hey
everyone I thought I'd show you what
I got done on the cookie
machine this
weekend so check it out
let's see the
rotor just started off in
some random
position I'm gonna tear the
scale and on
my UI here I have this set
to dispense
approximately half a gram
with white
sugar so I'm gonna press
the dispense
button and the rotor will
spin into the
correct spot the white
sugar spot and
start dispensing sugar so
you can see
the readout on the scale
going up and if
all goes well we should end
up with
about half a gram so after
it stops the
scale keeps reading a
little bit further
and I think that's that's
partially due
to the filtering so I'm gonna
have to
make my program a little
bit more
adaptive but there you go
that's five
hundred and thirty
milligrams and the
target was 510 so I'd say
that's pretty
good
within 20 milligrams it
dispensed that
amount of sugar in in about
five seconds
or so one thing that I
didn't anticipate
was how slow it was
actually gonna come
out of here this is the the
salt station
and the target value for
that is like 50
milligrams of salt per
cookie about and
there's a small drill bit
inside here
and as the drill bit rotates
it works
like a screw auger and
basically you
know pulls the granular
material out and
I did the same thing for
sugar but for
some reason in my testing I
thought that
I having a larger drill bit
was going to
be enough like I basically
targeted
about five seconds to
dispense each one
of these ingredients total
but as it
turns out this thing only
dispenses
about 50 milligrams per
second and in an
average cookie there's
about 4 or 5
grams of granular sugar so
this is gonna
have to spin for like 20
seconds to get
enough sugar out but oh
well maybe I'll
you know grind away a
little bit more of
the drill bit or something
to make it to
make more volume in there
but anyway
this is this is a lot of
progress and so
I got a gray code scale
attached on to
the drum here and on the
back side there
is a sensor with for
optical reflective
sensors and that feeds
those four bits
of information in
to the propeller chip for
some reason I
originally thought it would
be great to
have absolute encoder
information but
this turned out not to be
very helpful
and I still may change my
mind but what
would be better is to have
one or two
index marks on the rotor
and then just
to have quadrature encoding
so that I
can have a fairly dense
position
information there and then
also pick up
the indices to figure out
where the
rotor actually is so that
would require
like a startup routine so
that when the
index passes when you first
turn the
thing on it has to find one
of the index
marks and then it knows
where the rotor
is just by keeping state
the benefit of
this absolute encoding
system is that I
was planning on the
computer not really
keeping state for the rotor
in theory
this thing could just look
at the rotor
no matter what's happened I
mean someone
could come in here and turn
this thing
and then you know the
computer wouldn't
really care because all it
does this
look at the the current
configuration of
bits from the optical
sensors and it
doesn't have to keep state
for the rotor
but as it turns out this
didn't work out
very well because these
these sections
on the rotor are so large
you don't know
if you're at the front or
the back of
that section and so to get
around this I
you know made the code
always approach
segments counterclockwise
so even if the
segment is behind where you
want to go
the program will move the
road or
counterclockwise and then
spin it very a
very tiny amount clockwise
to get it to
the right spot so stupidly
enough this
is basically assuming again
it's making
the assumption that the
rotor isn't
going to move after the
computer is done
moving it one reason to go
through all
this hassle is that there's
going to be
electrical cords going up
the middle of
the rotor to all the motors
on the on
the rotor so you can't spin
it around
and around because the
cables would be
twisted off so the computer
has to keep
track of where the rotor is
and make
sure that it doesn't go
around a full
revolution most of the
motors in this
project only have to go one
direction so
for example this motor this
motor this
one even these guys only have
to spin
one direction to dispense
the product so
for those I'm just going to
use either
simple one transistor motor
control
things
or maybe I'll try one of
those
Darlington dip package type
things the
only motor that has to go
back and forth
actually controls the rotor
itself and
for that I used my favorite
l62 o3h
bridge driver which is
seriously
seriously overkill for this
application
but I already had one of
these sitting
on a board and so I just
plugged it in
another little benefit is
that the
temperature is warm enough
now or I
don't have to put a heater
unit on the
butter dispenser so when I
started this
project in the winter I the
butter was
too solid to squeeze out
from that metal
syringe but luckily now the
temperature
is warm enough where I don't
have to
worry about that so I first
want a
problem solved there so
there's still a
fair bit left work left to
do a lot in
the GUI I'm going to set it
up so that
the thing will actually
give the user
instructions like okay the
the
ingredients have just been
measured out
you know mix these first
three
ingredients and then it
will dispense
the rest of them I think
I'm not gonna
do mixing on the unit to
start because
there's really just not
enough time
before Maker Faire and that
would
require a lot of mechanical
development
that I just don't have time
to do so
what it's going to do is
just dispense
the ingredients and then
the user will
take the mixing vessel and
you know
manually mix them and then
put the
vessel back down and it
will move to the
next station
I still don't have a
dispenser for the
baking soda and there still
needs to be
a fair bit more work done
on the brown
sugar and chocolate chip
dispenser I
liked your guys idea of
dispensing
molasses instead of brown
sugar so I
could just use all white sugar
maybe
even use this as like the
course white
dispenser and then this is
like the fine
white sugar dispenser that
actually
sounds like a pretty cool
idea since
this one is going to be
good for high
accuracy and you just take
too long to
dispense that much sugar
okay see you
next time bye
s
Posts
No comments:
Post a Comment