Subject:
|
Re: Pneumatics Vs. Hydraulics
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Sat, 31 Jul 2004 17:08:08 GMT
|
Viewed:
|
1140 times
|
| |
|
|
In lugnet.robotics, Jon Gilchrist wrote:
> Some of this is not right, some is not very clear. I've been working
> in fluid valve design for 14 years. Let me clarify a couple things.
>
> On Wed, 28 Jul 2004 05:28:11 GMT, lego-robotics@crynwr.com (Mr S)
> wrote:
>
> > I don't think that pressure shares the same fluid
> > dynamics as liquids.
>
> First, some definitions. Pressure is a force, fluid is a type of
> medium, and liquid is a type of fluid. To say "pressure shares the
> same fluid dynamics as liquids." is like saying that voltage shares
> the same properties as copper. I think you meant "I don't think that
> gases share the same fluid dynamics as liquids" which is correct, they
> don't. Similar, but with differences.
>
> > Certainly on the larger scale,
> > fluid works much better at evenly distributing
>
> I think "fluid" here must mean hydraulic fluid, a liquid.
>
> > pressure even though pnuematic systems are capable of
> > very strong movements indeed. The viscosity of
> > hydraulic fluid is such that it does not compress well
> > where air does compress.
>
> Yes. This is one of the fundamental differences between hydraulic
> (liquid, whether oil or water) and pneumatic (air or other gases)
> systems. Hydraulics are essentially incompressible.
>
> > Even on a smaller scale, the
> > air must compress a certain amount before it fluidly
> > distributes pressures, and that is the reason that
> > LEGO has the air tanks. The more stored pressure that
> > you have in the form of compressed air, the more work
> > you can do with the LEGO pnuematics.
>
> No. Compression has nothing to do with distribution of pressure. The
> pressure is distributed in a fluid system because of the nature of the
> fluid to exert pressure evenly in all directions, not because it
> compresses.
>
> The air tanks are used as a reservoir to store volumes of compressed
> air, because each actuation of a cylinder exhausts some air into the
> atmosphere. The air tank in an air system is similar to a water
> tower. It stores air and allows you to have a smaller compressor
> because the tank evens out the system demand.
>
> > Fine adjustments
> > are capable with hydraulic systems and pnuematic
> > systems are used generally in open or closed
> > situations. If you have a need to have the valve half
> > open, hydraulics is your best bet. There is the
> > differences in a nutshell.
>
> Fine adjustment and position feedback are difficult in both hydraulic
> and pneumatic systems. The compressibility of air further complicates
> pneumatic systems.
>
> Having the valve half-open may or may not work in a given application.
> A half-open valve is a crude flow-control. But granted, it is easier
> to control flow in a hydraulic system with a half-open valve than in a
> pneumatic system. Not ideal, but it can work. (Upon re-reading this,
> I realize that it really depends on the valve design. Some valves are
> designed to use as flow controls, but most aren't. Lego valves are
> not.)
One way to change pressure in Lego pneumatics by a small amount is to switch in
a short tube (2cm) with an antenna stuck in the end. This lets only a small
amount of air out of the cylinder, whilst not exhausting any to the atmosphere.
> > It is not possible to build hydraulic systems on the
> > scale of LEGO for the cost of LEGO. Some modelling
> > enthusiasts do have small scale hydraulic systems, but
> > they are NOT cheap.
>
> This is probably true. Hydraulics tend to be more expensive than
> pneumatics, and more complicated. You have more of a mess to deal
> with (ever spill/spray hydraulic fluid all over a room? How about
> spraying air all over a room? Which is the bigger mess?) with
> hydraulics. Your system needs to be tighter to eliminate leaks. On
> the other hand, you can operate at far higher pressures with
> hydraulics. A typical air system runs at around 100-250 psi.
> Hydraulics go upwards of 5,000-10,000 psi.
I think the robot wars rules allow 1000psi pneumatics and 3000psi hydraulics, if
this gives some comparison. Most roboteers use CO2 cylinders with pneumatics,
or inject liquid CO2 to get about 570 times the expansion ratio, in order to
flip other 100kg robots out of the arena. Very few robots use hydraulics,
probably because of flammability, complexity and reliability, given that the aim
of the game is to destroy other robots. Sadly, liquid CO2 at freezing
temperatures would ruin Lego parts!
> > BTW, you can make your own air tanks. The air tank in
> > the LEGO pnuematic systems with the two port actuators
> > is really only there to supply stored pressure. With a
> > 12 or 20 ounce plastic soda bottle and some silicon
> > sealer, you can make a huge air tank. This makes it
> > possible to build a LEGO pnuematic machine which has
> > the capacity to do all the work needed before
> > recharging the tank, depending on how much work is
> > required, of course.
>
> I've been meaning to do some burst tests on this. It is possible to
> overpressure a bottle and make it burst, which is an ugly and
> dangerous thing. In general, a smaller smooth bottle should be
> stronger than a large dimpled one. You can use multiple small bottles
> to get the same effect, it's the volume you need, not the shape or
> single container.
>
> > Hydraulic systems make use of a principle of physics
> > where a cylinder with a 1 inch square plate on one end
> > and a 10 inch square plate on the other, takes 10 psi
> > on the one inch side, and applies 10 psi to each of
> > the 10 square inches on the other end, thus giving a
> > power boost by a factor of 10. This is idealized, but
> > close to accurate, and the reason that hydraulic fluid
> > is such that it does not compress. Any compression of
> > the 'power transfer fluid' and the boost ratio is
> > decreased accordingly.
>
> I found this not quite clear, but essentially correct. It works for
> both pneumatic and hydraulics. If you exert a 10 psi pressure on a 1
> square inch cylinder piston, the *force* that the piston rod exerts
> will be 10 pounds. You don't exactly get a "boost" (although there
> are products that do boost pressure this way), you get a force. But
> there is a force multiplier for each cylinder that you can use to
> figure out what force you get out based on what pressure you put in.
> The force multiplier is equal to the piston area in square inches (or
> similar units if you're working in metric). And you do lose some
> power because of compressibility, and you lose some because of the
> friction of the cylinder seals. Usually you should oversize the
> system by about 20% to accomodate this.
>
> The formula is (input pressure) X (piston area) = (output force)
>
> For the extend stroke of a cylinder, the piston area is the area of
> the bore of the cylinder. For the return or close stroke, it's the
> area of the bore minus the area of the ram.
>
> > Because of the inherent leakage in the LEGO or any
> > pnuematic system, it requires not only air under
> > pressure, but large volumes of it, replenished often
> > or continuously, while a hydraulic system is a closed
> > system that requires replenishment only rarely.
>
> Leakage isn't exactly the right term here. Leakage is for
> unintentional losses, like around a fitting or seal. What is being
> explained is exhaust. A pneumatic system is generally an open system
> where the air is compressed, fed to an actuator, and then exhausted to
> atmosphere. So by "inherent leakage" I think you were talking about
> exhaust. And yes, pneumatic systems use huge volumes of air.
> Luckily, air is cheap.
If TLC did make hydraulic parts, each valve would need to be a reverser with 4
ports, with the fourth port returning fluid to the reservoir. The pump would
need to be immersed in the reservoir to fill with fluid. Air bubbles would also
be a problem.
The system would have to use a non-toxic fluid such as water, since it's meant
to be a toy. Hydraulic fluid is also highly flammable and brake fluid is the
most flammable liquid in your car (drip petrol onto the exhaust manifold and it
smokes, brake fluid flares). Unfortunately, water is more compressable than
hydraulic fluid, but for a toy this shouldn't matter.
Is it true that water is used for point to point hydraulics but only oil is used
for variable movement? This would make sense if water is plentiful but
compressable. Are there examples of variable water-controlled hydraulic
systems?
> > Okay, see what the Captain can do?
> > The answer is that LEGO does NOT make hydraulic parts.
>
> Chances are if you used real hydraulics with your LEGO parts, you'd
> break a lot of parts because the forces could be very high.
I once tried using a non-return valve from the old pneumatic system to pump
water from a tank to a hose. The valve soon expired! I don't recommend the use
of water or any other hydraulic medium with Lego parts.
> Here's some info I have determined about some LEGO pneumatic parts.
> These are the current cylinder and pump designs. All dimensions in
> inches.
>
> Pneumatics Info
>
> Large cylinder
>
> Bore diameter .472
> Ram diameter .196
> Stroke 1.101
> Extend
> Area = .175 sq. in.
> Force @ 50 psi = 8.75 lbs.
> Force @ 10 psi = 1.75 lbs.
> Displacement = .193 cu. in.
> Retract
> Area = .145 sq. in.
> Force @ 50 psi = 7.24 lbs.
> Force @ 10 psi = 1.45 lbs.
> Displacement = .159 cu. in.
>
> Small cylinder
>
> Bore .229
> Ram .125
> Stroke .595
> Extend
> Area = .041 sq. in.
> Force @ 50 psi = 2.06 lbs.
> Force @ 10 psi = .41 lbs.
> Displacement = .025 cu. in.
> Retract
> Area = .029 sq. in.
> Force @ 50 psi = 1.45 lbs.
> Force @ 10 psi = .29 lbs.
> Displacement = .017 cu. in.
>
> Large pump
>
> Stroke =.976
> Bore .472
> Displacement = .171 cu. in.
>
> Small pump
>
> Stroke =.667
> Bore .229
> Displacement = .027 cu. in.
Useful info! I think Kevin should use this to calculate the number of cylinders
for his scorpion.
> It occurs to me now that the small pump is probably double-acting (it
> pumps in both directions). I don't have data for it in the other
> direction, but it should be trivial to determine.
I don't think it is. The ones I have aren't.
> I don't have the numbers handy, but I think the large pump can
> generate somewhere between 50 and 60 psi. I still need to test the
> small pump.
I run Lego pneumatics off a car tyre air compressor at 25psi (30psi for short
periods only). I think much more than that would break the seals. Therefore get
the one with the variable pressure limit if you can.
The cylinder probably needs a bit of extra pressure capacity since the system
suffers from the balon effect of only flexible hoses are used. Use flex tubing
for straight bits wherever possible.
>
> -Jon Gilchrist
Mark
|
|
Message has 1 Reply:
 | | Re: Pneumatics Vs. Hydraulics
|
| (...) I'm not clear on this. It sounds interesting, I just can't picture how it's done. Where does the tube go? Is it acting like a little variable volume reservoir? (...) I agree completely here. Hydraulics would be *much* more complicated, (...) (20 years ago, 2-Aug-04, to lugnet.robotics)
|
Message is in Reply To:
| | Re: Pneumatics Vs. Hydraulics
|
| Some of this is not right, some is not very clear. I've been working in fluid valve design for 14 years. Let me clarify a couple things. (...) First, some definitions. Pressure is a force, fluid is a type of medium, and liquid is a type of fluid. To (...) (20 years ago, 29-Jul-04, to lugnet.robotics)
|
10 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
Active threads in Robotics
|
|
|
|
|
|
