Pumping Liquid Nitrogen

[Paul Hovnanian P.E.]

The circumference, inside and out, decreases. It gets thinner, but the
inside diameter does not increase; that would be absurd.

Absurd? That's how one gets a press fit bearing race on a shaft (by
heating it).

If the rotor and stator are made of the same material (they are in an
induction motor, made of laminated steel) and are the same temperature,
then the gap will get smaller as the temperature decreases.

If the materials are not equal (e.g., aluminum housing or rotor, permanent
magnets, etc.), behavior may differ.

That may be the case. I've seen generators fail at high temps by rubbing
the rotor/stator due to decreased clearance. At first, it seemed to be
counter intuitive.

Most likely, in a DC motor, the rotor heats up more than the
stator/housing, so the gap will tend to shrink, regardless of the outside
air temperature. Different types of induction motors (wound, fixed and
permanent rotors, etc.) may behave differently.

That's possible as well. Behavior of a stator and rotor in contact with
LN2 may have different effects than a machine cooled mainly from the
outside.

 

[Rich Grise]

Not at all. I have been reluctant to reply to you, but the reason has a
lot more to do with me than with you.

Emotionally (and not a particularly rational emotion) my gut feels like
I have tried and tried to be friendly and have been rewarded with a
series of distinctly unfriendly comments.

Well, I've never had any intent to hurt you, and if some of my offhand
comments, intended as humour, were a little sharp or hurtful, then I
apologize.

Sorry to hear about your medical condition - I wish I knew the right
thing to say.

Get Well, OK?

Thanks,
Rich

 

[Tim Williams]

Absurd? That's how one gets a press fit bearing race on a shaft (by
heating it).

Not simultaneously. The bearing (or other socket) is kept at room
temperature or heated, while the shaft (or peg, or...) is cooled or kept at
room temperature, respectively. Heat makes the inner diameter expand, not
cold.

Tim

 

[Spehro Pefhany]

Not simultaneously. The bearing (or other socket) is kept at room
temperature or heated, while the shaft (or peg, or...) is cooled or kept at
room temperature, respectively. Heat makes the inner diameter expand, not
cold.

And you have exactly one chance to get them into the correct position
before the temperature equalizes...



Best regards,
Spehro Pefhany

 

[Guy Macon]

And you have exactly one chance to get them into the correct position
before the temperature equalizes...

....which is why we design and use jigs rather than trying to eyball it.

 

[Glen Walpert]

There we go.

How large is the diameter of the pipes in the percolator?

I don't recall the pipe size anymore but it was way too big for just a
few liters per day. You could calculate the optimal diameter using
the rather complicated methods described in "Two Phase Flows and Heat
Transfer" by Ginoux, or you could wing it and try 1/4" OD 3/16" ID
tubing or therabouts. Effective insulation will be the hard part.

Having once spent 3 days on a single cryo transfer choking
calculation, I suggest the "wing it" approach .

 

[Frithiof Andreas Jensen]

I have been offered a small (about the size of a water
cooler) liquid nitrogen generator at an attractive
price. It generates four and a half liters per day,
using a gravity feed to keep a half liter dewar filled.

Lucky bastard! That what I want for chrismas.

I think I need a small pump that can take the cold and
not add too much heat to the liquid nitrogen. Any ideas?

How about a "bubble pump" or "vapour lift pump"?

Basically the principle used by aquariums. That way you "only" need to pump
very cold nitrogen gas - which could be the gas boiled off from the CPU
cooler

 

[Nico Coesel]

Others have done it without any problems. Semiconductors as a
rule do not mind LN2 temperatures as long as you avoid thermal
shock and icing. Besides, overclockers pretty much expect to
lose a few if they push the envelope.


It will buy me minus 196 degrees, and others who have tried it have
reported a stable system with a 3GHz. CPU overclocked to 5GHz.


Nope. that's for systems with a *source* of nitrogen. A nitrogen
generator takes as much out of the room air as the boiling nitrogen
puts back in. (I would ventalate it anyway, but for the opposite
reason; to avoid any Oxygen concentration around the generator)

That makes me wonder. You are in fact creating some sort of heat-pump
which doesn't sound very efficient (you will be heating the room!).
Isn't it easier to use a modified air-conditioner? All you need is to
evaporate gas in the CPU's heatsink and chipset's heatsink.

 

[joseph2k]

Everything shrinks proportionally in all dimension. "Everything"
includes the gap.

Pretty much, but it still depends on the materials TCE.

 

[Paul Hovnanian P.E.]

Not simultaneously. The bearing (or other socket) is kept at room
temperature or heated, while the shaft (or peg, or...) is cooled or kept at
room temperature, respectively. Heat makes the inner diameter expand, not
cold.

That's how you get a press fit bearing race _off_ a shaft as well. Not
much chance of maintaining a temp difference between them in this case.

 

[Tim Williams]

That's how you get a press fit bearing race _off_ a shaft as well. Not
much chance of maintaining a temp difference between them in this case.

Except for certain circumstances: I've heard of CNC heads which have
ferrite transformers in the cutting head to inductively heat the spindle,
expanding it away from the cutter's shank quickly. Sweet!

Tim

 

[Guy Macon]

How about pushing the liquid nitrogen instead of pumping it?

Imagine an insulated chamber, half-filled with liquid nitrogen, having
a lid with two holes. One hole is for the nitrogen output tube, which
goes all the way from the lid down to within say 1/8 inch from the
bottom of the chamber. The other hole is for the input tube, which is
just a short tube. This is where you blow.

Now you just need a way to blow measured amounts of gas into the short
tube, pushing the liquid nitrogen up the output tube.

Compressed air tank with a needle valve... aquarium pump... CO2 gun
cylinders... ?

I believe that the back pressure would stop the LN2 generator from
putting any LN2 into the dewar flask. The above scheme cannot
result in liquid going out of the output tube at a higher pressure
than the liquid coming in to the input tube, and thuse is not
actually a pump.

 

[Guy Macon]

I was thinking this..

Release +-----------------------+
Valve ||<----+A |+---------------------+|
+---------||-----||----||-+ ||
| N2 pressure | ||
|~~~~~~~~~~~~~~~~~~~~~~~~ | B ||
| LN2 +->-----------------||
+--------------------------->------------------+
~CPU~

'A' is a pressure sensor which operates an electrically controlled
valve.
'B' is a one way valve.

The above system has equal pressure at all points and thus
will not do any pumping. You would have to do this instead:

Release
Valve
||<----+A
+--------||-----||--------+ Vent To
| N2 pressure | Room Air
|~~~~~~~~~~~~~~~~~~~~~~~~ | | |
| LN2 +------------------+ |
+----------------------------------------------+
~CPU~


Alas, you left out the part that caused me to ask
this question:

+------------+
| Liquid |
| Nitrogen |
| Generator |
+----+ +-----+
| |
C--->| | Release
| | Valve
| | ||<----+A
+--+ +---||-----||--------+ Vent To
| N2 pressure | Room Air
|~~~~~~~~~~~~~~~~~~~~~~~~ | | |
| LN2 +-------------------+ |
+-----------------------------------------------+
~CPU~


"C" is whatever pressure the liquid Nitrogen generator
puts out. If C has enough pressure to fill the tank,
I can replace the avove system with:


+------------+
| Liquid |
| Nitrogen |
| Generator | Vent To
+----+ +-----+ Room Air
| |
| | | |
| +--------------+ |
+------------------+
~CPU~


If C does not have enough pressure to fill
the tank, the tank will run out of liquid
and my liquid Nitrogen generator might as well
be something inert and useless, like an
engineering manager.

 

[D from BC]

The above system has equal pressure at all points and thus
will not do any pumping. You would have to do this instead:

k.. gonna go back to the electronics..cause my gas/liquid physics
sucks..

Release
Valve
||<----+A
+--------||-----||--------+ Vent To
| N2 pressure | Room Air
|~~~~~~~~~~~~~~~~~~~~~~~~ | | |
| LN2 +------------------+ |
+----------------------------------------------+
~CPU~


Alas, you left out the part that caused me to ask
this question:

+------------+
| Liquid |
| Nitrogen |
| Generator |
+----+ +-----+
| |
C--->| | Release
| | Valve
| | ||<----+A
+--+ +---||-----||--------+ Vent To
| N2 pressure | Room Air
|~~~~~~~~~~~~~~~~~~~~~~~~ | | |
| LN2 +-------------------+ |
+-----------------------------------------------+
~CPU~


"C" is whatever pressure the liquid Nitrogen generator
puts out. If C has enough pressure to fill the tank,
I can replace the avove system with:


+------------+
| Liquid |
| Nitrogen |
| Generator | Vent To
+----+ +-----+ Room Air
| |
| | | |
| +--------------+ |
+------------------+
~CPU~


If C does not have enough pressure to fill
the tank, the tank will run out of liquid
and my liquid Nitrogen generator might as well
be something inert and useless, like an
engineering manager.

D from BC

 

[[email protected]]

I believe that the back pressure would stop the LN2 generator from
putting any LN2 into the dewar flask. The above scheme cannot
result in liquid going out of the output tube at a higher pressure
than the liquid coming in to the input tube, and thuse is not
actually a pump.

Do you have the specs on what back-pressure the LN2 generator can
tolerate?

Michael

 

[Guy Macon]

Do you have the specs on what back-pressure the LN2 generator can
tolerate?

I have a request in to the manufacturer. They appear to be out
until 02 January.

 

[Rich Grise]

So where does the pressure come from?

If I could pull my emitter follower up entirely by its boostraps, I
wouldn't need a battery anymore.

Maybe with a seperate expansion chamber and keen connections, but
offhand, it doesn't work: internal pressure is internal pressure.

How about two dewars? One gets filled while the other drains onto
the heat sink. They just have to be higher than the computer - I
wonder how big this LN2 genny is? It would have to be on a top
shelf, but from there gravity should do all the work that's
needed.

And if the CPU is insulated from air properly, once it's chilled,
the boil-off rate should decrease some.

Cheers!
Rich

 

[Rich Grise]

+------------+
| Liquid |
| Nitrogen |
| Generator | Vent To
+----+ +-----+ Room Air
| |
| | | |
| +--------------+ |
+------------------+
~CPU~


If C does not have enough pressure to fill the tank, the tank will run
out of liquid and my liquid Nitrogen generator might as well be
something inert and useless, like an engineering manager.

You've just drawn the diagram I was too lazy to draw. ;-) Just insulate
the bejabbers out of the CPU, and use that container as your dewar,
with room for extra as needed.

Would using the exhaust N2 enhance the operation of the genny?
It's sure to be quite cold, you know.

Have Fun!
Rich

 

[Rich Grise]

I believe that the back pressure would stop the LN2 generator from
putting any LN2 into the dewar flask. The above scheme cannot result in
liquid going out of the output tube at a higher pressure than the liquid
coming in to the input tube, and thuse is not actually a pump.

LN2 drip ========
\\
||
______ || ______
| | || | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | cold N2 | |
| | | |
| | | |
| |~~~~~~~~~| |
| | | |
| | | |
| | LN2 | |
| | | |
| | | | <- Styrofoam
| | | |
_______| | | |_______
| | | |
| | CPU | |
| | ======= | |
==================================================== PCB
| |
| | <- more Styrofoam
| |
|_______________________________________|

Cheers!
Rich

 

[Rich Grise]

I have a request in to the manufacturer. They appear to be out until 02
January.

Got pix? Or any other kind of specs?

Thanks,
Rich