Alt-BEAM Archive

Message #06205



To: alt-beam@egroups.com
From: Benjamin Edward Hitchcock beh01@uow.edu.au
Date: Tue, 28 Sep 1999 07:28:52 +1000 (EST)
Subject: [alt-beam] master/slave bicore walker problems


I would try puttin a 0.1 uF cap across the bicore chip (74HCT240). This
cures a lot of woes due to motor noise.

So the capacitor goes from one power pin on the chip to the other power
pin. This smooths out the ripples in the supply that can wreak havoc with
the timing.

Ben

[Charset iso-8859-1 unsupported, filtering to ASCII...]
> I'm new, so please forgive my ignorance...
>
> I put together the master/slave bicore circuit (Using a 74HCT240)on
> beam-online. I have the motor leads running into a 74HCT245 to bump up the
> signal, and it's driving two R/C servos.
>
> When I turn it on, the bot will take two or three steps ok, then the back
> legs will turn about 30 degrees too far. In no time it's on it's side and
> helpless.
>
> Being my first walker, I'm lost. I tried putting in some mechanical stops,
> and they would just paralyze the thing. I read that the 74HCT's don't do so
> well in the feedback area. Could that be the problem? The HCT's are all I
> could find, so I'd like to make them work.
>
> I also tried changing various resistors, but I couldn't do much more than
> speed it up or slow it down.
>
> Please, impart your endless knowledge. I need to learn!
>
>
> ------------------------------------------------------------------------
>
> eGroups.com home: http://www.egroups.com/group/alt-beam
> http://www.egroups.com - Simplifying group communications
>
>
>
>


--
Nothing is foolproof to a sufficiently talented fool.



6206 Mon, 27 Sep 1999 18:52:09 -0400 [alt-beam] Re: master/slave bicore walker problems beam@sgiblab.sgi.com Richard Weait At 10:59 AM 9/27/99 -0500, "Phillip A. Ryals" wrote:
>I haven't tried springs OR pots. I thought about springs, but wasn't sure
>how to implement them. What's the general idea there?
>
>Curious... when I hold the bot in the air, the motors seem to be just fine.
>It's just when I put it on a surface when it starts freaking out. I assumed
>that meant it was getting feedback. Is there something else going on?

You may be asking too much from the '245. Sounds like all is
well until the motors (servos) are loaded. They draw more
current when loaded, and that may be over heating the '245. A
quick test is to put another '245 in parallel with the existing
one, and see if that helps. You can do this by stacking, without
solder, for a temporary test. (Make sure the power is off when
you line them up :-))



6207 Tue, 28 Sep 1999 06:20:50 -0400 [alt-beam] Re: master/slave bicore walker problems beam@sgiblab.sgi.com SG At 02:53 PM 9/27/99 EDT, you wrote:
>In a message dated 9/27/99 1:21:50 PM Eastern Daylight Time,
>phillip@ryals.com writes:
>Most of what is termed "feedback" is really the quirkiness of
>these coupled oscillators. Stop believing what was written over a >year or
>more ago. Most of it has been disproved in practice. True feedback can be
>achieved, but not with these setups.


my master-slave bicore seems to me truly has feed back!
slow or hold the master legs, more power goes to the slave set.
very obvious.

one thing the original guy might try is to MAKE SURE that the distance
between the contact point of the front legs and the rear legs are the same
(as measured out from the center point on the body)

one of my front legs was biased, but just because the back leg didn't have
the leverage to push it along. (due to the fact it was only 1/4" short one
one side!

-Sparky

Home