Using a liquid nitrogen boiler to power a pneumatic humanoid robot 3

[BenjieN]

I am working on a project to develop a system that a pneumatic robot could carry that could supply it with pressurized gas. We are considering having a tank of liquid nitrogen, which we than take small amounts of and then heat with either ambient air or with fossil fuels to raise the pressure (the robot consumes about .61 mols/sec (about 1.2 SCF / sec) at 100 psi.)

One problem is how to get the liquid nitrogen into the heat exchanger. As far as I can see I can either pressurize the liquid nitrogen dewar (which will make it heavy) or have some kind of pump that can push a few grams a second of liquid nitrogen through at cryogenic temperatures, but I don't know if that even exists.

These people (

http://www.aa.washington.edu/AERP/CRYOCAR/HomePage/Index.htm)

were doing something similar, if you want more background.

So I guess the question is, anyone have any crazy ideas? Know of a lightweight, high pressure, cryogenic pump?
Is this just completely nuts?

-Benjie Nelson
-bmnelson at gmail dot com

 

[KiwiMace]

You say you need 100psi anyway. How were you thinking of getting the operating pressure? Sounds from your description like you had subcooled liquid N2 at room pressure and you were just going to collect the vapour.

 

[pmover]

more energy will be expended in making the liquid N2 than what energy is extracted for auto use (ok, i note that you stated a similar situation).

ever thought of using gravity to pour or drain the liquid N2 into the exchanger rather than using a pump?

good luck!
-pmover

 

[BenjieN]

The idea is to take some of the liquid Nitrogen and then heat it in a vessel so that the pressure increases as the nitrogen turns into gas.

Simplest version: Liquid Nitrogen is poured into a large boiler at atmospheric pressure. The boiler is sealed with a flame under it. The flame adds heat, vaporizing Nitrogen and increasing the pressure to above 100psi. Gaseous nitrogen is pulled off the top through a 100 psi regulator and heated further to increase it's specific volume and to prevent frosting. This is used to power the robot.
Problems with this are:
Where to find/how to make a light weight vessel that can take more than 100 psi, cryogenic temperatures, and burner under it.
Added complication with two burners that we want to be light and also safe for people to be around

Second idea:
Liquid nitrogen in kept at room pressure, and a small amount is pumped into a 100 psi heat exchanger where it expands to close to ambient temp. This is better(?) than just compressing air because the pump is moving a few grams of liquid rather than a few grams of gas (much more work to compress). so it can be smaller, lighter and use less battery (which the robot also has to carry). Or at least that is the hope.

Hope that clarifies.

 

[Transient1]

Is this a commercial or academic venture?

 

[MintJulep]

Seems impractical to me.

How about dry ice in a pressure vessel.

Regulator on the outlet to the consumers. Pressure relief valve to prevent it from exploding.

 

[BenjieN]

Transient1: Commercial (anybots.com)
MintJulep: It seems that dry ice has much the same problems as LN2 just worse. It takes more heat to vaporize, still needs to be brought up to a temperature that won't frost, and Carbon Dioxide is heavier per mol than N2 so you get less Work per kg out of it.

 

[moltenmetal]

100 psig isn't all that much pressure- not much of a pressure vessel required relative to what you'd need just to contain the vacuum that a dewar requires to function. Use the LN2 to pressurize the dewar and a control valve to regulate how much LN2 goes into your evaporator.

You're going to have static (heat gain through the dewar) losses of LN2 to worry about regardless- better to vent this at 100 psig than at atmospheric.

1.2 scf/s is a sh*tload of N2. The dewar's going to be large in volume, and heavy, for any reasonable operating period. Done any calcs on the volume of LN2 required?

 

[vpl]

I would wonder about the commercial application if liquid nitrogen is needed. Seems like it would make the system rather cumbersome.

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.

 

[BenjieN]

Moltenmetal: yeah, I calculate that I use a little more than a liter a minute walking, so I am looking at refueling every 7-10 min at the moment, which is bad, but better than compressed air, which only gets me about 5 min operation in equal weight, plus I need to swap tanks, rather than pour in more fuel.

In regards to your suggestion: It seems a little like a chicken and egg problem. In order to force the fluid into the evaporator through the valve the evaporator needs to be at a lower pressure than the dewar, but if I am using the evaporator to pressurize the dewar then it needs to be at a higher pressure. I suppose I could use the hight of the LN2 to push the fluid if I take from the bottom.
Also 100 psig is not that much, but it is significantly larger than the -14 psig that is a total vacuum.

Actually at that speed of use maybe I don't need a dewar at all. I am trying to heat it up to create pressure, maybe a dewar is just making things harder by insulating. I can just pour in the nitrogen when I need it. Then I only have to worry about frosting all over everything.

 

[Compositepro]

Such a system would be poor in overall energy efficiency but if you want to use liquid N2, for whatever reason, the answer is a pressurized dewer and ambient air vaporizer.

This is how nitrogen is stored and used in most mfg. plants.
The liquid is stored at 100 psi and simply passes through a heat finned heat exchanger to absorb heat from the air (the delta T is quite large). A fan can be used on the air side but is usually not needed. The product is 100 psi N2 vapor.

 

[btrueblood]

Laugh all you want, but the idea is not as crazy as it sounds at first glance. LN2 is still cheap, as it is a byproduct of LOX production (mainly for medical purposes); even at the time it was built, the UW LN2 car was not much costlier to operate per mile than the gasoline powered version (but it was very slow, underpowered in comparison). The mail jeep shown in the link operated quite nicely, cheaply, and cleanly on the campus at U of Wash (the vehicle was built by a team led by my old aero. professors, so I got to hear about it whenever I dropped in for a visit).

Their biggest problem was icing of the ambient-air heat exchanger in the humid air up here. I'm pretty sure the dewar was vented to atmosphere at a fairly low temperature, and the fluid was pumped to the ambient-air HX by a small positive-displacement pump. The motor for the car was an ancient marine triple-expansion steam engine, unmodified. You could try and contact professor Adam Bruckner, or Carl Knowlen at <theirlastname>@aa.washington.edu if you want more details or advice. They will probably just send you along to the people here:

http://www.mtsc.unt.edu/CooLN2Car.html

They use your concept #1 (pressurized liquid storage tank), apparently. I doubt the air motor is as efficient as the UW engine, which is similarly probably less efficient than something purpose-designed.

 

[MintJulep]

How about liquid Hydrogen.

No need for a separate fuel source for heat.

 

[2stroketech]

what is the advantage of this over jsut using compressed air?

 

[dcasto]

you'd be better off with an air compressor.