| |
In lugnet.technic, Kevin L. Clague wrote:
> In lugnet.technic, Mark Bellis wrote:
> > In lugnet.technic, Kevin L. Clague wrote:
> > > In lugnet.technic, Mark Bellis wrote:
> > >
> > > [SNIP]
> > >
> > > >
> > > > Here are the pneumatic gate circuits I promised you:
> > > > AND / OR / NAND / NOR gate:
> > > > http://www.brickshelf.com/cgi-bin/gallery.cgi?i=1346510
> > > >
> > > > XOR / XNOR gate:
> > > > http://www.brickshelf.com/cgi-bin/gallery.cgi?i=1346511
> > >
> > >
> > > I can't wait until they are viewable to compare AND gates. SSC1 ended up using
> > > my design of that. If mine differs, I'll post it too.
> >
> > (SNIP)
> >
> > Oops! I forgot to post the jpg links rather than the item links, since it's a
> > while since I posted stuff! That's what comes of doing too many exhibitions!
> >
> > The AND etc... gates are dual no-leak bidirectional gates that perform the
> > function of a full electronic logic gate. They're not designed to be for gating
> > four leg positions, for which I'd use 1 switch per leg, daisy-chained.
> > >
> > > Would you like to work together on pneumatic documentation styles for Brick
> > > Journal? I will be writing for Brick Journal both on CAD and MOCs.
> > >
> > > Kevin
>
> Mark,
> I did a lot of thinking about pneumatics in my long drive home over the last
> few days, and am going to try to share them here (yah, Steve, there I go giving
> away my competitive edge ;^).
>
> I started thinking about a piston and its ports. Each port can have pressure,
> or release. Since each piston has two ports, we can list all possible states at
> the port.
>
> Expand Contract
> Port Port
>
> release release - piston state remains unchanged, unless some outside
> force acts upon it.
> pressure release - piston expands, unless it was already expanded
> release pressure - piston contracts, unless it is already contracted
> pressure pressure - Dave Schilling's parlor trick, the piston expands
> slowly
>
> In the simplest of pneumatic circuits, a piston is always in either expand
> state (pressure/release), or contract state (release/pressure). This means that
> a piston is always under pressure on one port, but not pressure on both.
>
> The kinds of gates you have designed, in general provide either an expand
> state, or a contract state. These outputs are rather like a differential pair.
> You have AND, NAND, OR and NOR gates that are six switch implementation of said
> gates. You call your implementation "no-leak". I studied your design and
> realized it would work very well for hydraulic LEGO.
>
> I'm not sure I understand why you feel "no-leak" is needed in pneumatic
> circuits.
>
> Lets start with the simplest gates. These gates have a single pressure
> output, unlike the differentiual gates you have describe. These are the gates I
> use to synchronize leg up and down movement. Here is a picture of the sinplest
> AND gate I know. Mark Tarrabain is the one who taught me this.
>
> http://www.brickshelf.com/gallery/kclague/pneumatics-tutorial/and_1.jpg
>
> It is a single piston and a single switch combined to give us pressure at the
> single output if both inputs have pressure. The pressure comes out the left
> port of the switch.
>
> After Mark taught me that about the single switch, I decided to create a
> single pressure output OR gate:
>
> http://www.brickshelf.com/gallery/kclague/pneumatics-tutorial/or_1.jpg
>
> The single pressure output comes from the center switch and is named Q. The
> input A controls the piston, and input B goes into the rightmost switch.
>
> I also designed an XOR gate with single pressure output:
>
> http://www.brickshelf.com/gallery/kclague/pneumatics-tutorial/xor_1.jpg
>
> Input A controls the piston and also goes to the leftmost switch. Input B
> goes to the rightmost switch. The output Q comes out the center switch. Notice
> that OR and XOR are physically identical, except for inputs to the leftmost
> switch.
>
> My pneumatic inchworm, pedmatic, quadrapeds, and hexapods are built
> exclusively using single pressure (single switch) AND gates.
>
> On my pneumatic adding machines, the gate technology I used (oh, so outdated
> ;^) provided two pressure outputs, a differential pair output.
>
> I provide my four switch version of a dual pressure output AND gate here:
>
> http://www.brickshelf.com/gallery/kclague/pneumatics-tutorial/and_2.jpg
>
> The picture contains the original single pressure output and_1 type gate, with
> three switches added to create the NOT Q output. Switch S1 releases pressure to
> the outside world, when S1's handle is to the left. S2, S3 and S4 are hooked up
> tight as a drum, and depend on the inputs B and NOT B, to release pressure to
> the outside world. I used this dual output pressure AND gate in SSC1 for the
> central timing circuit. It worked flawlessly at BrickFest. Does my AND gate
> above qualify for by your definition of "no-leak", if not, why not?
>
> I'm trying to guess what "no-leak" means, so I tried a third AND gate design,
> just in case.
>
> http://www.brickshelf.com/gallery/kclague/pneumatics-tutorial/and2nl.jpg
>
> I have two pressure output OR gate as well:
>
> http://www.brickshelf.com/gallery/kclague/pneumatics-tutorial/or_2.jpg
>
> Notice this OR gate only adds one switch to the single pressure OR gate.
>
> I have a two pressure XOR gate:
>
> http://www.brickshelf.com/gallery/kclague/pneumatics-tutorial/xor_2.jpg
>
> This is just like a single pressure XOR gate, but with one more switch.
>
> http://www.brickshelf.com/gallery/kclague/pneumatics-tutorial/xor_2.jpg
>
>
> Can you explain what you mean by "no-leak", and why it is important in
> pneumatic design?
>
> Thanks,
> Kevin
>
>
>
>
>
>
>
>
>
> 1. Differential output (for lack of a better term) - The gate has two inputs,
> each with a differential pair. In each pair, one
> >
> > Sounds good. I've applied standard electronic circuit diagram techniques to my
> > pneumatic drawings, such as avoiding too many crossovers. I hope my use of flex
> > tubing also helps, especially where tubes cross. Having said that, the gate
> > circuits have a lot of overlaps because they are complex circuits with multiple
> > parallel paths. Have fun tracing the air paths in 1-0 or 0-1 settings!
> >
> > Do you have Visio at all? If so I could send you a template of the parts. It
> > makes it nice and quick to draw complex circuits, without having to worry about
> > objects snapping to studs etc... I can just draw a new part if I need one. I
> > drew the hand pump today for basic circuits that I'll post soon, not being used
> > to using one for my MOCs!
> >
> > Perhaps you could add your walkers to this BrickWiki page:
> > http://brickwiki.zapto.org/index.php/Pneumatics
> >
> > Mark
Kevin,
I think "differential output" is what I really meant, though I called it "no
leak" because the supply never leaks, one output or the other is always
pressurised and there is never a state where both outputs leak. There is also
only one place where the leaking output leaks on the XOR and XNOR gates, from
the single switch - subsequent parity stages have no leaks at all. The idea
behind it was to have full control of both outputs in all possible states, since
that is also the goal of a finite state machine. It suits me being a control
freak :-)
I use differential output gates because the output is suitable for driving the
input of the next gate, so these circuits are building blocks. To invert the
output, swap the output hoses. To invert the input, swap the input hoses.
My differential gates are suitable for driving load-bearing cylinders because
there will be minimal pressure loss in the gate when the inputs change to
another state that has the same output. That was another requirement of the
design, which is helped by using as much flex tube as possible. The
load-bearing functionality was the main reason for having no leaks.
Another thing was to avoid any gate being transparent. This means, similarly to
electronic flip-flops, that the input can't go through to the output without a
proper change of state. In a transparent electronic D-type flip-flop, if the D
input changed, the Q and !Q outputs would change with it. In a non-transparent
flip-flop the change of Q and !Q has to wait for the data at D to be clocked
into the flip-flop, such that it changes state. This avoids logic races in
electronic flip-flops and it was also something I wanted to avoid in pneumatics,
in order to be sure of the next state of the system at all times and prevent
movement into an undesirable or lock-up state.
Another feature of my designs is that the input need not necesarily be a
pneumatic function. All that is required is something that moves a set of valve
switches, so it could be a motor or hand input. I've not yet tried adding
machines but I could make one on the same basis as a binary full adder from an
electronics book, using and AND gate and an XOR gate for each of two half adders
and and OR gate for the summation.
Don't blame me, I'm just an electronics engineer :-)
I do use single switch ANDing on enable lines, where I need only one pressure
and not a differential output. I use enable lines to interupt the air supply to
a set of cylinders, so the enable is a gated air supply to the middle port of a
switch, such that a cylinder can't move either way if it is not enabled. I
haven't yet applied an enable line to only one port of a cylinder because the
system might end up with either pressure on both ports or a leak on both ports,
both of which I want to avoid. For enable lines I apply the supply to the
middle switch port and take the output from one side with a stopper on the other
side. The supply must never leak.
On the first of your gates, are you feeding pressure A into the cylinder and
also through a B switch to the AB port?
Also, you seem to be addressing hose pressures as the objects that have a state.
I have always used valve positions as the objects that have "0" and "1" states.
This means that, assuming the centre port is pressurised, in state "0" the right
port is pressurised whilst the left port leaks, and vice versa.
I looked at your 4-switch dual pressure AND gate, assuming that the "B" input is
from another valve switch, but it appeared to cause both outputs to leak in 0-1
or 1-0 states, due to the mismatch between the two hoses from the B switch (into
B and !B ports) and he states of A and B. What parts are used that don't appear
in the diagram?
Solving the 0-1 and 1-0 mismatch was what pleased me most about my AND gate
design.
Mark
|
|
Message has 2 Replies:
Message is in Reply To:
32 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
|
|
|
|
