REFLECTOR:weak gas struts / beating a dead horse dept.

[reflector@tvbf.org]

At 09:40 AM 1/2/04 -0500, you wrote:

>It is still correct that if you take a hydraulically locked system (no air,
>no water or gas contamination, no seal leakage), that the plunger will NOT
>move until you pull with a very large force; one great enough to vaporize
>the hydraulic fluid.  Given that the vapor pressure of hydraulic fluid is
>very close to a perfect vacuum, it will take very large, even massive,
>mechanical force to form a bubble.

It all depends on your definition of "large, even massive."

Imagine a hydraulic cylinder floating in space, surrounded by vacuum.  The 
piston is fully retracted. Both ports are open, so there's nothing but 
vacuum inside.  You plug both ports, and pull on the shaft, what 
happens?  It pulls out with no resistance.  The pressure comes from 2 
places - atmospheric pressure on the shaft, and atmospheric pressure on the 
fluid in the top half of the cyl.

If there's no vapor pressure to worry about - and I agree, with modern 
hydraulic fluids, the pressure is so low we can ignore it - the force is 
going to be 1 atmosphere (14.7lbs at sea level) times the area of the 
piston.  If you have a 2 inch ID cylinder, and a 1/2 inch shaft, and you're 
at sea level, it will be 46.18 lbs (plus friction, but I'm ignoring 
that.)  At that point the fluid will cavitate, a void will form, and will 
get bigger as the shaft is pulled out.  A big force, but not that big, 
especially if there's some leverage on the shaft.

Now, suppose instead of plugging the upper end of the cyl, we plug the 
lower end?  So as you try to pull the shaft out, you're trying to compress 
the hydraulic fluid?  Then, as you correctly point out, you're trying to do 
the impossible, and the shaft won't move until you have enough force on it 
for something to burst.


>I'm pretty sure this is all true, but then, I didn't think they'd find
>Saddam in a spider hole either.

To be totally accurate, they didn't.  It was even better poetic 
justice...http://www.sftt.org/cgi-bin/csNews/csNews.cgi?command=viewone&id=336&database=DefenseWatch%20Archive%2edb