RTV2
A Photopopper Turbot
sort a thingy
-Based on Flippers success/failure, I pushed "RTV2", a
work in process.
The drive behind Flipper and RTV2 was that I wanted to get away from the wheel (how many wheels do you see in Nature?), and I wanted something that could almost climb.
It just so happens that both desires could be achieved
with a motor and leg sealed together into a flagella like limb. Sealing the motor and leg all
together for protection from the environment comes close to achieving both desired goals. Having
the leg inside the same seal of the motor eliminates
exposure of the motor thought the drive shaft joint, but of course your leg must be free to move within the
seal, and the seal must be strong enough for the bot
to walk on.
Learning a hard lesson on micro-voids and moisture
from Flipper, my second attempt at a climbing photopopper was to be a pair of photopoppers (as in twin bots) similar to Flipper; however, with the motors available, I
could hardly get them to move the mass of the bot body. So, I up sized the motors
to Bgmicro Lenz motors and joined the two photopoppers (total of four solar engines)
into the single, slow but powerful chassis (I only had two lens motors to spare). I run one
photopopper on 40000uF, and the other on 1F (I had to
use two 1F 2.5V gold caps to obtain the trigger
voltage of the 1381J).
With the 2 X photopopper configuration functioning as intended, I formed an aluminum chasse for the BGmicro lens motors. I sized the legs, by first using an extra long piece of aluminum and trimming it back. To determine how long the leg could be, I put one leg back, and let the other leg pick the bot up onto a step. If the motors winds back, or the SE's can't pick the bot up with their fire power then the legs are too long. One thing to keep an eye on is the wind-back, and
by this I mean, after a SE has fired and pick up the
bot, the weight will force the leg back around. This
will not be a problem for a worm gear, but I have not
found a pair of worm gears to use yet, so it's just
something to look out for. It takes a little time, but the earlier you find a problem the less likely you are to walk away :-)
Don't skip on the testing.
If the geared motor driven by your selected circuit
can not pick up an equivalent weight before assembly,
then there in no way it will be able to do so after
its assembled with the added weight of sensors, cells and an environmental seal.
With the mechanics selected and leg sized, I began the environmentally sealed leg construction. I formed an arc from a piece of stainless steel rod (1/16 gauge welding rod, or heavy gauge stainless steel guitar string). The end of the arc, should not exceed the leg length previously determined. Then I crimped the leg onto the motor.
To do this I found a tube that allowed motor shaft and
rod to just fit (some machining required), and crimped
down the tube with both leg and motor in place. I
attached a collar around the crimped leg joint directly to the gearbox with epoxy. This collar is the stationary point of the environmentally sealed leg to the chasse. I used a stainless steel technical pencil cap as a foot and
joined the foot to the collar with a surgical elastic
tube (you know the old brown tubes you can make a
sling shot with). The rod is then free to rotate within the tube of the environmental seal and the seal remains constant throughout the rotation of the leg (pretty clever Eh :-) Other construction comments:
So now this cheap little bot
has four photodiode eyes ($1.5/ea), four solar engines
($2/ea), uses four 24X22 solar cells ($4/ea), four
10000uF caps ($3/ea), two 1F caps ($6.5/ea), two
Bgmicro lens motors ($10/ea), 1 pack RTV ($6), and
Stainless steel flagella ($4). Grand total: $85
bucks… and he only moves about 1ft/day and can handle
a 1.5" step. Well humph… I knew there was a good
reason I didn't have twenty of these things crawling
around.
Problems:
Flipper's page.
To test the 2 X photopopper configuration with a little more speed (the BGmicro motors were very slow to trouble-shoot with), I built a quick chasse out of an old tooth-brush (of all things for a rudder) and a couple of motors from CD players. I used the large pulley as the wheel with good effect, but had hoped it could roll over something substantial which was not the case. I believe this to be one of the cheapest mobile configurations possible, given the high availability of CD players (BGmicro has the CD player innards for $1 each).
You can see a panorama GIF of this configuration here.
Then I covered the whole thing with
a thin coat of RTV. This is just one way to do it. One thing to keep in mind when selecting the external tube for the leg is that the force required to bend the tube will be the minimum constant load on the motor. It is extra work for the motor, so select the most pliable, or softest tube you can find for this job.
Although I sealed up the
photopopper circuits once each was completed, I left all the other wiring
exposed, such as: caps, cells, eyes, and sensors.
These are under a splashguard, or shell, but are still
mostly exposed to the elements. I did this for two
reasons. 1) I may need to pillage some of their
parts, and for now they are readily accessible. 2) I
wanted to see how bad they'd corrode, and after eight
months in the back yard I can see no visible
corrosion, just a little dirt from rain splashes.
RTV seal failed at the collar gearbox interface after
the collar gearbox epoxy failed (clean parts are
important for good glue attachment :-). One of the
motors to leg crimps failed due to over crimping (but
it could have been miss handled by a two year old),
need to be prepared.
Grass and twigs constantly tangle in the legs. I'm
not sure how to combat this problem.
Regarding protection to the environment. Seal it up
tight, or leave it open. Sealing it up will allow the
opportunity for micro void corrosion. Leaving it open
to the elements allows some corrosion over time, but
also allows a rapid escape for moisture. Remember
that in all wet or all dry, corrosion is slow, it is
in the transition for one to the other where the most
corrosion will occur.
And after all that, just look at him go.
Last updated:
02/16/03
Back to
my BEAM page.
Home.
Contact
Marty at mastervulk@yahoo.com
VEnterprise