Subject:
|
Re: SSClagorpion - Compressor
|
Newsgroups:
|
lugnet.technic
|
Date:
|
Mon, 17 May 2004 18:41:32 GMT
|
Viewed:
|
18351 times
|
| |
|
|
In lugnet.technic, Steve Hassenplug wrote:
> ok, so we have about 200 messages in this thread. I'm going to start changing
> the subject to actually include the... subject. But still keep SSClagorpion,
> for those who chose to skip the thread.
Hi Steve,
Were you able to come up with anything in your compressor experiments?
The topic of compressors is an interesting one. As you know there are many
many solutions to this problem.
The most popular is Ralph Hempel's design that you can find on Dr. Soh's
website. It uses small pumps and RIS motor(s?).
There is an on-going discussion/debate about small pumps vs. medium pumps.
One aspect of a functioning compressor is highest achievable pressure. The
pressure depends on the force of compression, and the area of the face of the
compressing piston. Small pumps and the proper downgearing of Mindstorms motors
make for a very good match of the motor's torque and therefore the force needed
to push the piston all the way in on small pumps. This lets small pumps get to
high pressure quickly. The downside is that the small pumps don't move a lot of
air. Any design that uses medium two port pistons, needs a large volume of
compressed air to work with rapidity.
Using modern LEGO hand pumps (less the spring if you like) and mindstorms
motors does not make for a very good compressor. The face of the piston in the
hand pumps is *much* larger than the face of the small pump. The total volume
of air contained within an expanded hand pump is *much* larger than the volume
of a small pump.
I'll have to do some measuring at home, but I'd like to know how many small
pump volumes it takes to make a large pump volume. At that ratio, I'd like to
know the sum of surface area of the faces of the small pump, vs. the surface
area of the large pump. I'd also like to know how many small pump piston faces
it takes to match the surface area of the hand pump piston face.
The surface area and pressure are what define the amount of force needed to
fully compress a pump. The force to compress is provided by a combination of
motor strength (torque) and any gear trains leading to the pump handle.
In a compressor you buy at the store, the pump runs very fast, because of
designs that match the motor torque, gearing, area of pump piston face, and
pressure requirements.
I've tried to make LEGO compressors that run the small pumps very fast. The
small pumps get *very* hot and the pumps function can get permanently damaged.
Small pumps are not very robust for heavy duty operation.
The large pumps expand and contract more slowly than small pumps. Certainly
the circumference of the seals is longer on large pistons than small pistons,
therefore increasing friction, but I don't think that is enough to explain the
difference.
I'd guess that the piston seal technology is different for smalls versus large.
It is my belief that small pumps are not heavy duty enough for making good high
volume, high pressure compressors. It is also my beleif that Mindstorms motors
do not provide enough torque to make a good compressor.
One downside is that building with large pumps takes a lot more structural
volume than small pistons.
Once you've fixed yourself with a given pump choice, the overall output of a
compressor depends souly on the number of pump compressions per minute. This is
constrained by the pressure needed to make a MOC work. The higher the pressure,
the more force needed to compress the pump. The large the force, the more motor
torque.
Using large pumps driven by motors is typically done by converting rotary motion
from the motor, to linear motion (compressing the pump) using either a large
pulley or 40T gears. 40T gears are better, because they provide a slightly
better range of motion than the large pulley.
One rotation of the 40T gear leads to one compression of the pump. The motor
has an easy job when it is expanding the pump. For the compression of the pump,
the maximum force is only needed when the pump is *mostly* compressed.
This means that the maximum torque is only needed for a small portion of the
rotation of the 40T gear. This allows us the ability to add *more* pumps to the
compressor without providing more motor torque, by timing the maximum
compression of each pump so it does not overlap maximum compression of the other
pumps.
For example, we could use two pumps that are 180 degrees out of phase from each
other, three pumps that are 120 degrees out of phase from each other, or four
pumps that are 90 degrees out of phase.
Packing four pumps into a compressor can make for a very large compressor.
To reduce the size of a four pump compressor, I chose to use a large (well
medium, but there is not really a large) dual ported (modern LEGO) piston. Al
large piston has two cavities: one for compression and the other for
contraction. All we need to make each of these cavities usable for as pumps are
a set of check valves. LEGO used to make these "pneumatic distribution bricks
with check valve" that provides exactly what we need.
By hooking the expansion port of a large piston to the center port of the
distribution brick, we make the dual ported piston into a single action pump.
By hooking the contraction port of a large piston to a second distribution
brick, we create a second single action pump using the same physical pneumatic
piston.
Now we have a dual action pump, meaning we get two compressions per single cycle
of compression/expansion of the piston. The two compressions are 180 degrees
out of phase. We effectively get two pumps out of the structural volume of one
piston (plus two 2x4 bricks).
By using a second large dual action pump (i.e. second piston+two distribution
bricks), with compression/expansion of the piston offset by 90 degrees from the
first, we get a total of four compressions with one complete rotation of the 40T
gear.
I've very successfully used two RC racer motors each with their own battery box,
turning 8T gears driving the 40T gears, driving two dual acting pumps, to make a
powerful compressor. My pneumatic inchworm requires a lot of pressure *and*
volume to function. Using the above compressor (with only three of the four
distribution bricks (more on the way)), inchworm ran almost as fast as I can do
running hand pumps by hand. This is the first time I've ever had a compressor
that came close to competing with my muscles.
I've got three more distribution bricks coming from bricklink..... Once I get
those, I'm going to try three dual acting pumps offset by 120 degrees (or 60
degrees between any two compressions), and see if the RC racer motors are up to
the task.
I sure wish LEGO still made those distribution bricks. I voted for them in
Jake's survey.
I'll let you know how things go with the three dual-pump compressor once I get
the needed parts.
Kevin
|
|
Message has 2 Replies:
 | | Re: SSClagorpion - Compressor
|
| (...) First of all, no. I have nothing interesting in the world of compressors. I got sidetracked building an all mechanical walker that has no gravity well. I think it's a cool idea, which I may revisit later, but I think this design requires more (...) (21 years ago, 17-May-04, to lugnet.technic)
| | | Re: SSClagorpion - Compressor
|
| (...) Airflow is what causes this. The two pumps both hook up to the same size hoses with the same size nipples, but an identical length of stroke on the larger pump is pushing quite a bit more air than the small pump. There's a limit to how fast (...) (21 years ago, 17-May-04, to lugnet.technic)
|
Message is in Reply To:
| | Re: SSClagorpion - Compressor
|
| ok, so we have about 200 messages in this thread. I'm going to start changing the subject to actually include the... subject. But still keep SSClagorpion, for those who chose to skip the thread. Kevin was talking about a compressor based on an RC (...) (21 years ago, 13-May-04, to lugnet.technic)
|
300 Messages in This Thread:
(Inline display suppressed due to large size. Click Dots below to view.)
- 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
|
|
|
|
