►
From YouTube: 3D Printer Extruder Design
Description
See https://wiki.opensourceecology.org/wiki/OSE_Machine_Design_Guide for links.
-----------------
What you see here at Open Source Ecology is an ambitious open source project for the common good. Join our development team:
http://opensourceecology.org/wiki/OSE_Developers
We run One Day Extreme Build workshops in many locations. Please put yourself on our map if you'd like us to host a workshop near you:
https://microfactory.opensourceecology.org/request-a-workshop/
Take a minute to subscribe to our email newsletter (updates, workshops, etc): http://bit.ly/1LtcM44
A
Okay,
so
take
two
today
is
the
session
on
the
design
of
3d
printer
extruders.
So
the
perspective
on
that
is
that
there's
two
main
items
on
our
side
is
the
ability
to
print
with
rubber,
which
is
thermoplastic
elastomer
rubber
is
printable
on
3d
printers
and
for
our
applications.
This
is
about
rubber
o-rings
up
to
tractor
tires
very
important
parts
of
the
modern
economy.
A
For
the
issue,
I
mean
accessibility
to
such
extruders
is
that
they
do
not
exist.
There
is
no
optimized
3,
millimeter
printing
extruder
and
in
3
millimeters
the
standard
large
filament,
and
we
also
want
to
go
larger,
they're,
printing,
larger
and
faster.
So
currently
we
can.
We
have
the
ability
to
buy
a
$250
extruder,
the
super
volcano
with
a
super
volcano
nozzle,
which
allows
us
to
print
about
20
pounds
per
day
of
plastic.
A
So
that's
the
industry
standard
state
of
art
that
can
be
done
with
today's
off-the-shelf
technology
and
I
mentioned
that
the
the
extruder
heater
block
costs
about
$110.
For
the
super
volcano
nozzle,
you
can
buy
a
small
heater
block
for
about
a
dollar,
but
if
you
want
to
print
10
times
faster
with
a
larger
one,
the
super
volcano
it'll
cost
you
$110.
So
there's
a
big
case
for
making
that
open
sourcing
that
and
making
it
ourselves
with
basic
machining
and
basic
basic
operations
and
we'll
show
you
an
example
of
a
simple
extruder
William.
A
So
we'll
start
with
going
to
the
3d
printer
extruder
design
manual-
and
let's
start
by
saying
by
talking
about
what
a
basic
extruder
is
made
of,
what
are
the
main
components
of
that
when
we
study
the
industry
standards
with
what
we
know
works
today,
and
how
can
we
build
upon
that
so
sharing
my
screen
now
with
the
remote
crew
and
people
here,
so
we
can
follow
what
I'm,
showing
so
general
design
features.
Let's
look
at
the
extruder
that
we're
actually
using
on
our
own
3d
printer
and
it's
called
the
Titan
Aero.
No,
not
that
one.
A
Components
of
the
3d
printer
extruder,
so
this
is
the
standard
that
we're
using
right.
Now
you
can
get
this
for
a
hundred
seventeen
dollars.
It's
not
cheap!
You
can
do
you
can
when
you
build
them
yourself,
you
can
do
much
better
than
then.
Therefore
the
question
becomes
well.
What
exactly
is
one?
Can
you
understand
what
enough
that
it's
a
simple
task
to
do,
and,
yes,
you
can,
if
you
know
what's
going
on
so,
let's
look
at
some
of
the
main
components.
You've
got
a
stepper
motor,
that's
driving
a
little
little
gear.
A
A
You've
got
the
the
stepper
motor,
that's
driving
the
drive,
drive
gear
that
feeds
filaments
down
through
the
system
and
through
to
the
nozzles,
so
you've
got
a
heater
block
at
the
bottom.
A
nozzle
at
the
bottom
you've
got
a
heater
element.
The
red
parts
going
into
the
the
block
you've
got
a
thermistor
that
detects
the
temperature.
It's
that
little
hole
back
there
in
this
design,
the
you
have
to
cool
after
the
heater
block,
there's
a
thermal
break
which
is
called
a
heat
break.
A
A
It
pushes
the
filament
below
it
through
to
the
nozzle
and
for
which
reason
the
the
tolerances
there
have
to
be
relatively
decent,
like
you
cannot
have
a
big
gap
where
the
the
the
filament
is
pushing
down
on
a
plastic
or
it
would
like
leak
around
it.
Think
of
a
syringe.
A
syringe
has
a
rubber
washer,
where
all
the
liquid
is
coming
out
to
the
desired
end,
not
back
washing
through
through
around
the
washer,
and
here
that
washer
is
your
filament.
A
The
your
filament
is
doing
the
pushing
it's
like
a
piston
so
and
you
want
that
to
be
relatively
tight
and
for
which
reason
these
are
somewhat
precise
devices
and
you
can
get
away
with
with
not
having
that
tightness
there,
but
at
that
point,
you'd
only
perhaps
be
able
to
print
very
slowly.
You
want
to
have
the
ability
to
push
nothing's
leaking
out
and
you're
pushing
with
large
force.
If
you
have
large
filaments
or
you
want
to
go
at
fast
rates,
you
want
the
push
and
no
leakage
around
the
filament.
A
So
when
you
get
to
performance
things
in
there
have
to
be
relatively
precise.
So
let's
look
at
a
another
image
of
an
extruder,
just
a
basic.
What
is
it
a
simple
diagram
images,
diagram
of
a
3d
printer
extruder?
So
here's
a
here's,
a
decent
one.
Let's
just
take
a
look
at
this
one
for
what
I've
just
described.
A
Let's
just
expand
this
and
just
take
a
look
at
this
this
picture,
so
it's
a
simple
diagram.
You've
got
a
stepper
motor
in
this
case
here.
What
you
have
is
a
gear
down
extruder
and
that's
actually
what
happens
within
the
e3d,
the
titan
arrow
III
D
Titan
arrows.
Is
this
one?
That's
what
we're
using
right
now
in
our
printers,
so
you've
got
a
you've,
got
a
a
small
gear
driving
a
larger
gear
which
then
feeds
the
filament.
So
you've
got
gear
reduction.
So
you
want
to
push
with
as
much
force
as
you
as
you
can.
A
A
This
kind
of
mechanism
is
quite
relevant,
so
you've
got
the
wire
feeding
down,
you've
got
a
heat
element
and
then
the
nozzle
and
the
thermistor,
because
you
need
to
detect
what
temperature
you
add,
you
have
to
know
what
what
to
set
it
to
your
software
takes
care
of
that.
But
you
have
to
have
feedback
to
know
what
temperature
you're
at
so
that's
a
generic
design
of
an
extruder
and
then
we
talk
about
there's
two
main
types
of
extruders:
one
one
is
direct
versus
Bowden.
A
So
this
is
the
generic
diagram
here
of
direct
drive
versus
Bowden
drive
and
direct
drive.
You've
got
the
gears
going
directly
into
the
heater
block
and
nozzle,
whereas
in
the
Bowden
style
you've
got
the
wire
feed,
that's
separated
from
the
heater
block
on
your
gantry,
so
the
the
wire
feed
the
filament
feed
can
be
mounted
anywhere
on
the
printer
farther
away
from
the
nozzle.
Now
you
can
guess
that
that
may
be
limited.
It's
got
advantages
and
disadvantages.
The
advantage
of
the
boat
and
setup
is,
then
that
that
the
tool
head
becomes
very
light.
A
You
just
have
to
carry
the
the
nozzle
pretty
much
the
heater
block
and
nozzle.
You
don't
have
to
carry
the
relatively
heavier
stepper
motor
and
all
the
other
drive
components.
Therefore,
you're
able
to
achieve
much
much
higher
speeds,
or
not
necessarily
that
you're
able
to
achieve
higher
speeds,
but
you
have
less
inertial
effects,
meaning
you've
got
more
stable
prints.
So
there's
less
lot
less
shaking.
As
you
reverse
the
registry,
you
can
go
much
faster
if
you
have
that
entire
extruder
on
a
gantry
like
we
do.
A
Your
frame
has
to
be
stiffer
to
accommodate
all
of
that.
But
the
advantage
of
the
direct
drive
is
that
you
have
a
very
short
path
between
the
feed
and
the
nozzle.
What
that
means
that,
for
example,
with
rubber
filaments,
you
can
do
well,
because,
if
you
think
about
the
remote
Bowden
extruder
setup,
if
you're
pushing
against
the
rubber
filament
that
thing's
gonna
crinkle
up
on
you,
it
does
have
a
guide
tube
but
you're
very
much
limited
to
what
you
couldn't
do
with
rubber
in
the
setup
of
a
boat
and
extruder,
especially
in
a
1.75
millimeter.
A
A
A
Absolutely
if
you
want
to
do
generic
brute
force
printing
with
a
large
nozzle
you're,
throwing
everything
in
there,
you're
gonna
get
a
mix
of
uncontrollable
qualities
which
may
be
acceptable
for
a
thing
like
outdoor
exterior
plastic
lumber,
which
doesn't
have
to
have
a
lot
of
precision
or
or
whatever.
But
typically,
if
you
want
any
engineered
control,
you
want
to
separate
your
trash
and
more
than
that,
even
though
what
kind
of
plastic
that
is,
there's
many
types
of
abs
or
PLA
blends.
A
When
you
think
about
3d
printing
filaments,
the
material
science
there,
you
can
have
a
blend
of
anything
with
anything
and
you
have
to
know
its
properties
if
you're
gonna
have
controllable
results.
Here,
for
example,
Sam
is
printing
directly
with
poly
propylene
I
believe
so
he
had
polypropylene
that
he
shredded
and
I
shred
it
up
and
put
it
into
the
system.
A
So
there
is
the
whole
material
science
aspect
of
that,
but
the
good
thing
about
it
is:
we
can't
experiment
a
lot
with
it
like,
for
example,
if
you
have
polypropylene,
which
typically
is
actually
quite
difficult
because
it
warps
a
lot,
you
can
mix
it
say
with
ABS
or
other
blends
and
that
that
goes
into
the
science
of
what
mixes,
with
what
there's
polar
and
nonpolar
and
less
polar.
When
you
talk
about
the
electronic
structure
of
these
plastics,
some
mix
really
well
others
totally,
don't
like
each
other.
A
If
you
have
one
that's
really
hard
to
print
with,
you
can
so-called
alloy
it
with
another
plastic
or
blended
with
other
plastics,
so
there's
a
whole
realm
of
unexplored
science
there
and
the
trick
to
that
would
be
well.
How
do
you
turn
the
waste
stream,
which
has
got
all
kinds
of
random
stuff
into
useable
plastic
I
mean
for
one.
You
do
have
labeled
like
PT
bottles,
HDPE,
any
milk,
jugs
and
so
forth.
A
So
you
can
know
some
of
what
you've
got
to
work
with,
but
there's
a
whole
whole
set
of
tricks
there
that
you
have
to
do
it's
the
whole
Production
Engineering
there
to
get
to
reliable
plastics,
and
that's
that
makes
it
very
difficult.
But
once
you
open
source
that
and
get
proven
protocols
of
how
do
you
do
it,
then
that
becomes
feasible?
And
but
that
takes
a
lot
of
work
and
it's
perfect
for
a
crowdsource
effort
because
say
a
plastic
plastic
in
one
country
may
be
different
from
plastic
in
another
country
and
so
forth.
A
In
the
direct
in
this
trash
extruder
yeah,
when
we
say
trash
yeah,
you
do
it's
I,
don't
know
mislead
to
say
that
oh
yeah
just
throw
anything
down
there
in
cases
where
you
have
a
blend.
That
has
so
many
different
things
and
it's
just
acceptable.
Yes,
you
can
do
that,
but
typically
you
want
to
separate
things
and
know
more
of
what
you're
doing
with
to
get
higher
higher
quality
results.
So
if
you
want
to
get
a
precision
part
for
a
3d
printed
calipers,
well,
you
want
to
know
exactly
what
you've
got
there.
A
You
don't
want
any
imperfections
in
your
print,
but
if
you're
printing
plastic
lumber,
you
can
throw
everything
in
the
kitchen
sink
in
there
and
it
may
work
depending,
which
is
what
you're,
starting
with
so
I'm
moving
on
here.
So
those
are
the
main
types
of
extruders
that
this
is
absolutely
open
source
here
there
are
open
source
versions
of
this,
but
nothing
is
well-defined,
well
well.
Refined,
I.
Think
Sam
is
doing
the
most
cutting-edge
work
on
this.
A
That
I've
seen
at
least
there's
a
lot
of
work
like
this
in
various
countries,
most
of
it
it's
like
partly
proprietary
and
stuff.
So
once
again
we
are
into
the
issue
of
people
saying
oh
yeah,
now
I
can
I'm
gonna
make
millions
and
I'm
gonna
hide
this,
because
this
is
too
good.
That's
a
typical
issue
with
open
source.
Once
you
get
something
that
actually
works,
maybe
99.9%
of
the
people
disappear
and
that's
a
that's
one
of
the
challenges
of
the
collaborative
literacy
in
our
world
today.
A
Ok,
so
moving
on
to
more
topics
here.
So,
let's
talk
about
designing
the
optimal
plastic
extruder,
so
we'll
go
to
a
page,
and
this
is
something
like
when
we
talk
about
an
optimized
plastic
extruder.
That
means
something
that
can
handle
both
regular
plastic
and
rubber
I.
Don't
see
any
reasons
why
you
cannot
optimize
for
both
in
one,
so
here's
what
what
I've
drawn
up
so
far
in
terms
of
what
optimization
would
look
like
and
it
looks
it
might
look
a
little
different
than
standard
one.
So
design
rationale.
A
A
If
you
want
the
higher
this
is.
This
is
now
for
larger,
faster,
better
extruders.
You
want
to
first
of
all
start
by
driving
from
both
sides.
Now,
maybe
you
even
want
to
do
a
second
set
of
drive
gears
to
get
double
that
drive
force
there,
because
the
thing
is
because
rock
now
plastic
is
soft,
it
will
wear
out
if
you
can
only
push
it
so
hard
with
a
set
of
teeth
before
you
actually
wear
it
out.
So
you
for
optimization.
A
You
definitely
want
to
start
with
double
sided,
draw
and
I'm
saying
this
with
respect
to
okay.
How
do
we
take
because,
for
example,
we've
built
one
here,
here's
an
extruder
that
we
built
for
five
dollars.
How
do
we
get
this
to
optimal
performance
so
start
with
dual
Drive?
Now
the
critical
aspect
of
an
extruder
and
and
so
yeah?
Let
me
get
into
the
the
distance
between
the
drive
and
the
heater
block.
You
want
that
distance
to
be
absolutely
the
shortest
possible.
A
So
if
you're
driving,
for
example,
rubber
filaments,
they
do
not
have
a
chance
to
kink
up
and
bind
up
before
they
get
out
the
heater
block.
So
here
you
have.
What
we
have
here
is
right
above
the
drive
gears
you
have,
it
goes
immediately
into
the
heatsink
and
in
fact
it's
kind
of
like
it's
got
this
upward
metal
part
where
it's
right
after
it's
driven
it
enters
the
cooling
block
and
gets
constrained,
so
it
doesn't
have
any
any
chance
to
wobble.
A
You
need
the
heatsink,
so
there's
our
heatsink
right
after
the
drive
gear,
then
after
that
you've
got
the
heater
block
and
the
goal
there
is.
You
are
absolutely
minimizing
as
much
as
possible,
but
with
a
caveat
that,
if
you
make
that
part,
this
heatsink
part
they're
too
thin,
you
may
not
get
enough
heat
rejection
to
cool
this
off.
You
might
just
the
heater
block
might
overpower
you,
so
this
gets
into
thermal
design.
How
thick
does
that
part
have
to
have
to
provide
enough
cooling?
A
What
you
have
here
is
conduction
to
this
other
part
of
the
heatsink,
where
you've
got
fans
blowing
on
it,
like
we
have
one
fan
on
a
current
extruder
blowing
on
their
heatsink,
but
here
I
drew
that
on
the
side.
So
why
am
I
drawing
this
on
a
side?
One
critical
aspect
here
is
maintenance.
I
want
to
leave
the
front
entirely
open
so
that
if
there
is
any
Jam
which
will
happen
from
time
to
time
and
hopefully
once
a
year
not
like
once
a
day,
then
you
can
maintain
it.
You
can
open
it
up.
A
A
One
so
like
this
one
here,
it's
got
an
open.
The
one
we
did
has
got
an
open
face,
but
that's
absolutely
unacceptable
from
an
efficiency
perspective.
Then
you're
becoming
a
servant
to
your
technology
as
opposed
to
doing
with
your
life.
What
you
need,
so
we
wanted
this
by
design.
You
want
to
open
an
open
front
face
design,
that's
why
the
heat
sinks
are
towards
the
side
and
those
ribs
or
whatever.
This
is
just
a
conceptual
design,
but
we're
taking
keeping
the
front
open.
A
Ok,
second,
second
part,
so
you
definitely
have
to
have
the
fans
for
cooling.
What
I
did
not
draw
here
is
a
print
cooling
fan
which
would
blow
under
the
nozzle
like
where
the,
where
the
filament
comes
out,
you're
cooling,
that
cooling
the
print
so
that
the
print
stabilizes
as
soon
as
the
the
layer
of
filament
is
deposited.
The
way
3d
printing
works
is
not
not
that
you're,
attaching
a
molten
piece
of
plastic
to
a
molten
bottom.
A
That
means
that
if
you
stop
your
print
for
a
day
and
everything
cools
off,
you
can
pick
right
up,
because
it's
the
new
filament
that
melts
and
fuses
into
the
layers
below
that's
how
that
works.
So
that's
called
fused
filament
deposition,
the
top
the
new
filament
fuses
into
what
your
your
existing
print
and
that's
an
important
concept
to
think
about,
because
that
kind
of
gets
you
to
understand
how
you
can
print
various
things.
A
Okay,
so
next
feature
here
is
I'm
using
bolts
here,
clamp
design,
I
told
you
about
the
clamp,
clamp
design
right
there
for
heavy-duty
bulldozers,
equivalent
of
cat-cat
v7,
and
here
we're
using
clamp
design
for
this
tiny
extruder.
Why
threads
are
tricky
threads?
Do
not
give
you
as
good
thermal
contact
as
metal
on
metal
contact,
I.
B
A
I'm
still
here,
we've
got
clamps.
The
clamps
are
the
little
bolts.
So
what
I'm
doing
here
so
there's
a
heat
break.
The
light
red
part
is
a
heat
break.
The
heat
break
is
the
part
that
that
connects
the
heater
block
to
the
cooler.
So,
basically
you
want
the
heater
block
to
be
hot
and
therefore
the
part
that
you're
connecting
to
the
rest
of
the
system
wants
to
be
as
thin
as
possible
to
have
the
least
conduction
and
thermal
design.
Conduction
is
much
stronger
than
convection
or
like
already
like
here,
radiation
conduction
is
like.
A
If
you
have
metal
connected
to
metal
that
that
conducts
fast.
If
you
think
about
something,
that's
you
know
a
hot
pot
on
a
stove,
it's
very
hot,
even
though
the
flame
is
on
the
bottom,
because
there's
conduction,
there's
pretty
rapid
conduction
conduction
is
what
you're
protecting
against
by
using
the
heat
break
being
as
small
as
possible.
Now,
what's
the
limit
to
small
the
smaller,
it
is
the
more
fragile
it
is,
so
it
has
to
have
enough
strength
that
you
can
still
make
it
workable.
So
there's
two
big
points
for
the
clamp
design,
you're,
avoiding
threads.
A
One
issue
that
you,
you
haven't
seen
it
but
I
have
I
do
our
current
extruder
has
got
threads
so
one
it's
weaker
thermal
contact
because
threads
aren't
like
a
hundred
percent
meshed
parts
of
the
teeth
are
meshing
and
a
lot
of
that
is
air
space.
So
the
conduction
there
is
not
great
so,
for
example,
on
a
heater
block
you're
trying
to
heat
the
filament.
That's
inside
that
heat
break,
but
you
can't
do
it
as
efficiently.
So
in
a
optimized
design,
you,
the
heater
block
itself
to
be
a
clamp-on
design.
A
The
second
aspect
of
that
is
that
you
can
arrange
the
angle
of
the
the
heater
block
as
you
need
to,
because
you
need
to
be
at
an
exact
angle,
because
the
thing
is
rectangular
unless
it's
spherical
the
angle
map,
unless
it's
cylindrically
symmetric
the
angle
matters,
because
then
you
are
either
heat
hitting
the
cooling
fans
or
something
else.
So
you
need
precise
positioning
of
that
angularly
with
threads
absolutely
hitting
this
and
I.
Don't
know
how
how
e3d
still
gets
away
with
that?
Maybe
most
people
don't
have
that
kind
of
a
requirement.
A
Here
we
have
a
requirement
that
we
need
to
position
the
heater
block
at
a
particular
angle,
because
there
are
cooling
fans
around
it
and,
if
and
and
the
nozzle
cooling
nozzle.
So
if,
if
you
don't
have
it
the
right
geometry,
you're
gonna
be
hitting
your
cooling
nozzle,
so
you
want
to
control
the
angle.
That's
the
second
reason
for
clamping.
A
Besides
it's
easier,
you
have
wires
coming
out
of
the
top
of
the
heater
block
if
you're,
spinning
them
you're,
hitting
the
wires
like
when
you
get
close
to
the
full
full
thread
on
you're,
hitting
all
your
wires
against
the
top.
So
it's
messy
here,
you
just
clamp
it
on.
You
know
you
put
the
wires
out
of
the
way
you
don't
have
to
worry
about
wire
management
when
you're
putting
on
the
heat,
heater
block,
Matthew.
B
A
A
A
Well,
then,
so,
let's
look
at
the
limits.
The
ideal
thing
is
when
there
is
no
boundary
there
when
you're
getting
all
the
heat
to
it,
but
so
my
claim
would
be
if
you
have
a
boundary
like
this
versus
a
boundary
like
that
I'm
not
seeing
a
difference
between
the
conductivity,
it's
one
of
those
things
of
the
sine
theta.
Think
of
the
angle
where
I.
A
D
A
Well,
I
think
it's
much
closer
than
threads,
and
so
you
know
we're
hand
waving
here.
We
don't
have
like
precise
data
to
show
it,
but
you
can,
at
the
end
of
the
day,
simply
take
thermal
measurements.
Like
you
turn
on
the
heater
and
you
measure
the
point
at
which
you
are
extruding
good
plastic,
so
you
can
do
that.
You
can
do
it,
an
experiment,
you're,
taking
the
exact
same
system.
One
is
threaded,
one
is
clamped
and
you're
simply
taking
data.
A
How
long
does
it
take
you
to
heat
up
and
that's
a
great
experiment
for
someone
here
to
do?
You
can
do
that
if
you
know
once
we
get
this
all
designed,
but
that's
the
kind
of
data
we'd
have
to
take,
because
maybe
maybe
there's
another
effect
here
that
says
it
absolutely
doesn't
matter,
but
the
first
principle
analysis
of
this
says
it
does
because
there's
a
lot
of
air
space
and
threads.
You.
A
A
Yeah
there
could
be
a
lot
of
things
but
from
the
first
principles.
I
would
start
with
this
and
why?
Why
also
it's
a
third
third
reason
for
doing
this:
it's
easier
to
drill
a
hole
than
to
thread
a
hole
right,
so
the
machining
requirement
here
is
much
less.
All
you
have
to
do.
There
is
just
like
with
our
split
shaft
couplers.
You
have
to
slit
that
block
either
partial
or
a
hole.
Split
I
would
do
a
clam
shell.
A
B
A
The
clam
shell,
okay,
details,
I'll
talk
about
the
scalability
of
these
nozzles
because
there
are
very
important
details
there,
because
for
one
thing,
the
way
these
systems
work
in
order
not
to
create
leaks,
the
nozzle
itself
and
and
I'm
gonna
draw
that
detail
right
in
there
right
now,
so
the
nozzle
itself
touches
so
there's
a
part
of
the
nozzle
there.
That's
screwed
in
that
touches
against
the
heat
break,
they're
actually
screwed
to
each
other,
so
there
is
absolutely
no
leak
between
them.
A
If
you
had
left
a
gap
and
there
were
threads,
then
plastic
would
leak
out
the
gap
it
does.
It
does
exactly
that.
If
you
don't
tighten
the
nozzle
against
the
heat
break,
you
have
leaks
and
your
prints
gets
messed
up.
You
eventually
get
a
whole
glob.
There
was
like
a
classic
I
should
find
this
on
the
LSD
workshops.
Facebook
page
I
showed
a
picture
of
what
happens
when
you
get
a
leak
out
of
that.
In
one
case,
it
was
the
the
funkiest
thing
I've
seen
the
entire
extruder
got
swallowed
up
in
plastic.
A
There
was
a
little
leak.
It
kind
of
kept
leaking
up.
The
whole
thing
got
encased
in
the
plastic:
I
should
show
you
that
for
what
happens
when
you
don't
have
that,
so,
let's
see
open
source
ecology
workshops.
What
is
that?
Because
it
did
anyone
see
that
Elysee
workshops,
if
I
can
find
it
so
search.
This
group.
A
But
just
to
show
you
that
that's
the
kind
of
features
you
have
to
pay
attention
to
in
a
system
like
the
threads
leak,
absolutely
clamping
will
not
leak.
So
you
remove
the
requirement
of
budding
up
the
two
against
each
other.
That's
another
reason
why
this
design
can
be
potentially
better.
So
if
you're
clamping
the
thing
on
top
of
the
thing
on
the
bottom,
they
have
to
have
the
same
diameter,
but
if
they
do,
when
you
clamp
up
on
them
with
a
straight
split
hole,
can't
find
it.
A
You
look
since
maybe
add
that
to
the
system
here
later,
you
remove
the
requirement
of
butting
up
the
the
nozzle
against
the
heat
break,
exactly
which
is
a
tricky
process.
I
mean
III.
D
has
instructions
for
doing
that,
I
actually
emailed
them,
say:
hey
guys,
you
can
never
get
the
angle
proper.
With
the
way
you've
got
this
design
and
I
said
yeah.
It
doesn't
matter
it
for
a
lot
of
people
that
doesn't
matter
if
they
don't
have
tighter
requirements
on
the
geometry.
A
For
us,
it
does
matter
with,
with
the
current
no
nozzle
on
extruder,
which
I
haven't
really
shown
the
extruders,
how
we
have
it,
but
we
have
a
nozzle.
That's
like
right.
Next,
not
the
fan
nozzle
for
the
print
cooling
fan.
It's
pretty
much
right
next
to
the
to
the
extruder
nozzle,
which
means
that
if
the
heater
block
has
turned
towards
that
nozzle,
it
interferes
so
we
can't
do
it
well,
let's
continue
so
skelet.
We
kind
of
talked
a
little
bit.
Let's
see,
here's
just
another
configuration
thinking.
A
Okay,
where
are
we
gonna
put
the
fans
so
that
we
can
still
cool
this
system
and
what
okay
and
printing
belts
on
a
small
small
area?
You
can
just
curl
up
your
belt
on
a
print
bed,
so
you
can
get
like
a
2
meter
belt
on
an
8
inch
by
8
inch
bed
make
some
metric
and
imperial
here,
but
that's
the
case
for
rubber
extruders
optimized.
A
This
deal
here
is
optimized
for
rubber,
the
shortest
possible
path
between
the
drive
and
the
heater
heater
part,
so
I'm,
including
the
clamp
on
both
at
the
heat
sink
area
and
at
the
heater
block
area,
which
means
that
the
component
there
is
not
a
threaded
long
tube.
It
could
be
just
a
tube,
which
is
much
easier.
You
don't
have
to
worry
about
first
threading,
that
too
so
absolute
simplicity.
This
is
the
next
generation.
This
is
gonna,
be
universal
in
a
few
years
and
you've
seen
it
here,
I
mean
it
has
to
be.
A
When
we
talk
about
improving
performance,
this
is
such
a
gap
in
the
community
right
now.
You
cannot
really
print
anything
large
with
rubber
right
now,
yeah
you
can
print
in
rubber,
and
so
the
people
who
advertise
oh
yeah,
you
can
print
rubber
yeah.
You
can
do
that,
but
very
slowly,
not
in
a
way
that
is
economically
feasible.
If
you
talk
about
actually
making
real
parts.
C
A
We're
starting
with
three
millimeter
filled
filament,
which
is
off-the-shelf
standard.
When
we
designed
this
ourselves
and
we
start
extruding
our
own
filament,
we
probably
want
to
maybe
go
to
four
or
five,
so
you
have
an
easier
way
to
de
print
when
the
filament,
the
larger
the
filament.
If
you
have
an
error
in
its
diameter
percentage-wise,
that
error
will
be
less
as
a
fraction
of
the
overall
width.
Therefore,
the
quality
control
requirements
for
producing
filament-
that's
larger,
are
much
decreased
compared
to
1.75
filament.
A
Thermoplastic,
urethane
or
other
thermoplastics
are
what
we're
talking
about.
This
is
not
the
same
as
the
a
lot
of
the
tire
rubber
which
is
from
rubber
plantations
rubber
trees,
but
some
tires
are
made
from
thermoplastics
like
I've.
Seen,
for
example,
a
lot
of
snowmobiles
have
thermoplastic
urethane
tracks.
B
A
Tpo
thermoplastic
olefins,
there's
various
rubbers
that
are
thermoplastic.
That
means
you
can
melt
them
and
then
remelt
them
and
use
them
again
and
again,
which
is
good
for
the
environment.
So
how
do
we
get
this?
If
you
want
significant
drive,
and
we
here
we're,
you
notice
we're
eliminating
gear
down.
So
how
do
we
do
that?
A
The
currently
the
way
the
e3d
Titan
Aero
is
set
up
which
wouldn't
use
it
has
the
drive
gear
like
I,
show
here,
and
then
it
has
the
gear
down
like
up
there
around
that
this
hole
here,
which
means
you've,
already
moved
the
point
of
drive,
because
the
point
of
drive
is
going
to
be
around
somewhere
like
up
there.
So
you
kind
of
like
doubled
the
distance
to
the
heater
block.
A
D
A
Necklace
design
the
neck
that
part
where
the
filament
gets
constrained
right
after
your
drive,
it's
part
of
the
heater
block
in
an
e3d
tight,
narrow,
there's,
a
separate
plastic
piece
where
the
filament
goes
into
that
before
it
goes
into
the
the
heatsink,
we're
eliminating
that.
So
for
another
thing,
we're
eliminating
and
yet
another
part
the
importance
of
that
is
you're,
shortening
down
the
distance.
A
So
when
III
D
had
they're
geared
extruder,
the
drive
was
somewhere
further
up
in
order
to
not
to
crinkle
that
filament
before
the
heatsink,
they
have
to
use
that
other
constraining
neck
piece,
basically
a
tube
that
goes
into
it,
constrains
it.
It's
a
small
piece:
it's
like
you
got
to
get
it
in
there.
It's
it's
adds
to
complexity
and
so
forth.
So
the
idea
here
is
start
with
a
direct
drive
system,
meaning
what
you
have
the
stepper
motor
there
is
already
geared
down.
You
want
to
use
a
light,
stepper
motor
like
a
NEMA
17.
A
D
A
That's
another
concept
to
consider
here,
but
whenever
you're
printing
and
you
jump
over
to
another
place-
you
don't
know
you
don't
want
a
little
strand
of
filament
to
keep
dragging
you
want
to
suck
it
back
up.
So
it's
retraction
in
a
motor
like
this.
The
only
challenges
do
you
have
enough
accuracy
to
do
retraction
properly,
I,
don't
see
why
not
you
spit
one
way
and
you
spin
all
the
way
it's
not
like.
A
We
may
have
some
difficulties
using
this
for
say:
the
axis
drive
I,
don't
know,
haven't
tried
it
so
you'd
have
to
get
make
sure
your
prints
are
pretty
good
in
order
to
be
able
to
retain
an
accuracy,
but
you
can
also
think
that
the
way
to
address
accuracy
here
is
you
have
a
you
know.
This
brute
force
printed,
3d
printed
gear
and
then.
A
You
can
have
n'stuff
direct
drive
of
a
pulley.
You
can
have
another
gear
down
where
at
the
end
of
the
day,
it
might
actually
work
out
because
a
gear
down
whenever
you
have
a
gear
down
that
step
motion
is
going
to
be
decreased.
So
right
now
we
have
10
microns
with
a
further
gear
down
after
this,
or
maybe,
if
you
gear
this
down
enough,
so
we
already
have
the
stepper
motor,
which
we
said
has
3200
divisions
per
revolution.
Well,
this.
D
A
Has
backlash
if
the
gear
down
can
compensate
for
that
and
actually
do
better
states
of
five
or
ten
fold
gear
down?
If
the
backlash
got
your
results
twice
as
bad,
but
the
gear
down
compensate,
you
can
possibly
do
actually
even
better
experiments
to
be
done.
So
so
the
idea
being
like,
if
you
engineer
this
or
you
play,
keep
playing
with
us
and
refine
the
system
you're,
freeing
yourself
from
depending
on
these,
which
are
much
more
complex.
So
we
got.
Let's
talk
about
scalability,
so.
A
A
A
A
That
there's,
you
can
buy
a
small
heater
block
for
one
dollar,
but
if
that
heater
block
is
about
five
times
bigger,
it's
like
two
inches
long.
It
ends
up
costing
$110.
So
so
this
heater
block
they
have
ones
that
are
that
long,
that's
the
supervolcano,
but
it
costs
quite
a
bit.
So
how
can
we
do
better
with
simply
stacking
through
these
together?
So
take
two
volcano
nozzles
next
to
each
other
and
stack
them
just
like
this.
So
there
you
go
single
40
watt,
heater
block,
you're,
doubling
them
tripping.
Why
don't
you
keep
going
the
limits?
A
Gonna
be
how
much
plastic
you
can
melt
through
that
like
in
the
thermal
contact
area,
and
then
you
need
more
force
to
push
this,
but
to
two
of
them
yeah.
You
could
absolutely
do
that
with
the
existing
Titan
arrow
just
put
two
blocks
on
there's
details
of
how
you
would
do
that.
So,
let's
take
a
look
at
that
we
mentioned
so
in
a
threaded
design.
These
are
details
for
threaded
design.
If
you
wanted
to
take
off
the
shelf
existing
parts,
it
cost
a
few
dollars
double
that.
A
D
A
A
B
A
You're
required
to
bought
up
everything
tight
and
screw
it
in
tightly.
You
can't
have
leaks,
so
this
might
get
a
little
challenging
like
because
you're
gonna
have
to,
but
that
up
these
two
up
against
each
other,
so
you
have
to
kind
of
screw.
It
thread
that
you
know
screw
it
in
through
that.
Well,
maybe
you
can
take
at
the
very
end
of
the
day,
just
take
the
volcano
nozzle
and
screw
it
in
tight.
So
it
tightens
both
this
gap.
In
that
gap,
tricky
I
mean
tricky
stuff.
So
maybe
this
is
like
not
that
easy.
A
You
have
to
get
good
at
it.
This
is
the
kind
of
stuff
we're
working
with
threads
is
just
hard
cuz
threads
behave.
You
know,
you've
got
the
threads
inside
this
heater
blocking
the
side.
This
threaded
compartment
I
can't
predict.
What's
gonna
happen
here,
but
theoretically
you
can
do
this
and
this
probably
will
work
if
you
get
it
right,
but
you
have
to
watch
out
for
leaks
like
the
budding
budding
of
these
where
the
breaks
are.
So
what
about
doing
this?
One
real
super
long
nozzle!
Yes,.
D
A
Nozzle,
you
can
use
the
super
volcano
nozzle,
so
you
can
eliminate
that
block
right
there,
but
that's!
This
is
once
again
where
the
next
step
see
like
right
here.
We've
got
an
example:
we've
got
the
heat
break
and
a
super
volcano
nozzle,
which
is
long
and
stretches
through
about
two
of
those
heater
blocks.
So
then
you
can
do
that
and
what
about
this?
Yeah
if
you
keep
going
further?
So
in
this
case,
so
you
can
kind
of
think
about
this.
A
Be
the
part
at
the
bottom,
where
you
want
the
fan
close
to
it,
so
maybe
you
can
do
it
or
maybe
like
mount
the
fans,
schooling
fans
just
far
away
so
design
that
angle
dependency
out
of
this,
but
then
you'll
get
pretty
much.
All
your
printers
will
look
a
little
different.
They
might
have
the
heater
blocks.
Each
of
these
here
blocks
might
be
at
a
different
angle,
which
is
just
visual.
It
could
still
work
I,
guess
what
you
have
to
consider
ease
of
cable
management.
A
What
I
would
suggest
is
do
exactly
this
with
clamp
blocks
so
now
you're
using
you
can
cut
your
simple
tube
to
be
as
long
as
you
want
at
the
end
of
that
tube,
you
still
need
the
contact
between
the
long
tube
and
the
nozzle.
So
that's
a
missing
link.
You'd
still
have
to
get
that
nozzle
tight
against
that
whole
tube,
which
is
incompatible
because
you
can.
How
do
you?
How
do
you
mix
clamping
with
bulky?
A
Maybe
you
could
maybe
even
thread
the
very
end
of
that
screw.
That,
in
the
rest
of
this
is
on
threaded,
when
you
slid
it,
it
might
just
end
up
that.
It
comes
back
now,
tricky
too
tricky.
So
here
in
a
scaleable
heater
block,
you
are
probably
left
with
you've,
got
a
long
tube
and
into
end
of
the
end
of
it
you
would
have
to.
Maybe
you
would
have
to
make
your
nozzle
part
of
that
long
tube,
which
I
guess
that's
a
little
bit
of
machining
there.
So.
A
Okay,
okay,
so
right
let
call
it
right.
So
thank
you.
So
we
said
that
if
you
have
a
clamp
design
here
which
I
don't
I
haven't
shown
any
clamp
designs
here,
because
I
was
thinking
about
using
off-the-shelf
parts
because
we
haven't
done
any
of
these
other
things,
so
something
that
anybody
can
get
without
having
you
having
to
slit
and
and
find
these
other
parts
are
easily
doable.
But
nobody
just
doing
that
right
now,
so
you
can
get
the
parts
so.
A
It's
clamped
on
at
the
very
end,
you
just
machine
your
nozzle
to
be
without
threats
period
done.
So
that's
that's
what
you
got
to
do,
but
that
depends
on
you
how
you
can
machine
that
little
or
find
somewhere
or
machine
that
nozzle
that's
without
threats.
All
of
them
come
with
Fred's
right
now.
Nobody
makes
unthreaded
heat
breaks
or
nozzles
I
know.
Now
the
other
company
called
J
head.
A
Those
guys
actually
make
clamp
systems
and
there's
arguments
that
J
head
is
actually
the
technology's
gonna
go
towards
what
J
Ted
is
doing
because
their
nozzles
are
pretty
high
performance.
I
talked
to
those
guys
at
the
Midwest
rap
rap
festival,
nice
guys-
and
we
were
sharing
these
thoughts
about
the
perennial
issue
of
threats,
and
they
they
completely
got
that
there
they
don't
use
threads
and
they
said
that's
a
bad
idea,
so
they
have
press-fit
parts.
A
You
know!
So
that's
so
when
you
depress
fit
it's
kind
of
hard
to
service,
because
once
you
press
fit
that
in
it's
really
hard
to
break
it
apart,
because
the
metal
might
fuse
into
itself
also
it's
kind
of
not
the
way
to
go.
Either
gotta
go
with
a
clamped
stuff,
so
I
think
personally,
I'd
say
the
clamp
designs
are
going
to
be
the
future
of
3d
printing
scaleable
nozzles.
Is
we
actually
get
to
real
printing
right
now,
everybody's
doing
hobby?
Printing?
Very,
not
a
lot
of
we
could
say.
A
Probably
the
industrial
printing
with
the
existing
open
source
3d
printers
out
there
is
less
than
a
hobby
style.
Definitely
a
lot
of
businesses
are
starting
to
print
real
parts.
Lowell
spot
is
a
great
case.
You
can
go
to
lulzbot
comm.
They
have
a
lot
of
case
studies
of
various
people
from
GM
to
the
mill
tree
using
our
3d
printers,
which
is
kind
of
interesting,
because
now
Jeff
Moe
is
servicing
the
military
industry,
but.
C
A
A
B
A
D
A
A
D
A
D
A
It's
okay,
if
you
were
not
heat
tightening
later,
but
there's
also
heat
tightening
when
you
tighten
this
up.
The
other
aspect
of
this
is
to
get
it
into
final
position.
You
need
to
tighten
it
under
heat.
So
that's
another
step
you
can
take
with
a
clamp
on
I.
Don't
believe
that
heat
tightening
is
a
necessary
step,
but
the
angle
I
can
show
you
on
the
real
printer
like
where
it's
where
it
gets
troublesome,
but
you
can't
get
the
angle
so
the
angle,
lower
thermal
conductivity
and
much
more
difficult
fabrication,
I
mean.
B
A
A
B
A
B
A
A
A
D
A
A
A
A
It's
really
thin
like
at
the
very
end,
I
mean.
How
are
you
the
machine
that
you're
gonna
be
wobbling
that
not
so
you
have
to
probably
in
the
machining
process
that
they
probably
are
using
some
form
of
rest
or
a
stop
against
which
that
thing
can
rotate
in
order.
So
when
you're
writing
it
like
machining,
that
there's
different
ways
to
machine,
but
there's
definitely
challenges
on
the
production
engineering
side
for
long
threads.
B
A
A
A
A
C
C
D
A
C
C
A
Of
a
homemade
extruder
heater
block,
it
houses
a
little
heater
element
or
your
nozzle,
your
heat,
brake
heat
sink.
There's
a
gear,
that's
missing,
but
this
is
the
filament
feed
that
Foreman
goes
in
there
and
gets
through.
So
this
isn't
a
wiki
under
simple
extruder.
Now,
let's
talk
about
filament
making
extruders
so
now
the
device
that
gets
you
the
filament
to
3d
print
with,
because
one
of
the
ideas
is
that
filament
is
twenty
bucks,
a
kilo,
ten
bucks,
a
pound,
you're
going
to
make
a
chair
or
a
piece
of
plastic
lumber.
A
A
The
precious
plastic
is
a
nice
project
out
there
that
does
that,
but
I
have
a
design
that
ya
start
with
a
shredder.
So
a
shredder.
This
is
an
open-source
shredder.
You
can
look
at
precious
plastic.
What
does
the
film
at
maker
look
like?
So
let's
go
to
lineman
filament
extruder
on
a
wiki,
so
this
is
what
we
have
built
and
we
have
used
successfully.
You
can
go
Lyman
filament
extruder,
that's
Hugh
Lyman.
We
visit
them
he's
out
in
Washington,
State.
D
A
Lemmon,
film
and
makers,
so
that's
the
cat
and
free
cat.
This
is
the
line
when
we
built
in
the
bottom-
and
you
know
this
is
what
we
had
going
on.
So
you've
got
the
hopper.
It's
a
wall-mounted
system.
It's
actually
in
the
workshop,
you
put
pellets
in
there
and
your
extruding
filament
and
we've
got
plus
minus
0.1
millimeter
out
of
the
system.
A
A
That's.
It
needs
a
little
refinement,
but
all
the
systems
are
there
to
make
that
work,
and
this
is
what
we
got
looks
beautiful.
This
is
from
pellets.
This
is
not
from
trash.
Pellets
are
a
dollar
pound
in
bulk
or
down
to
50
cents,
a
pound,
50
cents
to
$1
a
pound.
If
you
buy
a
ton,
if
you
buy
small
quantity
they're
like
four
four
bucks,
a
pound
on
eBay,
so
at
four
bucks
son
on
eBay,
you
know
you're
you're
getting
close
to
the
ten
dollars.
A
This
is
great,
but
we
don't
stop
there.
We
said:
okay,
Hugh,
let's
simplify
this,
that's
a
little
too
much
there.
So
how
do
we
do
the
absolute
simplest
version
of
that?
This
is
the
next
iteration.
They
call
it
the
simple
3d
X
3d
printer
filament
maker
extruder.
Let's
take
a
look
at
that
this
is
unfree
cad.
A
Clearly
you
would
go
to
the
Lincoln
and
a
Facebook
page
2018
simplified
version.
The
CAD
where's,
the
CAD.
There
is
the
CAD
you
can
download
it.
Let
me
download
it
what
happened
there,
download
this
file
and
show
you
in
3d,
but
I
said:
okay,
there's
our
filmin
maker
yeah
I
still
to
complicate
it's
make
it
more
simple.
So
this
is
the
simplified
filament
extruder
and
using
now
much
more
standard
off-the-shelf
parts.
Hugh
is
printing
the
casing
and
putting
all
the
components
in
there.
A
But
how
do
you
do
it
with
like
two
by
fours
and
while
retaining
heat
heat
safety
and
all
that?
So
let's
take
a
look
at
this
one
I'm
just
stripping
down
any
extraneous
details
like
the
3d
printed
parts.
So,
instead
of
doing
what
Hugh's
done
there
I'm
taking
I'm
doing
this
so
there's
like
a
2x4
in
the
back,
your
power
supply
is
on
the
back.
If
you
open
that
out,.
A
How
do
I
hide
that?
It's
not
hiding
on
me!
Well,
basically,
I
attached
a
controller
like
right
on
the
side.
There's
now
we
were
starting
to
see
the
insides
of
it,
but
basically
the
space.
This
is
two
by
fours
or
two
by
sixes
there.
You
got
a
hopper
here,
so
you
fill
this
whole
thing
and
there's
a
there's,
an
extruder,
basically
a
drill
bit
right
there,
and
in
this
case
there's
a.
Let
me
hide
that
there
it
is
so
you've
got
a
coupler.
You've
got
your
auger,
which
is
a
one
inch
drill
bit.
A
I
would
make
this
into
a
one
inch
system.
Hugh.
Does
this
with
a
half
inch
bit
the
precious
plastic
guys
use
a
one-inch
bit?
Why
one
inch?
So
you
can
just
throw
more
nasty
stuff
in
there
with
more
resilience.
So
you
wanted
larger
here.
With
this
Lyman
filament
maker,
you've
got
a
single
heat
band.
You
can
make
multiple
heat
band
say
for
a
one
inch
extruder
like
this,
but
the
compartment
in
there
carries
a
few
pounds
of
pellets.
You
throw
that
in
there
it's
wall-mounted
the
controllers
next
to
it,
there's
a
metal
plate.
A
So
this
gets
hot
right
here,
so
you
got
a
flange
and
then
you've
got
a
piece
of
metal.
Thank
you
so
metal
against
wood.
But
by
here
it's
it's
getting.
You
know
if
it
gets
too
hot.
We
can
put
fans
on
this,
but
this
is
oh,
no,
no
detail!
So
you've
got
the
metal
flange,
underneath
it
put
an
insulating
piece
of
metal
and
then
on
the
other
side.
So
essentially
there's
no
contact
between
the
hot
part
and
this
metal
here.
C
A
Can
put
like
yeah
welding
blanket,
for
example,
is
a
good
insulator.
It's
a
high
temperature
thing
that
that
thing
was
that's
like
thousands
of
thousand
Celsius
or
something
there's
heater,
yeah
called
welding,
blanket
and
I
think
it's
made
from
what
carbon
fiber
is
it
it's
a
carbon
fiber
metal
of
welding?
Well,
Welding
blanket
you
can
use
like
a
ceramic
thing.
Maybe
standoffs
ceramic
standoffs,
but
this
is
the
hot
part
you're
talking
about
200,
C,
200
C
and
would
do
not
mix
pyrolysis
begins
around
100,
C
or
lower.
A
Like
a
DC,
you
start
breaking
wood
apart,
and
this
is
200
C
so
but
I
like
the
wood
idea.
It's
like
this
is
just
simple
metal,
plate
and
wood,
and
there
you
go
now
for
the
motor
there.
That's
the
motor
that
we've
used
like
in
the
pictures,
but
look
at
this
I
mean
that
is
super
simple.
You
can
build
that
in
a
few
days
from
scratch.
The
controller
is
here.
So
what
do
you
have?
In
a
controller?
You
have
the
same.
Solid-State
relay
you
have
an
a
knob
for
selecting
temperatures,
so
there's
a
temperature
controller.
A
So
this
is
essentially
your
thermal
probe.
So,
like
the
the
thermistors
in
our
printer
system,
you've
got
an
on/off
switch.
You've
got
a
power
supply
in
the
back.
This
is
12
DC
for
off-the-shelf
heater
blocks,
but
we
are
learning
about
the
what
we're
doing
in
the
shop
right
now
we
have
built
the
heat
beds,
which
are
nichrome
embedded
in
fiberglass
sleeve,
so
get
rid
of
the
power
supply.
Use.
120
AC
use
the
universal
controller.
A
So
now
that
I'm
thinking
about
I
haven't
come
to
this
since
months,
but
now
I
would
just
put
the
OSC
Universal
controller
on
this
and
forget
about
all
this
novelty.
Here
you
can
have
the
capacity
to
turn
on
a
solid
state
relay
and
120
AC
I
would
get
rid
of
this,
which
is
you
know
a
whole
whole
bunch
of
different
things.
Arranged
differently,
use
the
existing
3d
printer
controller,
and
then
you
have
an
LCD
screen,
so
you
can
be
more
precise
in
what
you're
doing
there.
So
that
would
be
the
next
step
on
this
awesome.
A
It's
this
one
of
the
easiest
parts
you
can
buy
and
off-the-shelf
like
five
dollar
heater
element
or
you
can
make
it
your
when
you
buy
it
off
the
shelf,
you're
limited
to
like
whatever
it's
a
hundred
watts
with
a
nichrome
wire.
You
can
make
it
500
watts
or
whatever,
but
hear
them
the
hardest.
Part
I
would
say
is
the
nozzle
where
you
have
a
threaded,
you
have
a
little
through-hole
like
one
millimeter
or
so
I.
A
Is
one
point
two
millimeters
I
guess
you
can
drill
that
out
to
three
millimeters
and
it
could
work
so
maybe,
instead
of
that
plug
brass
plug
you're
using
another
fitting,
that's
reduced
down
to
where
you
can
screw
in
the
six
millimeter
3d
printer
nozzle.
That
might
be
an
interesting
thing
to
do,
because
otherwise
you
have
to
drill
something
from
scratch.
Whereas
if
you
already
have
that
nozzle,
you
can
ream
out
that
hole,
it's
it's
self
guiding.
It
will
be
very
straight,
so
there
might
be
an
advantage
of
doing
that.
So.
A
Then
you
can
screw
in
the
different
nozzles
for
different
filaments.
So
that's
that's
the
basics
of
the
3d
printer
filament
maker,
one
of
the
critical
things
you're
controlling
a
heat
element,
you're
augering,
the
pellets
that
go
in
there
and
before
that
you
want
to
shred
so
you're
working
from
trash
and
the
shredder.
Let's
go
check
out
the
shredder,
but
what
you
need
to
do
for
the
shredder
is
look
at
precious
plastic
and
replicate
that
so
that's
doable,
but
the
current
design
of
the
shredder
the
admission
there
is
that
it
wears
out
too
fast.
A
It's
got
quarter-inch
blades
I
would
do
do
this
system
and
then
cut
out
half
inch
blades
that
when
scale
get
gets
you
to
your
rock
crushers
and
things,
but
do
a
little
bigger
than
this.
What
people
found
was
this:
these
blades
on
this
tend
to
wear
out
too
fast,
they're,
just
quarter-inch
thick
when
you're
talking
about
even
grinding
plastic.
You've
got
rubbing
and
stuff
like
that.
So
you
want
to
harden
that.
A
So
maybe
after
you
cut
the
if
you
cut
with
a
torch,
there's
already
some
heat
hardening
happening,
because
a
flame
hardened
steel
that
it
cuts,
even
if
you're
doing
mild
steel,
so
not
laser
cutting
but
torch,
cutting
with
a
carbon
containing
oxy
fuel
gas.
So
like
oxy
acetylene,
the
carbon
that
goes
comes
from
the
flame,
actually
hardens
the
steel
itself,
so
shredders
precious
plastic.
If
you
haven't
heard
about
it,
Google
that
there's
great
videos
on
there
for
how
to
build
all
these
machines
and.
A
Things
like
that
in
the
extruder,
some
people
call
for
a
breaker
plate.
So
in
the
more
advanced
versions
you
would
have
at
the
very
end
of
the
tube.
You
would
have
a
breaker
plate
so
a
little
plate
with
holes
in
it,
so
that
it
balances
out
the
pressure
and
maybe
blocks
impurities
of
their
impurities
in
your
plastic.
A
But
I
would
go
inside
like
right
past
the
auger
bit
dog
herb
it
reaches
to
within,
like
you
know,
just
a
few
millimeters
or
half
an
inch
of
the
nozzle,
so
you're,
just
all
going
and
just
pushing
it
down
and
gravity
helps
you
have
to
use
less
force
in
terms
of
augering
here.
So
I
think
the
wall-mounted
designer
is
decent
for
doing
as
opposed
to
horizontal.
Where
you,
you
won't
get
gravity
assisting
you
here.
It's
like
you're
gonna
have
a
more
uniform
flow
because
in
the
horizontal
system
you
know,
gravity
draws
things
down.
A
Does
so
so
because
we
have
in
our
system,
you
have
basically
like
this
thing:
dribbling
down
from
your
wall
like
a
meter
of
this
filament,
depending
on
how
hot
that
the
room
is,
if
it's,
if
you're
in
the
Arctic-
and
it's
really
you
know
it's
near
freezing
I
mean.
Obviously
that's
gonna
get
you
different
results,
so
you
want
to
control
the
temperatures
of
that.
Somehow.
C
A
A
A
A
Okay,
so,
but
getting
back
to
extruders
one
important
part
about
extruders
is
the
retraction,
because
that's
what's
going
to
get
you
perfect,
let's
look
at.
Let's
talk
about
retraction,
maybe
a
wrap-up!
That's
one
good
thing
you
have
to
think
about
because
this
here
this
video
here
shows
you
the
different
results.
When.
A
Okay,
that
let's
well,
but
let's
refresh
that
page,
but
this
cover
page
here-
shows
you
the
that's
when
you
don't
have
enough
retraction
and
you're
skipping
between
one
post
and
the
other.
You
have
a
little
piece
of
filament.
That's
just
dangling
making
this
complete
thing
that
shouldn't
be
there.
That's
perfect
retraction,
you
retract,
just
a
little
bit,
so
the
filament
pulls
away
and
you're
clear
to
jump
to
the
next
place.
A
You
have
too
much
retraction
you're,
actually
pulling
enough
filament
that
you're
not
able
to
extrude
enough
when
you
start
printing
again
and
then
you
get
very
weak
parts.
So
do
this
test
they'll,
be
like
your
ultimate
arbiter
of
your
retraction
settings
in
the
cure
up
the
interface,
you
select
the
the
length
of
retraction
and
speed
of
retraction
and
that's
some
of
the
settings
that,
when
you're
doing
Production,
Engineering
you're
open
sourcing,
that
that
goes
into
your
initialization
file
that
you
save
and
give
to
other
people
to
say.
A
Okay
now,
I
figured
how
to
print
this
exactly
these
are
all
the
settings
this.
This
will
be
one
of
the
settings
because
you
know,
depending
on
your
many
factors,
what
filament
you're
working
with
all
the
filaments
will
behave
slightly
differently.
So
you
want
to
optimize
this
for
every
filament
and
save
that
settings
file.
So
you
can
do
that.
You
don't
have
to
do
the
figuring
out
the
next,
so
I
think
I'll
wrap
up
with
that.
Maybe
take
any
questions
so
here
in
this
session,
you've
got
some
insights
into
what
the
optimized.
B
B
A
Get
off-the-shelf
1.75
and
3
millimeter,
the
Titan
Aero
can
actually
run
with
1.75
the
3
3
millimeter
system
it'll
be
a
little
worse
than
a
dedicated
you
can
you
can
print
the
adjustment
happens
within
cure.
You
tell
Kyra
what
thickness
filament
you
have.
So
if
you
made
your
own
filament
and
it's
2.5
millimeters
you
put
that
into
Kyra,
and
then
it
extrudes
it
according
to
that
speed.
But
you
cannot
do
four
millimeter.
A
Room
in
extruder
for
it,
but
you
want
to
match
like
I
talked
about
leakage
and
the
plunger
idea,
so
you
want
to
be
as
close
to
the
physical
system
as
possible.
So,
for
example,
4
millimeter
will
not
work
the
neck
and
the
holes
in
there.
Don't
let
you
put
something
bigger
than
3
millimeters
in
now,
when
you
design
your
own,
you
can
make
a
larger
one
that
does
handle
yeah.
A
The
other
thing
about
extruders
is
there
are
filament
with
sensors,
so
you
can
add
a
sensor
that
detects
the
filament
with
as
it
changes
throughout
so
say
you
make
crappy
filament,
you
add
that
and
that's
a
feature
already
available
within
Marland.
So
you
can
add
this
other
component
to
your
extruder,
which
then
adjusts
in
real
time
for
the
thickness
of
your
filament.
So
in
principle,
when
this
is
all
worked
out,
you
can
make
any
kind
of
crappy
irregular
filament
and
get
really
good
prints,
really
good
to
a
limit.
D
A
B
A
A
D
A
D
A
So
that
that
would
be
pasted
so,
like
they'll,
be
concrete,
paste,
extruders
yeah,
you
can
build
houses
with
3d
printers
dude,
that's
a
little
different.
That's
the
syringe
based
or
are
you're
pumping
right,
a
lot
of
yeah
you're
pumping.
Actually,
let's
show
that
as
a
what's
the
bench
top
example,
look
like
there's.
A
A
This
is
what
we
want
to
do.
This
is
fully
open
source.
So
that's
a
that's
your
syringe
and
it's
feeding
into
a
small
auger.
So
instead
of
a
filament
of
plastic,
this
is
a
modular
setup
that
allows
you
to
do
that.
That's
awesome!
It's
modular!
So
you
have
this
device
driven
by
a
stepper
motor
you're
limited
to
certain
size
prints,
because
you
can
only
hold
so
much
paste
in
there,
but
supersize
it
gear
down
and
you
can
print
large
objects.
A
A
You
can
make
ceramic
offsets
for
our
filament
maker
custom,
3d
printed
ceramic
offsets
in
that
case,
you'd
want
to
bake
them
here
the
process-
maybe
you
can
do
maybe
raw
pastes,
but
if
you
want
ceramic,
that's
bonded
like
brick
like
ceramics,
you
need
to
fire,
it
so
say
we
can
3d
print
our
own,
like
the
toilet,
for
example,
or
sink
for
our
open-source
bio
digester.
There
you
go
supersize.
This
nozzle
use
ceramic
paste
like
porcelain.
A
As
modular
as
possible,
so
you
might
have
a
system
here
where
you
you
might
have,
for
example,
another
syringe
that
does
your
water
purge
at
the
end,
and
you
can
perhaps
automate
that,
but
otherwise
you
just
have
to
clean
that
out
after
you're
using
this,
so
you
printed
it.
If
you
keep
using
it,
you
just
keep
refilling
it,
but
if
you
let
it
dry,
you
want
to
clean
it.
So
that's
that's
the
challenge
here.
A
You
don't
just
leave
it
like
3d,
printing
filament,
which
just
solidifies
here
you
have
to
have
an
extra
measure
for
cleaning
or
have
a
self-cleaning
system.
For
example,
a
self-cleaning
system
would
be
a
system
where
the
extruder
is
near
down
so
hard
that
you
can
break
through
the
clog
upon
startup
or
have
some
kind
of
a
wash
through
system.
So.
A
That's
where
the
scalability
to
solid
parts
and
the
ability
to
modify
this
is
useful,
because
then
you
have
absolute
control.
What
you
can
do
with
this,
as
opposed
to
maybe
some
limited
applications,
but
this
Penn
State
work
I
just
recently
found
out
about
I,
didn't
even
know
about
this,
but
it
appears
to
be
absolutely
open-source
the
guy
this
guy.
This
is
what
he's
printing
with
this
kind
of
stuff
like
past
work,
I
guess
he's
doing
3d
printing.
This
is
clay.
A
D
A
Can
do
a
lot
with
this
I'm,
not
sure
he's
used
he's
using
this,
how
this
metal
object?
I,
don't
it's
metal,
but
you
can
do
metal,
embedded
pastes
where
you
Center
that
at
the
end,
so
you
bake
it
at
the
end
and
paste
part
evaporates
and
you're
left
with
a
hundred
percent
metal
object,
so
that
is
very
powerful.
That
would
require
some
development.
There's
companies
doing
that.
The
open-source
version
of
that
would
be
very
welcome,
so
I
think
I'll
and
with
that
any
other
questions.