Alt-BEAM Archive

Message #09138



To: beam@sgiblab.sgi.com
From: Richard Piotter richfile@rconnect.com
Date: Thu, 13 Jan 2000 09:50:19 -0600
Subject: [alt-beam] Re: microcore experimentation kit


What's a projected total cost? At $5, I'd be limited to small
quantities, say 10 to 20 max, but if they were around $2.50, I wouldn't
have a problem saving up for 30 to 40 or so. If they were more
expensive, it'd be more dificult to purchase large quantities.
Unfortunately, I'll have to wait for them, whatever the cost, till I
build up some spare cash. Anyhow, I think the demand would be
impressive. Wouldn't supprise me if Tilden started making them for
himself if they fared successfuly on the list! It's simply the next
logical step in expirimentation. It's too dificult to form complex
pathways of neurons when you are limited by the 6 NV core's physical
orientation. I could also imagine these being made into visual NV/Nu
demos. You can arrange the neurons according to a pattern that shows the
pathways for demonstration purposes. You could form 3D circuits much
more easily. Board design would then be nothing more than having the
power supplies and the pathways for signals. Well, one more thing. you
realy need to include plugs for both capacitor and resitor. When
switched around the Nv becomes an Nu. Realy a requirement for expirimentation.

I'm looking forward to such a thing. It would greatly reduce the
dificulty in BEAM expirimentation.

I'd like to see a variable resistor on the board, but if there is one,
and it's soldered, you'll need to offer two versions of the board, Nv
and Nu. Also, not everyone uses the more wide spread negative supply
connection for the resistor. I came up with a good chip pin out.
__________
| ,--. |
{+) o|-. o `--|o Out
| `-/ \--. |
| /___\ | |
Bias o|-. | ,-+-|o Polarity
| `o-+ o | |
| | | |
In o|-o `o `-|o (-)
|__________|

If you need to use it with negative polarity like most people do, then
you just tie polarity and (-). If you want positive, you tie it accross
to the (+). Bias can be driven fron the output of other neurons or from
other sources. Input is on the same side, as both can recieve inputs.
Out is at the top corner. Chaining these inline wouldn't be a problem at all!

(+)-+-----------+-----------+----------+
| | | |
| ___ | ___ | ___ | ___
`-| |---o `-| |---o `-| |---o `-| |---o
| | | | | | | | | | | |
-| |-. | -| |-. | -| |-. | -| |-. |
| | | | | | | | | | | | | | | |
.-------|___|-+ `---|___|-+ `---|___|-+ `---|___|-+ |
| | | | | |
| (-)---------+-----------+-----------+-----------+ |
| |
`---------------------------------------------------'

simple microcore out of these things. Easy to wire up, and easy to expand.



> David Perry wrote:
>
> Now that this has been raised a few times i'd like to go into it
> further.
>
> Isn't it a pain wiring up microcores? Using a 74hc14 chip is very
> cumbersome.
>
> What if (as someone suggested) i create a small circuit board complete
> with a single schmitt inverter and either a plug to put in your own
> resistor or an adjustable pot built in. Plus a plug to add your own
> hysteris. A SMD LED is on board. It would connect to a breadboard (or
> protoboard) with some PCB pins, a little like the BASIC STAMP. With
> this you could easily grab a few, plug them in, connect them in your
> own pattern an you have a microcore. Would people be intrested in such
> a thing? I could start a production of them. Keep in mind though they
> may be rather expensive compared to buying a 17HC14 and individual
> parts. This is just because the parts themselves are rather expensive.
> (A single schmit inverter is 68 cents).
>
> David

--


Richard Piotter The Richfiles Robotics & TI web page:
richfile@rconnect.com http://richfiles.calc.org

-- Make Money by Simply Surfing the Net or responding to E-Mail!!!
-- Click below!!!

http://www.alladvantage.com/go.asp?refid=ATL147
http://www.spedia.net/cgi-bin/dir/tz.cgi?run=show_svc&fl=8&vid=329630



9139 Thu, 13 Jan 2000 10:02:43 -0800 [alt-beam] Re: BEAM LEGO was microcore experimentation kit "'beam@sgiblab.sgi.com'" Wilf Rigter What about BEAM LEGO?

For example, put the Nv complete with trimpot and LED into a suitable block
with GND,V+, INPUT, OUTPUT, BIAS POINT contacts in such a way that the
blocks can be put side by side and interconnections made with a SOLARCELL,
NiCAD, SUPERCAP, PD SENSOR, DRIVER and MOTOR block. The default contact
arrangement should allow standard uCore walker to be assembled in BEAM
PLUG'nPLAY fashion. A JUMPER and PROTOTYPE block can be used to configure
experimental BEAM LEGO Bots. In this way standard BICORE, PHOTOVORE and
other BEAM bots can be bundled into standard kits.

So what do you all think?

wilf


> -----Original Message-----
> From: Bruce Robinson [SMTP:Bruce_Robinson@telus.net]
> Sent: Wednesday, January 12, 2000 11:46 PM
> To: beam@sgiblab.sgi.com
> Subject: Re: microcore experimentation kit
>
> > David Perry wrote:
> >
> > What if (as someone suggested) i reate a small circuit board complete
> > with a single schmitt inverter and either a plug to put in your own
> > resistor or an adjustable pot built in.
>
> So, what exactly are you gaining, David? Just MORE stuff to plug in.
>
> Why not include the pot (make sure it isn't one of those 50-turn-life
> things) AND the capacitor. Provide 4 connection points:
> - capacitor input (Nv input).
> - pot input (Nu input).
> - inverter input (so you can inject or kill pulses.
> - inverter output (obviously).
> And power & ground, of course. That's a six-pin board.
>
> You may want two or three versions with different size capacitors.
>
> But if you're going to do that, why not use a hex surface mount and
> supply six circuits similar to the above? That gives you 26 pins.
> Perhaps you could size the board to plug into a 28 pin socket.
>
> Hmm ... Is this really getting any easier? :)
>
> Bruce



9140 Thu, 13 Jan 2000 20:29:07 CET [alt-beam] Re: BEAM LEGO was microcore experimentation kit beam@sgiblab.sgi.com "johannes urke" sounds pretty nifty wilf.
and by the way will a in4001 or something diode work with the d1 solarengine
you made?


>From: Wilf Rigter
>Reply-To: beam@sgiblab.sgi.com
>To: "'beam@sgiblab.sgi.com'"
>Subject: BEAM LEGO was RE: microcore experimentation kit
>Date: Thu, 13 Jan 2000 10:02:43 -0800
>
>What about BEAM LEGO?
>
>For example, put the Nv complete with trimpot and LED into a suitable block
>with GND,V+, INPUT, OUTPUT, BIAS POINT contacts in such a way that the
>blocks can be put side by side and interconnections made with a SOLARCELL,
>NiCAD, SUPERCAP, PD SENSOR, DRIVER and MOTOR block. The default contact
>arrangement should allow standard uCore walker to be assembled in BEAM
>PLUG'nPLAY fashion. A JUMPER and PROTOTYPE block can be used to configure
>experimental BEAM LEGO Bots. In this way standard BICORE, PHOTOVORE and
>other BEAM bots can be bundled into standard kits.
>
>So what do you all think?
>
>wilf
>
>
> > -----Original Message-----
> > From: Bruce Robinson [SMTP:Bruce_Robinson@telus.net]
> > Sent: Wednesday, January 12, 2000 11:46 PM
> > To: beam@sgiblab.sgi.com
> > Subject: Re: microcore experimentation kit
> >
> > > David Perry wrote:
> > >
> > > What if (as someone suggested) i reate a small circuit board complete
> > > with a single schmitt inverter and either a plug to put in your own
> > > resistor or an adjustable pot built in.
> >
> > So, what exactly are you gaining, David? Just MORE stuff to plug in.
> >
> > Why not include the pot (make sure it isn't one of those 50-turn-life
> > things) AND the capacitor. Provide 4 connection points:
> > - capacitor input (Nv input).
> > - pot input (Nu input).
> > - inverter input (so you can inject or kill pulses.
> > - inverter output (obviously).
> > And power & ground, of course. That's a six-pin board.
> >
> > You may want two or three versions with different size capacitors.
> >
> > But if you're going to do that, why not use a hex surface mount and
> > supply six circuits similar to the above? That gives you 26 pins.
> > Perhaps you could size the board to plug into a 28 pin socket.
> >
> > Hmm ... Is this really getting any easier? :)
> >
> > Bruce

______________________________________________________



9141 Thu, 13 Jan 2000 15:01:07 -0500 (EST) [alt-beam] Re: BEAM LEGO was microcore experimentation kit Wilf.Rigter@powertech.bc.ca "Mark W. Tilden" BEAM Lego is a good idea, but alas anything that smells of LEGO gets
a legal department up the wazzo. So caution, experimentation only.

There's another problem theough. Lego only assembles from bottom to
top, so it severely cramps the permutations possible for a successful Nv
design. Nv is best when freeformed in 3space, and if you make some Nv-lego
mockups out of post-its, you'll see what I mean.

Later.

markt.



9142 Thursday, January 13, 2000 2:28 AM Re: SV: BIPEDS Was: Pressure Sensors beam@sgiblab.sgi.com Evaristo Westplate
>Hi,
>
>Lets see if I follow you
>
>>if the ultimate purpose of the sensors is to 'level out' the forces on a
>>foot by activating motors to shift the center of gravity so the heel, toe,
>>left and right sides of the foot all have equal weight distributed on
them,
>
>Yes.
>
>>take for example just the heel/toe axis. as the heel strikes the ground
the
>>toe needs to be forced down to make contact until both the heel and toe
have
>>equal weight or force exerted on them.
>
>>so if the force on the heel is converted directly into mechanical force,
>>instead of electrical that is in turn converted back into mechanical via
>>motors/servos, like what happens when you connect two air/hydraulic
>>cylinder's ports together. i'll try to explain what i mean.
>>
>>if you have two identical air pistons, and connect air line between the
>>inlet port of one to the inlet port of the other, when you push one of the
>>pistons in air is forced out the inlet into the second piston, causing it
to
>>extend. disregarding friction for a moment...and substituting a
>>noncompressing fluid for the air, the two cylinders will try to equalize
the
>>forces between them.
>>
>>so depending on how they are coupled to the foot, as the heel strikes the
>>toe end is raised or lowered until the front back pressure is equalized.
>
>{I'm just thinking this over}
>
>So the cylinders are connected to the legs and the pistons to the foot.
>Pushing the heel will force the cylinder to the toe to extend. The foot
>will rotate and the angle of the leg remains fixed. So you need something
>to regulate the pressure so that the leg will change its angle. This could
>be a valve but what will control this valve?
>
>Lets see. Suppose the foot is flat on the ground and the leg (plus robot)
>is upright. Pressure in both cylinders are equal. Now the robot starts to
>fall slightly forwards. Pressure in the heel forces the toe down so
>pressure in the toe forces the heel down. If this is true then the robot
>will push itself forward to the ground. Hmm. Could you make a system that
>will push the toes when pressure is applied to the toes? Hmm.
>
>I don't know. Did I overlook something? Sounds to me you need something to
>control the pressure in the cylinders.
>
>See'ya
>
>
>
>
>Evaristo
>
>Gizmo homepage: http://www.crosswinds.net/~evaristo
>Biped photos : http://www.crosswinds.net/~evaristo/biped.html
>


Home