Alt-BEAM Archive
Message #08971
To: beam@sgiblab.sgi.com
From: "Ben Hitchcock" ben@wollongong.apana.org.au
Date: Mon, 10 Jan 2000 15:14:30 +1100
Subject: [alt-beam] New(?) FLED variant
--MS_Mac_OE_3030362070_687663_MIME_Part
content-transfer-encoding: 7bit
All,
I have been experimenting with the FLED Solar Engine for the past couple of
days, and I've improved it a little bit. Gone are the days of watching your
FLED blinking away precious energy and making your SE lock up. Attached is
a .gif of the new circuit.
What's more, you can now use this FLED SE as the basis for a photopopper -
no more chasing exotic 1381's and ultra-efficient motors!
Full details of my popper are available at:
http://wollongong.apana.org.au/~ben/fred/
I just realised that the last attachment I sent was encoded in some weird
"AppleDouble" format. This one is sent as a UUencode. Please let me know
if you want the attachment(s) sent again, maybe in MIME this time. (Damn
Outlook Express!)
Ben
--MS_Mac_OE_3030362070_687663_MIME_Part
Attachment: schematic.gif
end
--MS_Mac_OE_3030362070_687663_MIME_Part--
8972 Sun, 09 Jan 2000 21:34:01 -0700 [alt-beam] Re: New(?) FLED variant beam@sgiblab.sgi.com Senior Very interesting! Do you know how it compares to a 1381 photopopper or
SE?
-Kyle
Ben Hitchcock wrote:
>
> All,
>
> I have been experimenting with the FLED Solar Engine for the past couple of
> days, and I've improved it a little bit. Gone are the days of watching your
> FLED blinking away precious energy and making your SE lock up. Attached is
> a .gif of the new circuit.
>
> What's more, you can now use this FLED SE as the basis for a photopopper -
> no more chasing exotic 1381's and ultra-efficient motors!
>
> Full details of my popper are available at:
> http://wollongong.apana.org.au/~ben/fred/
>
> I just realised that the last attachment I sent was encoded in some weird
> "AppleDouble" format. This one is sent as a UUencode. Please let me know
> if you want the attachment(s) sent again, maybe in MIME this time. (Damn
> Outlook Express!)
>
> Ben
>
> ------------------------------------------------------------------------
>
> Name: schematic.gif
> schematic.gif Type: GIF Image (image/gif)
> Encoding: x-uuencode
8973 Mon, 10 Jan 2000 00:38:12 -0500 [alt-beam] parts source beam@sgiblab.sgi.com Jonathan D Rogers new parts source:
Marlin P. Jones & Associates, Inc.
www.mpja.com
They have pager motors which I think are Namiki, they seem to have the
same specs, though they are surplus so they have weights attached.
However, they're only $3.95 each!
That's the best price I've seen! And it's not that hard to get the
weights off, is it?
They also have a very cheap panel meter($1.95) that I suppose you could
use the coil from like Ben is talking about.
They also have a 100RPM 6VDC gearmotor that looks like it might be good
for a walker for $6.95. Though I've never built one so I wouldn't know.
They have a rather unusual assortment of parts that should be of interest
to any experimenter, and good prices to boot.
Check 'em out!
Jonathan
8974 Sun, 09 Jan 2000 22:04:35 -0700 [alt-beam] Re: BIPEDS Was: Pressure Sensors beam@sgiblab.sgi.com Senior Bruce Robinson wrote:
> I imagine there is a fair amount of dynamic balance involved. A quote
> from their web site (keep in mind that Honda engineers studied walking
> motion for a year before they even started conceptual design on these
> robots) ...
>
> "We have revealed that no significant effect is caused to walking
> even without toes. More important support is ensured by base
> sections of the toes, i.e., balls of feet, and joint areas" ++
>
> So they decided not to have toes. That immediately limits the type of
> walking motion they can develop.
If they say so, I suppose their is. I didn't see it in the videos but
it's not VERY noticable with a human either. I disagree with what they
say about toes. I find myself using them a lot, especially with my shoes
off, to give a tiny extra push right before I lift it off the groun when
I'm walking. But I can't say I'm like everyone else! Do you find the
same?
> More to the point ...
>
> "A human eases the impact of walking with a combination between
> structures and functions of movement. ...
> ... Although the robot must feature similar shock-absorption
> mechanisms, the structural measure was not welcome because it
> might deteriorate the robot's stability. Impact absorption was thus
> ensured through precise control of each component." ++
> So Honda engineers opted to avoid the extra complications and control
> the limb motion to minimize shock. This inherently limits the types of
> motion the robot can achieve.
Good idea :) I notice we tend to aid ourselves in shock absorbing with
padded shoes and soft carpet. If it walkd with static balance then
padding is not needed, so I guess their need for shock absorption proves
that P3 walks with dynamic balance (IE it is off balance and falling at
one point).
> Here's where pressure sensors come into the picture. Many of the
> suggestions (including mine) involve some kind of compressive foam --
> which is a spring. Unless this foam is very thin, you will get
> significant motion changes as the weight distribution shifts. There will
> be a magnifying effect as well: divide the height of your robot by the
> narrowest dimension of a foot ... this is how much any deflection of the
> pressure sensors will be magnified by.
Ah very true! Did you see Wilf's post? His pressure sensors seem like
they wouldn't cause this problem too much because theydon't compress a
heck of a lot as compared to foam. Also I don't beleive this would be
such a problem with our small, lighter bipeds.
> By the way, a biped walker is not that difficult to construct. It has
> been done with two servos and mechanical linkages. Needless to say, it
> doesn't look very humanoid, and it sure can't handle uneven terrain.
Also true. I read through Hondas whole page, and am a bit more
impressed! I'm still hoping to prove them wrong :)
I came up with some notes while watching myself walk in the mirror for
hours (Make sure you're home alone if you do this...):
1 - A lot of people seem to think we walk with our feet a shoulder
length width apart. I think this is why Hondas bot wobbles from side to
side so much. It is much more efficient, however, if the feet are
brought closer, less weight shifting is needed. I tend to walk that way.
2 - That just about sums it up!
-Kyle
8975 Sun, 09 Jan 2000 22:34:52 PST [alt-beam] Re: BIPEDS Was: Pressure Sensors beam@sgiblab.sgi.com "Timothy Flytcher" >
>Good idea :) I notice we tend to aid ourselves in shock absorbing with
>padded shoes and soft carpet. If it walkd with static balance then
>padding is not needed, so I guess their need for shock absorption proves
>that P3 walks with dynamic balance (IE it is off balance and falling at
>one point).
>-Kyle
I don't see how you get this assumption??? After all a car has shock
absorbers but is never "off balance"???
Timothy...
______________________________________________________
8976 Monday, 10 January 2000 2:50 Photovore parts beam@sgiblab.sgi.com Jonathan D Rogers
>Hey, I'm about to build my first photovore, and I wanted to compliment
>Chiu-Yuan Fang on his excellent BEAM site, especially the section on
>building a free-form photovore.
>I assume he's on this mailing list...
>Anyway, it's been very helpful and I love the free-form design. Very
>cool.
>Does anyone have any suggestions on how to mount those cool touch sensor
>feeler things that Solarbotics sells to the design? I mean, it looks like
>you could glue them to the cap...but they're not in the circuit...
>
>Jonathan
Home