►
From YouTube: Thunderhead Filament Extruder Working Meeting
Description
The Thunderhead Filament Extruder produces 3D printing filament from plastic bottles. The extruder is designed for PET plastic.
Here is a meeting with Matthew Rogge of Tech For Trade - considering running an Extreme Build of the extruder in the form of an OSE Extreme Manufacturing workshop.
Notes at http://opensourceecology.org/wiki/Thunderhead_Filament_Extruder
Tech for Trade website - http://www.techfortrade.org/
Tech for Trade Github - https://github.com/techfortrade
Similar idea for insulation mentioned in discussion - http://www.thepolyflossfactory.com/
-----------
What you see here at Open Source Ecology is an ambitious program based on a voluntary effort. To help us reach the goals - please consider joining as an OSE Developer in 2017-
http://opensourceecology.org/wiki/OSE_Developers
Take a minute to subscribe to our email newsletter (updates, workshops, etc): http://bit.ly/1LtcM44
A
Okay,
starting
the
recording
here,
we're
here
with
Matt
from
the
Thunderhead
that
this
is
the
Thunderhead
filament
maker
from
a
collaborating
project
with
open
source
ecology.
We
were
working
on
on
doing
a
workshop
where
we
do
it
one
day
build
not
a
one
day
but
a
weekend
build
of
the
Thunderhead
filament
maker.
B
B
A
Collaborating
so
so,
the
good
news
here
is
that
this
is
officially
all
open
source.
So
we
can
collaborate
and
look
at
bringing
this
open
source
filament
maker
to
to
the
public,
and
this
one
works
with
PE
T
and
there's
a
whole
discussion.
Why
PE
T
was
chosen,
and
that
is
because
it's
a
very,
very
widely
accessible
plastic,
but
it's
also
much
harder
to
extrude
and
that's
why
this
this
device
that
we're
talking
about
right
now
has
a
water
bath.
It's
much
more
complex
than
the
linemen
filament
maker,
which
is
just
simple,
air-cooled
Matt.
B
A
Sorry
mind
if
you
can
just
provide
some
perspective
like
where
you
are
with
the
project,
and
just
just
the
website
where
you
guys
are
to
get
all
together
with
the
project
is
on
our
side.
We
pretty
much
did
a
partial
build
of
Liman,
we
kind
of
stalled
it
because
actually
one
of
our
PID
sensors
wasn't
working.
So
we
couldn't
build
it
that
one
day,
but
anyway,
we're
looking
at
this
as
joining
forces
and
integrating
this
knowledge
into
a
workable
product
that
can
do
all
kinds
of
filaments.
So
oh
yeah
go
ahead.
B
Work
with
texture
trade,
which
is
a
video
that
does
a
lot
of
working
doctor,
cut
and
power
kind
of
goals
around
developing
this
obscure,
is
to
make
3d
printing
accessible
anywhere
in
the
world
anywhere
in
the
world
and
most
of
plastics
industries
based
on
pellets,
even
in
the
recycling
industry.
Look
the
first
thing
face
the
plastics:
it's
get
it
cleaned
up
and
turned
into
pellets.
Then
it
goes
on
to
remanufacture.
B
A
B
B
B
A
B
B
A
B
A
A
B
A
B
B
A
B
B
B
B
You
know
anyhow,
so,
basically,
when
you
run
that
processing
file
while
connected
while
the
computers
connected
diameter
sensor,
you
can
control
the
diameter
sensor
from
the
computer
and
the
computer
will
run
run
the
calibration
standard
across
filament
diameter
and
record
it
just
gobs
and
gobs
of
ADA
about
you
know
where
the
shadows
are
so
basically
corrects
for
the
inconsistencies
in
the
lens
and
the
laser,
because
the
intensity
of
laser
light
is
not
even
across
the
entire
sensor.
So
it
basically
takes
all
that
data
and
then
crunches
it
down
and
then
produces.
B
B
A
B
A
A
B
A
B
Just
kind
of
a
snapshot
of
how
the
diameter
sensor
works,
so
there
is,
you
know
so.
There's
one
kind
of
drawback
is
actually
scanning
that
standard
across,
although
I,
to
be
honest,
I've
only
calibrated
like
3
diameter
sensors,
so
I
haven't
done
a
lot
of
work
into
figuring
out
like
exactly
how
much.
B
A
B
And
so
it
says:
okay,
that's
that's
the
edge
and
now
you're
going
to
be
at
a
certain
pixel,
and
so
then
it
says:
okay,
this
pixel
is
that
such
in
such
location
and
the
way
I
know
the
location
is
because
the
carriage
for
the
for
the
that
moves,
the
standard
back
and
forth
is
driven
by
a
screw.
So
I
basically
know
how
many
revolutions
I've
done
location
so
I
know
well.
A
B
A
B
A
B
A
B
There's
main
arduino
that
controls
everything
and
it
also
controls,
so
it
it
coordinates
everything
and
it
controls
everything
that
happens
on
the
main
extruder
body.
So
the
different
parts
of
barrel
running
the
screw
I
guess
that's
pretty
much
about
it
on
the
main
barrel,
so
it
does
those
things
and
then
it
also
displays
the
menu
system.
A
A
A
B
B
A
B
B
A
B
A
B
B
A
B
B
B
B
Xa,
like
the
lion,
my
men,
where
we
actually,
this
kind
of
the
early
early
version
of
this,
was
for
ABS
they're
cool,
the
they're,
the
the
filament
kind
of
gradually
cools
down
to
its
final
temperature.
So
you
get
this
slowly.
Changing
melts,
drink,
yeah,
so
filming
he
goals
at
all
doesn't
really
cause
any
problem,
whereas
with
us
we've
got
about
three
centimeters
of
liquid
filament
coming
out
of
nozzle
and
then
right.
What
hits
the
water
pretty
much?
You
know
at
least
the
outer
surface
is
solid.
B
B
A
B
A
B
It's
groovy
so
basically,
at
the
end
of
the
screw
right,
there's
there's
there's
basically
one
flight
on
the
screw,
so
as
that
screw
goes
around
the
circle's,
it
creates
a
pressure
wave
that
has
a
frequency
of
the
revolution
of
a
screw
so
that
pressure
wave
causes
varying
diameter
and
different
thicknesses
of
filament
bend
with
different
curvatures.
So
when
the
film
goes
around
the
roller,
sometimes
it
bends
more
easily
and
sometimes
doesn't.
B
A
A
B
A
A
A
B
A
B
A
A
B
A
B
A
B
Exactly
and
you
know
so,
it's
kind
of
like
the
way
you
could
combat,
that
is
by
making
an
extruder
that
just
pushed
more
filament
through
you
know
so
kind
of
like
using
just
turning
off
the
faucet.
But
then
you
got
into
the
bigger
heaters
and
a
bigger
motor
to
push
it
through.
Then
all
your
power
supplies
get
bigger.
You
know
it's
kind
of
chain
reaction
upstream.
A
So
there's
there's
that
issue
for
us.
I
think
it
the
most.
The
most
important
thing
is:
is
we
want
to
be
printing
larger
scale
items
so
probably
3
millimeter
is
desirable
for
that,
just
also
in
the
manufacture
of
it,
I
mean
it's.
If
you
have
much
less
length
seems
like
it's
easier
to
at
the
end
of
the
day
to
extrude
like
the
spools
like
it
seems,
it
seems
like
that
if
you
have
less
land
that's
easier,
it
makes
it
technically
a
little
easier
to
to
manage
the
thicker
filament.
Or
is
that
not
the
case.
B
B
B
B
B
B
A
B
B
B
The
other
thing
that
happens
is
as
the
as
the
screech
going
around
in
circles
of
polymers
beats
by
now
gray
and
all
those
that
kind
of
orient
the
chain
polymer.
Even
though
they're
liquid,
you
know
it's
like
spaghetti
or
something
like
that.
So
so
it
kind
of
has
this
spiral
shape,
and
so
you
here
just
to
let
it
come
out
of
the
extruder,
as
is
you
would
get
through
that
it
looks
like
spiral
macaroni
and
are
you
could
make
our
AV
Mystik
spiral?
B
B
A
B
A
A
B
B
Yeah,
so
that's
the
that's!
The
function
of
the
breaker
plane,
men
in
most
industrial
machines,
there's
actually
a
melt
filter
in
front
of
that
which
is
basically
just
like
a
a
piece
of
wire
mesh
like
like,
actually
something
that
you
might
use
and
other
small-scale
extruders
have
now
it's
like
it
will
get
a
stainless
steel
wire
mesh
for
like
a
faucet
mm-hmm.
A
A
B
B
B
B
This
is
not
just
a
power
extruder
like
flow
isn't
some
times
it
seems
like
it
should
be,
like
sometimes
plastic
might
melt
earlier
or
a
little
later
for
the
barrel
earlier
or
later
or
screw.
It's
not
supposed
to.
You
know
like
different
things
like
that
can
cause
the
filament
barrel,
not
a
varying
rate
yeah.
A
B
B
A
A
B
A
A
A
B
A
B
B
B
B
B
B
B
A
A
B
B
B
B
A
B
You
know,
there's
just
if
you're
trying
to
make
a
uniform
product
that
pulse
quite
a
bit,
and
so
my
thought
was
okay.
Well
we're
not
using
pellets.
How
can
we
turn?
How
can
we
make
flakes
that
are
as
close
to
pellets
as
possible
right
also,
if
you
look
at
the
size
of
the
flakes
on
me,
they're,
like
tiny,
four
millimeters,
whereas
if
you
look
at
the
size
of
the
plastic
coming
out
of
Russia's
plastic,
we
shouldn't
go
that
it's
far
larger
right,
then
the
other
thing
that
happens,
but
so
you
say:
okay
well,
I!
B
My
grinder
only
makes
big
flakes
family
gets
smaller
stuff,
I'm,
going
to
run
it
through
again
and
that
helps
you
pay
some
fines.
So
as
you
get
like
fine
particles
in
there,
those
things
melt
more
quickly
than
the
rest
of
lakes,
so
they
may
to
barrel
earlier,
and
then
you
may
even
get
yeah
other
thing.
I,
don't
know!
If
you
know
this
too,.
B
B
B
B
A
B
And
then
you
start
running
the
machine
as
the
plastics
cooling
down
it'll
freeze
on
to
the
screws
plug
the
plastic
goes
around
with
it
doesn't
move
down
the
length
of
the
barrel
but
yeah.
So
you
want
the
plastic
to
slide
on
the
screw
and
rip
on
the
barrel
and
that
translates
into
the
plastic
being
forced
down
the
barrel
towards
them.
A
B
B
They
also
you
can
play
with
temperature
right.
So,
if
you
think
about
a
plastic
that
melts
or
gets
soft
at
a
lower
temperature,
what
you
might
do
is
be
wrote
to
a
certain
temperature
and
actually,
you
know,
actively
maintain
the
feed
throat
that
I
don't
know.
Whatever
temperature
and
plastic
may
soft
him
enough
on
it
that
it,
you
know,
gets
a
better
grip
barrel
than
it
was
on
the
screw.
B
B
B
B
B
B
B
B
B
B
B
B
B
A
A
B
B
A
B
B
B
Actually
they
use
a
they
use.
A
different
type
of
measurements
called
melt.
Next,
sorry,
what
index
melted
next
and
actually
for
me,
a
t
they
use
even
more
specialized
turn
which
intrinsic
viscosity,
because
they
have
this
crazy
way
of
measuring
basically
an
analog
of
the
viscosity
of
the
plastic,
because
when
you
heat
it
up,
it
starts
breaking
down
and
changing
so
the
actual
viscosity
you
measures
so
dependent
on
the
particular
situation.
You're
measuring
it.
B
A
B
Those
chain
extenders
just
grab
one
end
of
one
chain
and
they
start
looking
and
as
soon
as
they
find
another
chain
they
just
grab
onto
that
and
so
like.
If
you
I
have
I
was
talking
with
some
folks
that
said
yeah
you
can
use
them.
You
just
have
to
be
really
careful,
because
if
you
get
too
much
in
your
plastic
will
set
up
like
epoxy
inside
of
your
extruder,
you
won't
be
able
to
melt
like
it'll,
be
done.
Yeah.
B
So
another
way
and
I
haven't
again.
These
are
just
like
things
I've,
you
know
say
like
seeing
brief
mention
of,
but
I
haven't
really
researched
them
because
they're
beyond
our
capability,
but
I,
think
if
you,
the
plastic
up
vacuum
and
hold
it
at
a
certain
temperature
and
keep
it
moving
for
a
certain
amount
of
time.
That
chains
will
naturally
start
joining
yeah.
B
Basically,
with
either
of
these
just
like
have
your
target
intrinsic
viscosity,
you
can
just
continue.
You
figure
out
how
long
you
have
to
process
to
get
there
yeah,
how
much
so
for
us,
it's
kind
of
the
opposite.
We
just
want
to
like
process
as
little
as
possible
to
just
maintain
the
quality
that
cap
yeah
are.
B
B
You
know
if
you
wanted
to
do
like,
if
you
think
about
say,
like
a
small-scale
recycling
program
where
you're
using
pellets,
then
what
you're
doing
is
you're
collecting
your
bottles,
sending
them
to
trying
recording
pellets
from
China
and
then
making
whatever
it
is.
But
at
that
point
it's
probably
easier
to
make
the
thing
in
China
as
well,
so
we're
like
nope
we're
just
gonna
do
go
straight
to
Lake
extrude
straight
from
flake,
and
if
all
that
middle
Business.
B
A
A
B
B
B
You
know
so
there
may
be
I've
been
daydreaming
on
a
way
you
know
like.
Let's
say
you
know,
if
you
have
a
paper
shredder
or
something
similar
to
that,
that
cuts
strips
instead
of
the
cross
cuts
yeah
some
sort
of
mechanism
that,
as
the
strips
come
out,
they
get
fed
in
perpendicular
on
another
one.
Then
you
would
get
nicely
diced
up
lakes
or
maybe
there's
a
cutter
geometry.
B
B
B
A
B
Flakes
won't
be
as
uniform
and
I.
Don't
think.
That's
absolutely
critical,
I
mean
I.
Think
world
may
be
a
little
bit
hyper
in
terms
of
like
trying
to
get
all
through
a
uniform,
feedstock.
I.
Think
part
of
the
thought.
That's
just
like
completely
eliminate
one
variable.
Well
until
we
until
we,
you
know,
keep
eliminating
variables
until
we
figure
out.
B
B
B
It's
like
another
issue
that
we
found
with
with
those
things
when
we
were
just
trying
to
buy
them,
that
the
screens
they
come
with
are
typically
four
twelve
millimeter
size,
bikes,
which
is
kind
of
a
standard,
but
it's
it's
really
big,
something
they
clump
together
and
they
don't
fit
into
our
extruder
anyway.
So
what.
A
B
In
the
industry,
so
what
they'll
do
they
have
like
huge
grunt,
where
the
size
of
that
flake
is
still
small
compared
to
the
rest
of
the
machine?
Okay
yeah,
so
you
fit
in
the
barrel
and
then
they'll
take
those
and
extrude
them
into
pellets
yeah,
and
that's
that's
probably
the
point
at
which
they
either
mix
in
a
chain
extension
compound.
B
A
A
B
B
B
A
I'm
thinking
actually
right
now,
just
if
you
talk
about
product,
ecology
is
a
system
where
you
have
a
hydraulic
power
cube
like
we
use
it
normally,
like
you
saw
on
the
tractor
and
use
that
for
hydraulic
drive
so
that
you
can
actually
an
interchange,
the
motor
from
the
granulator
to
the
extruder,
even
in
the
extreme
case,
but
Hydrox
would
be
an
easy
way
to
drive.
It
I
mean
hydraulic
engines.
Hydraulic
motors
are
relatively
inexpensive,
they're,
like
a
hundred
bucks
for
multiple
horsepower
scale.
A
B
There
was
a
like
an
old
plastics
recycling
bin
or
you
can
build
them
yourself,
and
then
it
was
like
kind
of
like
almost
it's
so
easy
that
you
know
they
didn't.
Oh
it's
probably
not
that
it
was
so
it
was
not
the
right
spot
to
go
right,
but
I
I
may
have
some
stuff.
I'll
dig
around
one
interesting
thing
that
I
did
learn
about.
Granulator
is
when
I
was
in
Tanzania.
B
B
It's
kind
of
like
luck,
if
the
smallest
he's
this
farther
than
the
screen
all
throughs,
sometimes
they
get
recycled
back
through
and
cut
even
smaller,
and
so
you
end
up
with
lots
of
fines
and
the
fines
are
a
problem.
I
mean
they're,
a
problem
to
the
point
where,
when
you
buy
flake
industrially,
a
typical
speck
is
the
percentage
of
fines.
B
B
A
B
It's
a
thing
it
would
be.
One
thing
that
would
be
pretty
awesome
to
be
able
to
do
is
to
use
like
the
most
common
blades
I
feel
like
that.
I
could
find
four
things
would
be
like
planer
blades
for
woodworking,
or
something
like
that.
You
know
so
you
get
like
a
good
hardened
steel
and
then
just
regrinding
edge
nunim
versus
like
trying
to
find
cool
steel
and
grind
your
blends
from
scratch.
B
A
B
A
B
Okay,
I'm
afraid
the
next
iteration
I
mean
there's
some
there's
some.
You
can
I'll
clean
up
kind
of
just
like
things
that
have
been
in
the
design
or
just
because
of
the
history
of
the
design
that
you
know.
I
just
want
to
design
them
for
this
iteration
of
machines
and
cleaning
up
frame
shape
all
the
new
electronics
fit
in
the
box
top
that,
in
terms
of
the
main
issues,
one
of
the
next
things,
the
diameter
sensors
performing
better
and
getting
that
feedback
control.
A
B
And
then
the
other
thing
is
on
the
printer
because
of
that
diameter
sensor,
there's
run-out
detection.
So
if
we
hit
the
end
of
a
section
of
filament,
the
printer
will
just
tell
you
to
add
more
solvent,
don't
lose
your
print,
but
yeah
I
would
say.
You
know
right
now,
probably
around
five
pieces
of
filament
that'll,
probably
change
pretty
quickly
as
the
diameter
sensor
kind
of
illuminates,
those
large
and
small
sections
and.
B
B
And
so,
if
it's
you
know,
say
the
barrel
that
the
center
bore
and
there
is
like
two
millimeters
in
diameter,
and
so
if
you
put
something
that's
one
point
two
millimeters
in
there
it'll
start
buckling
inside
of
the
inside
of
that
space.
It
makes
it
so
that
adjusting
extrusions
doesn't
really
work.
Yeah.
B
A
B
A
Yeah
I
mean
I
was
trying
even
like
what
I
mean.
What's
the
limit
of
the
thickest
filament
I
mean
say
we
make
our
own
extruders
for
3d
printers,
which
we
can
I
mean
we
probably
get
to
that.
But
I'm
thinking
for
larger
scale,
prints
well
I
mean,
what's
the
maximum
practical
limit,
I
mean?
Is
it
four
millimeters
five
millimeters?
Maybe.
B
So
they
make
these
big
huge
printers
called
gigabits
yeah,
and
we
were
talking
about
you
know
they.
Printing
large
issues
there
in
terms
of
throughput
will
become
the
cooling
of
the
plastic,
while
you're,
printing
and
I
also
actually
talked
with
with
Ultimaker
years
and
years
ago
they
had
a
huge
printer
that
you
know
their
idea
was
basically
the
same
thing.
You
throw
bottles
and
they
get
crown,
but
nothing
boom
printed,
and
they
said
basically
another
big
problem
is
the
crystallization
of
plastic.
B
B
B
A
B
B
B
A
A
A
B
A
B
B
A
A
A
A
B
A
B
Unless
we
can
find
a
way
to
maybe
not
use
a
lot
there,
but
anyhow
you
got
to
make
sure
it's
dry
and
then
also
when
you
add
a
heat
cycle,
so
it
grades
more,
some
that
may
or
may
not.
It
may
not
be
an
issue
for
3d
printing
I'm,
not
sure,
but
then
it
just
adds
another
step
process.
But
you
potentially
like
one
thing
we're
thinking
about
doing
is
all
that
like
when
you
start
up
the
Machine.
B
You
end
up
like
as
the
Machine
kind
of
attained,
the
brim.
You
end
up
with
a
bunch
of
filament
that
you
extruded
that's
not
in
spec
and
so
that
stuff
we
want
to
just
grind
back
up
and
use
those
pellets
for
extruding.
So
one
of
our
thoughts
would
be
that
we
would
save
that.
You
know.
So
that's
every
X
number
of
times
you
extrude!
You
know
you
end
up
with
a
bucket
of
pellets
and
then
you
can
extrude
those
that
stuff,
maybe
for
the
extrudes
better
in
the
flakes.
Maybe
we
just
dude
everything
twice
right.