Alt-BEAM Archive
Message #10576
To: beam@sgiblab.sgi.com
From: Bruce Robinson Bruce_Robinson@telus.net
Date: Mon, 21 Feb 2000 11:55:11 -0800
Subject: [alt-beam] Re: BEAM: Tendency toward miniaturization
Timothy Flytch wrote:
> But I do question your scaling practice ...
> ... Yes it is true that not the same
> materials are use but you get the idea???
You're quite right about the relative weights and materials.
The point of the exercise is to do a systematic analysis and see what
the implications are, rather than just wing it. Scale up the robot (on
paper, in your head) or scale it down. Keep everything else the same.
Then see what happens.
Scaling up in this manner gives you problems, some of which I listed.
Stresses go up in proportion to the scale factor. Strength does not.
Sooner or later you hit a limit. Material costs impose a limit. Momentum
imposes a limit. So we are pretty much forced to begin changing the
design.
In order to make viable large robots, we are FORCED to use hollow
members and lattice structures and lighter materials. Our simple control
systems may become more complicated.
Scaling down in the same manner, these particular problems don't occur.
Stress vs. strength ratios improve, momentum issues disappear. We aren't
FORCED to deal with the same problems. We will, however, begin to run
into other problems. As dimensions decrease, mass is decreasing faster
than surface area and length. Air viscosity and velocity start to become
an issue.
So if we scale up, we are forced to deal with certain issues that get in
the way of what we're really trying to do -- build robots. And if we
scale down, we have entirely different issues to deal with.
It seems that by chance (or possibly not), we started at a point where
the complications of scaling up occur sooner than the compliciations of
scaling down. So a lot of research has been moving to the smaller side,
as Jean noted.
This type of mental exercise can also jump-start design efforts. For
example, if you want a robot to clean up a nuclear waste site, you need
to worry about radiation. It can affect your electronics and even your
materials. Less surface area per unit of volume is an advantage in this
case. So you mignt start by planning for a few large robots rather than
thousands of tiny ones. On the other hand, you can also predict some of
the design issues for large robots (if you take the time to think it
out).
Knowing that small robots might not last as long, you can start to make
LOGICAL comparisons, instead of emotional ones. Big robots = higher
survivability & more difficulty to build; small robots = low
survivability & (possibly) less difficulty to build. The questions then
tend to focus on issues of survivability and disposibility (which, if
the nuclear industry had paid attention to them, would have had a
radical effect on nuclear power plant design and policies).
So that's what my arguement was about ... a systematic analysis. What I
described is merely one way to go about it. I gave a mechanical approach
because that's what I happen to know best. Other's will have different
approaches. They are all viable, as long as they are systematic and
based on fundamental facts. Often the best answers come when different
approaches are used and then the results are compared.
Regards,
Bruce
10577 Mon, 21 Feb 2000 11:50:46 -0800 (PST) [alt-beam] Arrgghhh!!! beam@sgiblab.sgi.com Max Inggal today I got my shipment of parts. I ordered enough to
make 6 or 7 photovores. But(there's always a but)it
seems Digi-key has run out of the 4700uf caps I
wanted. They won't be getting any till May. This
conpletely sucks!! All I have parts for is maybe a 4
segmented 3 motor snakebot. Anyone out there know
what other 4700uf caps are good for photovores that I
can get from Digi-key.
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com
10578 Mon, 21 Feb 2000 14:20:59 -0600 [alt-beam] Re: More solenoid junk beam@sgiblab.sgi.com "Paul Atkinson" I mentioned the shutter release solenoids. The disposable cameras I've dissected
don't have them. I took apart a couple of broken 35mm autofocus point-n-shoot
types and that's where I found them. If you like I'll try to scan the couple I
have so you can get an idea of the size. I don't know where you could find them
other than in a similar camera.
Paul
BUDSCOTT@aol.com wrote:
> Right so the solenoids do actually work in my circuit. its just that even
> thought they are three volt solenoids, my circuit just doesn't have the balls
> to drive it. i was thinking about what somebody said earlier, camera shutter
> release solenoids. has anybody tried these, or even seen one. i was wondering
> if they were in disposable cameras. cause then i could just walk over to the
> photo place and ask if they had any empty disposable cameras that i could get
> parts from. these would definitly be more efficient than my surplus store
> ones. thanks for any response
>
> -Spencer
Home