Design of Simple -40 Freezer System with LN2 and Styrofoam 1

[dctherandomguy]

The goal is to design an LN2 Freezer system that is portable and inexpensive as possible. I want to use Styrofoam (polystyrene) walls with a certain R-value to be built around a unit with a certain heat gain. The idea is to purchase components such as a tank, hose, (pump?), solenoid, and nozzle, to inject LN2 into the system. I want to use existing controller boards (that I have), use thermocouples as input and the solenoid control as the controller output. I have not worked with LN2 before, nor am I particularly savvy on the subject of thermodynamics.

Given a chamber volume, heat resistance of chamber walls (R-value), and heat gain (in watts) of the unit under test;

how can I calculate the necessary flow of LN2 to set the chamber at a certain temperature?

I'm assuming LN2 tanks can be manually filled, pressurized, and LN2 flow can be controlled, all with commercial products, which I can (hopefully) buy separately. Manufacturers often include a controller with a stock LN2 injection setup, but in the interest of saving money and customization, I want to use existing controllers.

Any suggestions?

 

[IRstuff]

Is this for school?

TTFN

FAQ731-376

 

[dctherandomguy]

No, it's for my job.

 

[IRstuff]

Ok, then I'm completely unclear how you're going to be able to control the chamber temperature. If LN2 is flowing into the box, it's at 77 K, which is a far cry from 233.15 K.

LN2 can be dangerous:
>> all LN2 materials need to be insulated to prevent freeze burns to personnel/operators
>> ventilation needs to be considered to ensure adequate breathing air


TTFN

FAQ731-376

 

[dctherandomguy]

The purpose of this system is to mount on a vibration table for product testing. We have thermal chambers, but not ones that have simultaneous temperature-vibration capabilities. We want to know the effect of vibration when the product is at -40º C.

I've never worked with LN2. My first thought to how this system should work would be to control the flow of LN2 through pipes or coils in the Styrofoam box. With a fixed surface area of the pipes, I could control flow of LN2, and while mathematically determining the approximate heat loss for the flow I'm providing with the gains from the unit and outside the box, use thermocouples to determine whether to increase or decrease the flow rate.

I was led to believe that we could simply inject a quantity of LN2 into the box, and some constant evaporation would facilitate the venting of the nitrogen gas. The determined rate of injection and evaporation could keep the box at an approximate temperature.

 

[MintJulep]

Seeing as you list yourself as an electrical engineer, I can understand the lack of experience with the thermodynamic elements.

However, if your goal is to maintain a given temperature in the enclosure, and you plan to have a closed-loop control system, why do you think you need to calculate the necessary flow in real-time? At any give instant you need to maintain the flow constant, increase it, or decrease it. That seems fully sufficient for a control system.

 

[dctherandomguy]

I guess I don't quite understand how a typical LN2 freezer works, are there any resources you can direct me to?

 

[btrueblood]

LN2 has a specific heat of vaporization, Cvap, and a specific heat, Cp, as it warms from 77 K to 233 K. Multiply the mass of LN2 injected by Cvap, and by Cp(delta-T), add the two numbers, the result is the amount of heat absorbed by the warming of the LN2. Google both terms, or look them up in the CRC Handbook of Chemistry and Physics.

Don't mount a LN2 dewar on a shaker head, if you shatter the vessel and spill the contents, you're gonna have problems. Modulating (pulsing) slugs of LN2 directly onto your device is probably going to cause a lot of frost buildup too, unless you take care to work in very low humidity.

Have you considered using dry ice? Pack a bunch of dry ice chips into a styro box, and monitor hardware temp. as it cools down, start testing when the temp. gets to -40F (remove the box and blow off excess chips at that point). Concerns for frost still apply; melting frost dripping down into the shaker table may not be a great idea.

 

[IRstuff]

A "typical" LN2 freezer runs at 77 K, not 233 K.

TTFN

FAQ731-376

 

[Zapster]

I recommend that you review all relevant safety standards before you start playing with liquid nitrogen. One place to look for good safety standards is

http://www.cganet.com/

 

[MintJulep]

While btrueblood has hit the fundamental basics, there are some practical concerns to deal with as well.

For instance: The surface temperature of your "evaporator" will be below the dewpoint of anything likely to be your ambient atmosphere. So it will very quickly become a block of ice.

 

[btrueblood]

Mint,

Yes, dewpoint is an issue...but I have seen it work okay (LN2 chilling) in very dry, cold, desert air, at high altitude. But, we designed all of the exposed surfaces to be easily scraped prior to functional (not vibration) testing, and had drip trays and gutters to contain and channel the inevitable runoff of melting ice/frost. LN2 slugs were directed into a steel block heat exchanger that was tucked under a insulation blanket along with the hardware to be tested.

 

[unclesyd]

Why not good old fashioned compressed air?

Why not use a Vortec Tube from Vortec or Exair that can achieve your required temperature. I've worked with both companies and gotten very good support.

I would also check with the low temperature freeze manufactures like Scientemp. I know some of them build small cabinets, around 1 cubic foot, that go to -85C.

http://www.vortec.com/solutions_cooling.php

http://www.exair.com/en-US/Primary ...ing/Vortex Tubes/Pages/Vortex Tubes Home.aspx

http://www.scientemp.com/freezers.htm

 

[dctherandomguy]

What I'm going to do is build a Styrofoam box around one of our units on the vibe table. It'll be secured to the table during vibration. I'll have a tank of LN2, a hose, and a solenoid valve, probably all 1/4". A simple controller will control the solenoid based on temperature. When the LN2 is injected, it will more or less vaporize, and the gaseous nitrogen will vent. I'll set up a PID loop to control the temperature this way. There is a distinct possibility it will overshoot, but the idea is to set it up and just make it "work." It's a quick and dirty solution, but it's apparently commonly used.

Any recommended suppliers in North America, specifically Canada, for LN2 tanks, hoses, and solenoid valves?