fuel may be contaminated with traces of water, so would suggest the min normal inlet temp of the coolant be > 0degC.

 

[Caloooomi]

R134a?
If you really want to use LIN, use it to cool a refrigerant in an intermediate loop between LIN and JP-8 so that you don't risk freezing the JP-8.

edit: In fact, that's what the Oswald guys did that you mentioned. They used Duratherm XLT-120 which is a low temp coolant.

 

[EdStainless]

I would look at two HX, one LN to JP, and the other cold N2 gas to JP.
Keep the LN HX at high enough pressure to suppress boiling, then flash it into the second one which would serve as the pre-cooler for the JP.
This would let the designer optimize the geometry and flow in each HX.
Or, use the LN to chill brine (by sparging) and then use that to cool the JP. A nice little Ti HX would handle that.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed