►
From YouTube: 3D Printer Quality Control Session
Description
-----------------
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
We'll
primarily
be
using
the
four
by
four
inch
square:
tubing:
half
inch
wall
like
half
inch
wall,
so
that's
pretty
pretty
strong
and
that
that
kind
of
attractor
becomes
scalable
to
like
bulldozer
grade
like
a
couple
100
horsepower.
If
we,
if
we
have
materials
for
four
power
cubes
of
27
horsepower,
so
we
have
four
engines.
A
We
have
the
pumps
that
go
on
those,
but
we
could
build
three
or
four.
My
goal
was.
The
proposal
was
about
70
horsepower
for
this
tractor,
which
would
be
around
that
that
number,
which
would
be
either
like
four
times
18
or
three
times
27,
so
three
power
cubes
that
are
larger
since
we
have
four
of
the
larger
ones.
A
It's
not
much
difference
to
build
the
18
between
between
18
and
27,
so
we
can
do
even
go
above
the
power
that
we're
looking
at
to
like
108
horsepower
with
four
power
cubes,
the
innovation
there
is
about
making
the
scalability
seamless
to
date.
We've
done
power,
cubes
and
they're
complex
thing,
with
the
entire
system
built
in
engine
hydraulics,
cooling,
automatic
ignition,
filtering
fans,
solenoids
things
like
that
hydraulic
outlets.
A
A
You
can
separate
it
into
the
fluids
gas
cooling,
like
all
those
hydraulic
support
systems
from
the
engine
pump,
which
makes
a
lot
of
sense
because
the
engine
pump
are
the
things
that
that
go
bad.
So
it's
better.
For
so
that's
that's
the
concept,
the
idea
that
we're
we're
gonna
make
that
more
seamless
and
it's
kind
of
a
crazy
idea,
because
nobody
does
this
in
the
world.
Nobody
uses
hydraulic
power,
scalable
power
units
in
the
world
that
I
know
of
just
when
people
build
a
tractor.
A
It's
like
100
horsepower,
10
horsepower
and
that's
it
that's
all
you
got
so
anyway.
That's
that's
the
innovation.
We're
doing.
We've
got
the
frames.
We've
got
the
power
cubes,
we've
got
the
drive
systems
hydraulics.
We
know
how
to
do
that
and
if
we
all
I
mean,
if
we
actually,
if
we
all
work
on
it,
then
it'll
be
like
five
days
that
we
we
will
crank
that
tractor
and
run
with
yeah.
D
C
A
Does
yeah,
I
know
yeah?
Actually
that
there
is,
I
mean
we
can.
If
we
want
to,
we
could
one
option
is
to
divide
into
two
teams
because,
for
example,
there's
a
steel
order
coming
in
today
we
can
bring
the
so
we're
going
to
use
these
we're
going
to
do
steel
wheels
for
like
lifetime
design,
so
we
can
bring
bring
out
the
materials
closer
like
the
engines,
bring
them
over,
bring
the
older
universal
rotors
which
are
in
the
other
work.
A
Other
storage
shed
like
we
can
be,
bringing
that
all
over
that
that's
part
of
the
setup
time
and
and
getting
out
the
steel
from
by
the
torch
table,
area
and
stuff
like
that,
so
well.
Actually,
yeah
people
can
be
doing
that
if
another
team
wants
to
do
that,
but.
E
A
Yeah
we
can
tell
jeff
what
all
to
bring
and
all
that
are
all
the
welders,
but
we
have
six
welders,
so
part
of
it
is
we're
getting
getting
six
welders
up
and
running
and
and
people
are
welding
cutting.
There
would
be
a
lesson,
basic
tools,
lesson
on
welding
and
other
things,
what
else
to
say,
but
on
the
printer
we
never
went
over
what
it
means
to
actually
get
this
thing
running
with
the
electrical
system
with
the
controller
and
we
never
went
over
what
it.
What
is
absolutely
critical
for
a
printer
to
print?
A
Well,
so
you
can
understand
it
I.e,
maybe
like
the
most
general
quality
control
points
that
everyone
has
to
get
for
their
printer
to
work.
We
went
through
pieces
so
far.
At
the
end
of
the
day,
you
have
to
talk
about
integration
and
what
is
critical
for
the
thing
to
work.
Well,
so
a
little
lesson
on
that,
like
you
know
like
an
hour
or
two
going
through,
so
that
people
are
not
asking
me,
we
can
go
over
this
and
here's
the
five
or
so
five
or
ten
quality
control
points
on.
A
I
would
say:
there's
the
there's,
the
tightness
of
the
axis
and
there's
how
you
tighten
the
belt
like
those
two
things
and
then
beyond.
That
is
the
controller,
how
everything
there
works,
but
we
have
actually
pretty
good
documentation
on
the
controller,
and
I
can
point
you
out.
We
never
took
a
look
at
what
that
is,
but
we
do
have
that
pretty
well,
but
those
three
things
if
I
can
transfer
that
knowledge,
then
that
that
could
propagate,
but
we
need
to
do.
We
need
to
go
over.
All
of
that.
H
H
G
A
A
I
mean
you
see
like
real,
tangible
there's.
A
I
So
what
I
would
suggest
is
we
quit
at
five:
don't
kill
ourselves.
A
From
five
there's,
there
is
a
few
hours
for
people
who
are
not
in
an
enterprise
session
to
keep
going
at
it.
If
you
want
to
do
the
printer,
so
that's.
A
A
A
And
the
tractor.
A
Gonna
be
like
half
the
people
left
working
on
it
because
it's
harder
so
we're
not
gonna.
You
know
conserving
our
energy
throughout
the
projects
being
smart
about
how
we
rest
and
and
like
quit.
D
I
D
A
A
Three-Week
tractor
built
this
november
14th
two-week.
H
F
A
I
A
A
Be
on
your
own
because
there's
not
going
to
be
a
lot
of
people
here.
I
don't
know
how
that
would
look
what
you
can
pretty
much
do
on
your
own,
but
I
mean,
as
far
as
everything.
A
Build
one
of
these
things,
the
part.
B
C
I
J
E
J
B
J
You
know
I
mean
in
terms
of
making
this
happen.
You
know.
A
A
That's
the
challenge,
we're
dealing
with
some
people,
like
yourself,
you're,
like
man,
let's
do
it
bam
all
day
at
night,
other
people.
F
A
A
little
different,
but
it's
like
the
you
know
when
you
plan
for
a
lot
of
people,
the
reality
is
like
kind
of
plan
on
have
the
people
really
working.
So
that's
the
bottleneck
right.
F
A
A
We
did
a
lot
of
the
quality
control
throughout,
like
just
the
very
basics
that
are
critical
and
now,
if
we
can
propagate
that
to
the
rest
of
the
build
and
do
the
controller
which
is
well
documented,
which
we'll
go
through
just
right
now,
I
think
we're
quite
close
to
it.
I
mean
I
finished
my
printer.
We
can
basically
look
at
what
I
did
and
then
okay
upload
the
code
and
here's
how
you
get
it
to
print
the
first
time.
A
So
that's
that's
what
we
can
do,
but
all
of
you
should
be
able
to
get
there.
The.
E
A
A
session
I
get
asked
10
to
20
to
30
times
the
same
question
when
we
explicitly
went
over
that,
I
said
pay
attention.
This
is
important.
Like
the
retention,
I
don't
know
how
to
make
it
work.
I
mean
that's,
that's
a
stumbling
block
right
now.
I
see
it.
I
always
have,
and
I
don't
have
it
on
me
right
now,
but
I
have
my
fat
ass
notebook.
That's
the
24.
K
A
That's
how
I
can
learn
much
more
rapidly,
but
that's
only
like
that's
one
detail.
I
would
propose
for
anybody
to
just
get
a
piece
of
paper
like,
like,
let's
say
we're,
building
a
tractor.
What's
the
dimension
of
this
thing,
we
gotta
build,
write
it
down,
you're,
not
gonna.
Remember
it's
like
people,
don't
remember
stuff
and
learning
is
repetition.
It's
like
just
blew
me
away
how
that
how
how
that
was
an
issue
at
this
time.
I
I
don't
know
what
to
do
about
it.
It's
can.
A
Yes,
that
that
too
yeah
could
try
to
to
be
more
and
more
deliberate.
It
really
kind
of
requires
yeah,
very
deliberate
effort
to
make
that
happen.
Yeah.
A
Take
notes,
that's
a
difficult,
I
think,
like
with
the
internet
culture,
we
kind
of
get
less
and
less
into
actually
writing
things
on
paper.
I
think
paper
is
a
critical
tool
also.
F
L
G
E
E
A
G
When
we
built
modules,
there
was
a
tablet
that
showed
the
dimensions
of
whatever
modules.
Okay.
So
after
he
knew
how
to
do
this,
then
we
left
and
we're
not
doing
it
anymore
yeah,
but
we're
not
using
it
at
the
level
of
intensity
we're
using
it
while
we're
building
modules.
So
nobody
is
really
opening
the
document
one
or
one
site.
Why
is
that
which
supplements.
A
Guide
that
you
can
follow
right
now
and
I
could
disappear
and
you
could
probably
even
do
it,
but
still
with
this
you
know
I
I
want
to
be
there
and
I
should
be
there,
but
this
is
kind
of
this
pictorial
thing
because,
like
for
example,
like
some
one
person
might
not
even
know
what
the
hell,
this
is.
Okay,
and
this
is
also
for
the
pro,
but
the
wiring
is
identical
for
the
universal.
A
So
with
a.
A
This
is
the
universal
that
we're
working
on,
but
the
controller's
identical.
It's
got
identical
components
and
all
that.
So
here's
a
complete
thing.
It's
got
a
video
for
every
single
step
and
there's
a
total
total
of
20..
You
can
follow
that
you
can
look
at
the
videos
at.
A
The
thing
is
like
this
was
made
for
the
pro
we
need
to
adapt
it
for
the
universal,
because
it's
like
somebody
might
not
know
what
the
hell
that
thing
is
up
there
like.
What's
like,
how
do
you
correspond
that
to
what
you
have
on
the
universal
there's
still
xyz,
but
it
looks
different
on
the
pro
than
the
universal
as
long
as
you
can
get
oriented
on
what
that
is,
the
the
controllers
are
the.
I
I
B
B
I
A
D
A
Level
of
person
a
person's
experience,
because
then
one
person
would
be
like.
Oh
yeah,
you
know
don't
waste
my
time
with
the
other
person
who
won't
even
get
it
and
stuff
it's
pretty
challenging,
but
we
could
do
that,
but
it's
like
the
feedback
so
far
throughout
this
you
know
this
summer
has
been
it's
like.
I
try
to
do
that,
but
I
felt
like
people
are
like.
What's
this
all
all
about,
because
it
really
depends
how
much
how
much
you
know
already
about
it.
A
That's
that's
the
challenge
there
and
the
documentation
is,
is
the
thing
that
there
should
be
documentation
for
every
level
of
from
the
very
simple
like
this
is
simple
stuff.
This
is
this
is
a
power
user.
This
is
somewhere
in
between,
but
we
can
only
do
so
much
and
there's
that's
where
the
idea
was
okay.
People
who
are
here
spend
some
time
some
time
documenting,
actually
record
the
technical
knowledge,
that's
being
generated,
and
that's
we're
not.
I
don't
think
we're
doing
too
great
on
that.
A
Overall,
it's
like
a
long,
drawn-out
process
where
only
a
few
people
end
up
doing
anything
meaningful
towards
that.
But
I
mean
I
don't
know
how
to
get
people
to
do
that,
because
people
are
just
so
busy
with
their
lives,
for
which
reason
it's
like
you're
only
going
to
be
motivated
to
do
it
really
well,
if
you're
actually
going
to
consider
doing
this
for
a
living.
A
I
D
A
Yeah
so
today
we
we
gotta
quit
at
like
4
45,
where
the
the
enterprise
team
is
going
to
the
city
to
to
look
at
lots
and
talk
to
an
architect
and
stuff
like
that.
A
The
ones
that
are
politically
connected,
like
the
people
that
live
there
and
their
property
values,
might
go
down
because
there's
houses
coming
in
for
100k
when
their
mortgage
is
for
180k
and
then
that
devalues
their
mortgage
and
all
that.
So
it's
like
we're
solving
that
issue
of
access
to
housing
where
we
help
in
general.
But
that's
that's
the
that's.
C
A
C
A
A
D
A
We
went,
I
think,
everybody
kind
of
got
the
hang
of
here's
the
we
got
data.
Two
two
pounds
are
under
for
the
stiffness
and
you
want
to
make
sure
it
does
not
wobble
if
you
can
push
it
from
one
side
or
the
other.
So
maybe
like,
let's
write
this
down
explicitly
like
the
ultimate
test
for
all.
A
A
So
okay?
So
let
me
what
is
that
so
you've
got
the
axis.
You
got
your
carriage,
which
is
in
the
middle
right.
So
that's
your
axis
and
by
the
way,
let's
share
this,
so
it's
editable
everybody
can
actually
edit
change.
It's.
The
link
is
in
in
the
chat
box
if
you're
on
zoom
here,
so
you
can
access
that.
H
F
A
Push
it
there
then
push
it
there
and
see
whether
you
got
a
torquing
on
that
axis.
So,
ideally,
you
want
no
torque
meaning
like
no
rotation
like
it
doesn't
start
to
jam
up
so
ideally
you're
under
two
pounds
and
even
like
one
or
less,
and
when
you
push
from
one
side,
you
don't
see
a
visible
torquing
like
like
this
thing
like
you,
don't
want
it
to
push
you
push
from
one
side
or
the
other.
You
don't
want
that.
You
don't
want
this
to
rotate.
A
You
want
it
to
behave
as
if
you're
pushing
from
the
center
and
it's
moving
straight
so
observe
for
that.
A
Loose
like
less
than
two
pounds,
but
you
can
do
that
very
easily
at
the
expense
of
being
so
loose
that
you're
just
wobbling
and
that
will
determine
the
quality
of
your
prints
like
if
you
can
see
visible
like
one
millimeter
wobbling.
A
A
If
you've
got
like
a
little
visible
wobble,
I
mean
it'll
still
work
for
like
a
big
part,
but
you
know
it
won't
be
as
clean
as
possible
and
super
neat.
So
do
this
two-sided
push
test
while,
while
under
two
pounds
of
friction,
so
we
measured
that
was
cool,
we
actually
measured
the
friction.
We
got
numbers
for
that
in
our
data
collection
form
there
do
that!
That's
the
first
thing.
So
we
went
through
this
extensively
like
if
you
get
get
all
your
axes
like
this
you're
gonna
have
really
nice
clean
prints.
A
What
else
can
fail?
So
so,
let's
go
to
qc2.
A
I
mean
it
will
probably
increase
a
little
bit
because
you
got
more
weight
but
try
to
be
under
two
pounds
for
the
most
excellent
results.
I
think
you'll
probably
still
work
with
like
four
or
five,
because
we
said
we've
got
18
pounds
of
available
force.
So
as
long
as
we
have
that,
but
what
th?
That
does
mean
the
torque
curve
of
the
motor
is
like
this.
It's
it's
not
the
same,
torque
at
all
speeds
when
you're
very
slow,
you're
near
full
torque.
A
I
A
You
can
print
slowly
and
if
you're,
only
doing
like
25
or
50
millimeters
per
second
perfectly
fine.
If
you
want
to
try
to
push
it
to
things
that
you
could
print
with
like,
say:
200
millimeters
per
second
like
4x,
the
speed
you're
going
to
see
your
layer
shifts,
which
means
that
you're
printing
and
then
you
you
jump
steps
and
the
layers
shift
and
all
that
so
so
qc2
is
belts.
So
let's
look
at
the
very
details
of
this
because
we're
going
to
have
about
a
100
to
500
of
this
question
throughout
the
day.
A
But
there's
like
three
points
so
one
location
of
location,
of
of
cylinder,
the.
I
A
A
D
A
Yeah,
it
should
be,
anyone
can
find
an
edit
it's
in
a
link
in.
I
A
Link
so
we
got
the
bearings.
K
A
A
The
one
bolt:
okay,
there's:
that's,
let's
make
that
into
the
the
actual
nut.
Now,
maybe
not
we'll
draw
the
little
bolt.
A
A
D
A
D
A
Next
to
that
bolt,
that's
step
part
one.
Why
not
put
in
the
other
side,
because
then,
if
you
try
to
put
the
clamp
on
that
side,
you'll
be
riding
on
a
bolt
and
it
can
move
here.
It's
important,
so
part
two
belt
is
by
is
between.
A
J
I
B
A
A
A
In
and
make
it
user-friendly
so
so
carriage
closure
bolt.
Is
that
thing
right,
so
is
that
drawn
right.
C
Well,
no,
no,
the
lower
belt
should
be
in
the
character
that
carriage
right
in
the
plastic,
whatever
that's.
I
A
A
A
I
H
D
H
A
D
I
C
A
C
Know
faces
where
head
to
outside
of
carriage.
D
A
A
F
D
I
J
Is
it
more
accessibility
in
the
future
or
is
it?
Is
there
an
operational
issue
if
the.
A
A
If
you
didn't
know,
you'll
appreciate
that
we're
trying
to
make
bolts
accessible
engineers
do
not
make
bolts
accessible
we're
trying
to
pay
attention.
So
if
you
have
the
head
out,
you
can
put
a
allen
key
right
there,
as
opposed
to
from
the
other
side
where
you
got
the
other
rods
and
the
belts
in
the
way.
So
that's.
A
A
Belt
packs
and
all
that
it's
actually
build
instructions.
It's
in
there.
A
A
F
A
Winding
the
belts
so
do
we
have
some
so
here's,
for
example,
a
picture
of
what
we
talked
about.
Just
now,
that's
73,
but
the
point
is
that's
73
out
of
many
more
numbers,
but
it's
like
it's
tmi.
It's
like
you
got
to
follow
it.
So
there's
only
like
in
the
cheat
sheet
notes
that
you
can
take.
You
can
only
take
some
of
the
most
critical
things,
but
you
don't
know
which
things
are
critical.
It's
a
very
hard
problem.
So,
but
here
it's
a
picture
of
what
we
just
described.
A
So
what
we
can
do
is
do
a
little
screen
capture
and
make
a
picture
of
that
happen.
See
step
73
in
the
build
manual.
A
And
then
going
further:
okay,
that's
not
right
there,
and
maybe
that's
before.
A
A
A
A
A
The
head
of
the
bolt
should
face
the
opposite
belt,
no
yeah.
I
do,
and
that's
I
wouldn't
do
that
anymore
and
then
that
should
face
the
outside
or
maybe
we
do
want
to
do
it.
So,
let's
revisit
that
which
is
which
do
you
think
well,
I
mean
which
is
easier
for
somebody
like
that
was
actually
easier
for
me
at
that
time,
because
what
you
do
is
you
tilt
that
belt
and
you
can
get
your
allen
allen
key
into
the
nut
the
bolt
head.
A
D
That's
practical
head
out.
D
D
C
A
Yes,
yes,
I
would
so
you'd
take
the
belt
and
squeeze
it
down
like
that.
Or
would
you
go
like
clamp
and
go
down
on
a
belt.
H
A
A
J
A
C
F
C
A
A
It
could
be
by
the
z
by
putting
the
cap
on
the
cap,
which
is
the.
A
G
G
Make
it
easier
to
to
slide
them
a
little
bit
so
that
you
can
do
the
locking
interlocking
in
a
different
way
and
strengthen
from
the
side
rather
than
have
to
go
in
there
on
the
on
the
line?
All
right,
this
makes
sense,
don't
require
a
cap
design
so
that
you
slide
the
slider
belt
down,
and
then
you
lock
it
on
one
side,
and
then
you
pull
from
this
other
side
trading
the
design
for
build.
A
A
A
Okay,
let's
do
what
we've
got
for
now,
so
so
we
said,
put
pull
it,
but
easier
way
is
to
grab
it
with
the
vise,
the
needle
nose
pliers
and
twist
it
that's
the
easier
way
to
do
it.
A
Twist
it
just
twist
it
if
you
grab
it
and
twist
it
with
needle
nose
pliers
you'll,
do
you
can
do
much
more
force?
In
fact,
you
can
go
so
hard
that
you
can
break
the
belt
by
hand.
You
have
so
much
strength.
You
can
do
it.
You'd
really
have
to
pull
hard
to
get.
What
it's
easy
to
do
with
twisting
of
pliers.
L
A
So
push
the
clam
next
to
the
carriage
while
you
grab
so
let's
say,
grab
well
no
we're
going
to
do
this,
since
it
would
be
like
that's
like
how
people
might
think
about
it.
Let's
go
strike
through
that
and
while
you
grab
with
pliers
with
needle
nose,
pre.
A
Okay,
tension
tension
data.
So
what
are
the
two
limits?
How
do
you
know
it's
too
soft,
like
the
strumming,
that's
cool,
but
you
need
to
have
a
meter
that
would
work
because
you
can't
you
can't
do
that.
You
can
hear
the
sound
nice
but
say
you
don't
have
that.
What's
what
are
the
two,
it's
useful
to
think
about
limits,
so
what's
the
strongest
you
can
get.
Let's
look
at
that
limit.
What's
the?
A
So
if
you
had
it
like
a
two
pounds-
and
you
know
you
can
see
it
move
easily
and
then
you
tension
the
belt
and
it's
very
hard
to
move
measure
it
and
don't
ask
me
measure
it.
If
it's,
I
would
say
max,
I
mean
don't,
allow
it
to
go
above
four
like
three.
Maybe
so
we
said
two
already,
but
here
is
where
we're
actually
tensioning
things.
That
was
two
was
a
start,
so
we
still
have
a
little
way
to
go.
I
wouldn't
go
above
like
three,
but
make
sure
so
we'll
say.
F
E
E
G
E
G
D
But
can
you
get
on
your
ball.
G
A
E
A
A
G
A
G
F
A
A
Okay,
so
after
all
this
like,
maybe
we
can
come
up
with
a
different
belt
clamping
mechanism,
but
but
I
think
this
is
cool
because.
A
I
don't
know
it's
like
field
adjustable
easily,
but
maybe
some
people
want
to
say:
okay,
we'll
make
this
carriage
design
different,
so
that
because
this
is
you
know
it's
it's
a
bit
challenging
to
get
it
first,
we
used
friction
clamps
like
wedges
to
wedge
the
belt
in
and
then
it
that
kind
of
gets
hard
because
you
got
to
be
very
precise
on
how
on
the
on
the
size
of
the
wedges
you
can,
they
can
easily
be
too
small
or
too
large
the
belt
might
get
loose
or
it's
too
hard
to
get
them.
L
A
I
A
We
had
another
way
to
do
it.
We
moved
to
this
we're
claiming
that
the
former
way
was
even
harder.
Okay,.
F
A
Right
so
it's
there's
detail
there,
okay,
but
the
point
is
that
if
you
have
the
carriage,
accurate
in
terms
of
tension
and
you
tension
the
belt
properly,
you've
got
really
good
motion
at
that
point
and
your
prints
will
be
pretty
good.
So
the
only
other
thing
that
that
can
degrade
your
print
quality
is
if
your
drive
gears,
the
pulleys,
are
not
tight,
but
we
checked
that
that's
all
good.
A
The
motor's
got
to
be
mounted
tight,
so
that
motors
are
not
wobbling
around,
but
those
are
the
three
screw
little
screws
and
that
should
all
be
good.
So
at
this
point
we're
gonna
have
pretty
good
prints
for
all
of
us,
so
the
next
thing
would
would
be
the
electronics.
D
J
You
just
quickly
explain
why
any
of
the
parts
that
we
just
discussed
being
mounted.
You
know
tilted
yeah
or
you
know
off
center
things
like
that.
None
of
matters,
oh.
A
Yeah,
oh
well,
it
does
matter
to
it,
makes
it
just
a
little
harder
well
like
we're
struggling
with
some
things
because
of
part
quality
like,
for
example,
it
made
it
harder
on
the
extruder,
where
it
was
kind
of
hard
to
get
some
of
those
bolts
in
because
the
things
weren't
they
were,
the
axes
were
skew.
So
that
means
I
was
printing.
I
printed
most
of
it
can
printed
the
good
parts.
A
A
However,
oh
yeah,
oh
yeah,
so
before
we
go
there
to
answer
your
question,
which
things
matter.
A
So
let's
answer
that
question
and
that's
going
to
be:
let's
do
that
slide
duplicate
slide,
so,
let's,
let's
examine
which
things
matter
for
because
there
are
some
alignment
alignment
procedures
such
as
the
bed
leveling
that
happen.
So
what
is
therefore,
what
is
critical
and
not
so
let's
do.
A
A
A
If
there's
a
different
bed
leveling
procedure,
it's
possible
that
your
bed
is
like
say
not
parallel
to
the
floor
and
you're
printing
it
that
way,
and
it
actually
turns
out
to
be
a
perfect
print,
you're
printing.
That
way
that
could
be
possible,
but
that's
not
how
the
code
works
right
now.
That
would
be
an
upgrade
to
marlin.
A
A
A
D
A
A
A
The
bed
leveling
will
correct
for
that,
but
meaning
it
will
follow
the
surface
of
the
bed
yeah
and
print
on
the
surface.
It
will
still
do
that
ken
any
controversy
on
that.
A
A
Maybe
check
this
with
more
there's
I
mean
there's
documentation
on
marlin
is
decent,
but
it's
not
great.
So
look
at
the
forums
see
if
we
see
anything
else,
because
maybe
by
now
they
actually
updated
other
bed
leveling
mechanisms-
I
mean
that's
an
open
source
project,
so
cons
like
right
now.
Somebody
could
be
working
on
a
different
bed,
leveling
mechanism
and
our
different
algorithms
for
doing
that.
A
A
A
This
is
like
this
kind
of
thing
is
a
visible
slant.
Is
perfectly
fine
like
that?
What's
the
limit
of
it
I
haven't
tested
the
limit
of
it
would
be
where
it's
so
badly
slanted
that
actually
it
starts
hitting
things
when
you
print,
but
because
the
nozzle,
so
the
nozzle
is
going
to
be
like.
A
As
long
as
the
tip
of
the
nozzle
is
like
below
any
slant,
and
what
is
that
distance
that
distance
in
the
current
setup,
it's
only
actually
a
few
millimeters,
but.
A
So
actually,
let's,
let's
look
at
so
this
is
okay.
This
is
actually
okay.
K
A
A
So
if
the
limits
are
geometrical
like
when
will
it
actually
parts
of
the
this
axis
actually
hit
your
prints?
That's
that's
the
limit.
K
K
A
You'll
see
that
you'll,
look
at
your
z,
you'll
visibly
see
it
moving
up
and
down
very
gently
to
follow
the
contour,
the
whatever
you
have,
and
if
you
have
like
little
you
know
say
you
got
a
concave
surface.
I
think
it
follows
that
too
it
will
follow
a
concave,
so
it
doesn't
doesn't
necessarily
have
to
be
just
slanted.
It
could
also
be
concave.
Convex
and
it'll
still
work.
A
Okay,
what's
the
worst?
How
close
do
you
have
to
have.
A
A
You
turn
the
knob
a
few
millimeters
up
or
down
until
the
print
is
exactly
at
the
surface.
You
record
that
number
and
you
store
it
in
memory.
So
the
question
question
I'm
asking
is:
how
accurate
does
that
number
have
to
be
for
you
to
get
like?
What's
the
tolerance
of
this
printer
for
inaccuracy
on
the
initial
z,
height
correction,
the
z,
height
leveling,
thing,
I'd.
A
Yeah,
but
you
can
be
way
more
off
than
that.
You
got
a
1.2
nozzle.
You
could
be
as
bad
off
as
like
at
1.2
as
long
as
some
of
the
bottom
is
still
hitting
the
surface,
so
in
other
words,
with
large
nozzles,
you
can
have
quite
a
bit
of
tons
like
1.8,
even
like
one
millimeter
you'll
still
get
good
adhesion.
If
it's
too
close
you'll
have
stuff
kind
of
like
squeezing
out
the
layers
will
be
a
little
fatter.
A
If
you
have,
if
you're
at
like
one
above
the
bed
and
you're
printing,
that
means
only
the
bottom
portion
of
your
filament
is
going
to
adhere
to
the
bed.
But
that
could
be
enough
because
the
the
stickiness
on
the
pei,
the
bed,
that
we
use,
it's
really
good,
so
it'll
still
probably
stick,
and
you
have
a
good
print.
So
we
actually
by
using
the
large
nozzles.
We
actually
have
quite
a
good
tolerance
for
the
off
that
offset
not
being
correct
upon
the
first
time.
So
I
would
say
that
limit
is.
A
I've
seen
it
where,
as
long
as
it's
like,
it
seems
like
if
even
if
it's
a
little
bit
above
that
1.2,
it
still
has
like
enough
of
that
push
that
it
actually
smears
it
into
the
surface.
So
I
I
know
more
than
that,
but
it's
it
could
be
like
around
1.2
or
a
little
a
tiny
bit
higher.
I
wouldn't
at
2.4.
No,
probably
you
might
not
stick
but
1.2
yeah
you're
likely
to
stick
very
likely
to
stick
and.
A
You've
optimized
your
your
printer
with
more
advanced
stepper
drivers
that
have
smoother
steps
which
don't
vibrate
as
much
and
things
are
more
steady
or
if
you're
going
super
slow
speed,
you
can
still
probably
get
decent
prints
under
like
extreme
conditions
and
stuff
like
that.
Anyway,
that's
just
to
get
an
idea.
You're
gonna
be
asking
like
how
okay,
how
accurate
do
I
have
to
make
that
offset
one
millimeter
under
is
good.
A
A
A
In
a
way
and
the
thing
that
you
have
to
pay
attention
to
like,
if
it's
hardly
stuck,
if
it
was
there
just
by
gravity
and
you
had
no
force
on
it,
that
would
still
work,
but
that's
not
how
it
works
in
reality.
If
you
have
like
a
little
blurb
like
a
little
filament
of
plastic
that
your
nozzle
actually
hits,
that's
where
you
have
to
have
the
good
adhesion.
A
Sometimes
I
notice
that
when
you
print
like
your
head
is
actually
hitting
the
print
because
it
didn't
get
the
off
the
initial
bed
leveling
correctly,
but
it
doesn't
necessarily
fail
because
if
the
prints
still
stick,
it
will
kind
of
like
ride
up
on
them
and
kind
of
bump
a
little
bit,
but
it
will
actually
succeed
if
you,
if
you
got
really
good
adhesion,
because
it
will
just
kind
of
like
that,
will
all
wash
out
the
higher
up
in
the
print
you
go.
So
that's
why
the
sticking
to
the
bed
is
important.
A
When
you
have
failures,
I.e
the
the
nozzle
hitting
your
print
because
there'll
be
little
inaccuracies,
little
burbles,
you
might
have
a
little
thing
of
filament
like
when
you
jump
like
when
you
do
retraction
filament
retraction,
like
a
little
gob,
spat
out
that
shouldn't
be
there
like
it'll
freeze,
it
will
get
solid
and
plastic
is
4000
psi
or
so
it's
one
tenth
of
steel
or
one
tenth
120th
of
steel.
Let's
just
say
one
tenth,
but
it
is
solid.
A
A
A
A
You
do,
is
you
have
to
take
your
your
bed
axis,
put
in
one
screw
and
rotate
to
the
right
angle.
Take
a
take
an
angle
to
determine
the
best
you
can,
but
that
would
be
a
weak
point
here.
We
don't
have
a
good
way
to
set
that
that's.
A
A
As
long
as
your
thing
is
moving
yeah,
it
doesn't
matter.
Actually,
let's
see,
let's
see
that's
this
way,
it
seems
like
it
has.
It
doesn't
no
not
at
all
played
because
it
doesn't
because
it's
the
same
plane.
A
A
Film
and
it
wouldn't
have
to
even
be
adjusting
if
you
have
this
perfect
at
this
little
angle,
and
this
one
is
perfectly
parallel:
it
won't
even
be
z,
adjusting
it's
just
fine
that
way
as
long
as
there's
no
physical
interference
between
extruder
and
your
prince.
So
it
can
be
like
this
because,
maybe
like
then
the
extruder,
actually
the
other
part
of
the
extruder,
actually
hits
your
bed
yeah
right.
But
no,
this
is
you,
don't
have
to
have
anything
here.
Okay,.
A
Good
as
to
be
as
level
because
that
complication
will
propagate
like
if
something
else
is
off,
you
might
get
confused
and
maybe
not
able
to
track
down
the
other
thing
like
that
might
be
an
effect
that
that
has
on
something
else.
I
can't
really
think
of
a
good
example
of
where
it
would
confuse
you
outside
of
your
perfectionist
sensitivity.
A
And
let
me
know
otherwise,
if
that's
the
case,
but
but
no
troubleshoot
yeah
the
number
one
and
two
so
skew
needs
to
be
corrected.
Z
vertical
is
not
necessary.
A
Vertical
meanings,
perpendicular
to
the
ground,
plane,
z
vertical,
is
not
necessary.
L
I
A
A
Will
be
skewed,
but
it
will
be
well,
I
mean
we'll
tilt,
but
the
z
correction
will
will
fix
that.
Okay
right,
I
think
so,
we'll
you
know
you'll
see
it
in
action
today.
A
Let's
see
some
now
cruiser
has
autos.
So
how
do
you
do
auto
ski
auto,
skew
correction,
so
bruce
brags
about
auto,
skew
correction?
They
actually
developed
it,
so
that
is
actually
open
source
software,
but
what
they
will
do
is
they
have
four
markers
on
the
bed
or
what,
however,
many
markers
they
probe
them?
They
find
their
location
by
like
whatever
inductive
sensor
or
whatever,
and
then
they
see.
Oh
this.
This
is
actually
a
parallelogram
and
they
will
correct
that
in
the
software,
so
you
can
actually
have
skewed
axes.
I
will
correct
that
out.
A
We
can
apply
that
to
this
printer
if
we
have
a
basically
sends
four
points,
so
you
have
to
fix
like
four
sensing
points
that
are:
maybe
how
do
they
do
that?
I
think
they
have.
A
The
bet
is
not
metal
like
it
would
be
because
the
way
a
an
inductive
probe
works,
it
works
on
metals.
Well,
so
you
could
like
probe
four
points
that
are
the
same
pieces
of
metals,
which
you
know
their
location
and
you
can
calculate.
A
D
I
A
A
That
yeah
or
you
could
3d
print
a
base,
that's
accurate
and
put
it
into
points
that
you
are
they're
already
pre-measured.
That
would
be
an
upgrade
to
this
system.
You
would
print
maybe
a
little
template
or
even
a
printout,
even
a
printout,
on
a
sheet
of
paper
that
shows
you
exactly
you
put
laid
over
the
bed
you
screw
into
exactly
where
it
tells
you
to
put
it
that's
easy
to
implement.
L
B
A
A
So
if
we
say
bed
axis
and
that's
the
z-axis.
A
A
A
A
A
A
D
D
A
A
G
A
A
You'll
see
d10
is
that
where
this
actually
goes
so
my
next
question
is
does
polarity
matter,
because
that's
the
only
thing
you
got
to
watch
out
for
polarity
doesn't
matter
in
cases.
In
other
cases
it
doesn't
in
cases
where
it
doesn't.
You
have
to
worry
about
it
in
case
it
does
you
better
get
it
right
or
or
it
won't
work,
but
so
an
important
question
to
ask
is:
does
polarity
matter
in
this
situation.
It's
a
resistor
does
polarity
matter.
No.
G
A
No,
so
you
can
plug
in
what
the
red
wires
from
the
extruder
heater
one
or
the
other
into
d10,
d10,
plus
or
minus,
doesn't
matter
any
questions
keep
going
extruder
blower.
So
now
the
blower,
that's
d9
polarity
does
matter
those
those
leads
are
black
and
red,
and
it's
minus
and
plus,
as
in
the
order
shown.
Maybe
we
can
zoom
up
a
little
bit.
A
E
A
E
Blower,
if
you
maybe
you
asked
what
happens.
A
K
I
was
gonna
say
that,
from
my
experience
when
I
was
doing
it
wasn't
clear,
which
was
which
so
I
had
actually
had
them
swapped
so
because
the
the
heatsink
fan
is
supposed
to
run
all
the
time.
D
K
E
A
K
A
A
Heat
bed,
so
what
are
we
doing?
There
we're
doing
a
solid
state
relay
now.
There's
two
things
here:
one
polarity
matters
and
two
which
side
of
the
solid
state
relay
are
you
connecting
to
there's
input
and
output?
So
you
have
to
get
that
right.
There's
it's
labeled
in
and
out
the
inside
has
to
be
dc
3
to
32
volts.
It's
a
small
voltage
dc.
A
A
Take
a
look
at
this
video,
but
there's
a
little
or
look
at
my
printer,
so
I
finished
mine
last
night
by
11
30..
It's
all
wired
up,
it's
ready
to
to
run
ready
to
be
connected
and
tested.
Here's
those
headers
and
there's
a
pin,
a
double
white
pin-
and
you
can
probably
see
it
in
the
video.
L
D
A
The
top
yes,
yes,
that's
five
volts
fed
through
the
usb
we're
so
we're
hacking.
This
we're
feeding
five
volts
to
trigger
the
heat
bed,
because
we
can
otherwise
you'd
have
to
put
12
volts
in
there.
Why
don't?
We
do
12
volts?
A
A
I
don't
know
if
that
makes
sense,
but
that's
five
volts.
That's
five
volts
we're
feeding
the
solid
state
relay
with
five
volts.
If
you
want
to
know
it's
a
signal,
oh
yeah,
before
it
was
power,
so
this
is
just
for
your
understanding
before
this
was
actually
producing
power
like
a
few
amps
like
10
amps,
to
do
heating
of
the
heat
bed
that
would
always
burn
out
these
terminals
and
these
terminals.
A
A
Okay,
five
volts,
so
a
little
double
plug
those
are
actually
readily
there.
They're
the
thing
from
the
fan:
we're
saving
those
the
fan,
fan
cutoffs
for
this
purpose,
putting
putting
what
on
it,
putting
the
ferrules
the
terminals
on
them
using
that
crimper
and
feeding
that
into
those
two
there.
So
all
these
terminals
have
barrels,
that's
a
way
to
get
a
nice
clean
connection
into
there
that
doesn't
fray
and
break
over
time.
Ferals
are
those
little
crimps
with
cylindrical
crimps
exterior
phantom
power
supply.
Okay.
So
now
we're
talking
about
the
fan,
the
the
extruder
fan.
A
A
A
A
A
If
it's
crossed,
we
gotta
take
a
little
pin
and
remove
the
connect
like
remove
the
pins
and
switch
them
on
a
black
side.
Black
side
is
the
white
you
can
disconnect
both
the
white
and
black,
but
the
black
one
is
easier
to
disconnect
so
make
the
switching
on
the
black.
I
put
a
cross
there
because
we're
typically
we're
not
using
that.
We
only
need
four
channels,
not
five
you'd.
C
D
A
And
for
some
people
we
used
so
both
of
both
of
those
both
this
one
and
that
one
drive
the
y,
so
you
could
connect
to
either.
Actually
so
you
could,
you
can
actually
choose
in
a
case
where
some
of
the
drivers
were
hard
to
insert
at
the
bottom.
You
just
put
it
at
the
top
because
it's
actually
a
little
easier
to
insert
it
at
the
top,
because
it's
not
sandwiched
between
two
other
ones
and
it's
kind
of
tight.
There.
A
E
D
A
A
A
The
y
is
for
the
bed,
so
why
this
is
the
bed
axis
plugs
into
either
one
of
those?
Actually
here
it
has
y
motor
one
and
two
because
there's
two
there
here
you
choose
one
or
the
other
just
plug
into
whichever
way,
wherever
you
put
in
the
the
driver,
a
little
chip,
little
board
watch
the
video
as
if
you
got
questions
z
motor.
Yes,
we
have
a
z
motor,
the
z
axis.
A
I
A
E
A
That
one
goes
to
right
here,
so
the
x
is
the
end
step
you
put
on
your
extruder
axis
black
green.
You
can
only
plug
it
in
one
way
on
the
end
step
itself,
but
there
you
can
plug
it
one
way
or
the
other
so
make
sure
the
green
is
facing
to
the
top.
Otherwise,
the
system
you
won't
get
proper
triggering
of
end
stops.
D
D
A
It's
just
what
it
is
like.
You
can
have
end
stops
at
the
minimum
position
like
zero
or
the
max.
So
those
there's
six
n
stop
plugs
because
you
can
put
them
x,
minimum
or
maximum
y
minimum
maximum
z
minimum
maximum.
What
we
have
is
x,
min
and
y
max
details,
but
it's
if
you're
wondering
why
it's
four
over
and
stuff
like
right
next
to
it.
So.
A
A
Zn
stop,
which
is
the
height
probe.
That's
an
end.
Stop
too
it's
just
an
inductive
sensor,
the
other
one's
triggered
by
a
physical
trigger
this
one
triggers
by
induction.
It
turns
on
when
it
senses,
metal,
meaning
it's
close
to
the
bed.
It
senses
between
four
and
eight
millimeters
triggers
as
far
as
eight
millimeters
away.
A
B
A
So
you
got,
the
bed
has
a
heat
sensor
to
tell
you
what
temperature
it's
at
and
you
control
that
to
make
it
60
for
pla.
Typically
extrude.
You
typically
run
at
like
230
c,
but
you
have
to
have
feedback.
You
got
to
tell
you
got
it.
It's
got
to
tell
you
what
temperature
it's
at,
so
you
can
turn
it
on
and
off.
That's
the
software
does
that
so
you
got
feedback
on
this
next
power
chords
in
bed,
okay,
power,
chords-
and
I
imagine
one
needs.
D
A
That's
correct
so
so
power,
so
we
have
we're
just
using
here.
We
use
these
plugs
well,
let's
see
which
there's
a
bunch
of
videos
there
on
this
connector.
We
have
the
obviously.
A
We
have
these
power
plugs,
which
you
open
up
and
do
the
wires
inside
I
like
to
get
away
from
it.
It's
kind
of
hard
to
fit
four
wires
in
them,
so
just
migrating
to
a
regular
plug
snip
it
off
and
make
two
connections.
One
goes
to
the
bed.
One
goes
to
the
power
supply,
so
basically
cut
off
the
plug
double
it.
One
end
goes
to
heat
bed.
A
Other
end
goes
to
inserts
of
power
supply,
but
the
details
are
so,
do
you
want
to
watch
these
videos
together
or
yeah?
Well,
I'll
show
you
out
there.
I.
A
Okay,
so
this
is
since
this
is
useless.
Let's
just
it's
semi,
it's
easier
yeah!
You
can
refer
to
it
once
you
once
you
see
the
thing,
so
maybe
we
can
concrete.
I
will
do
much
better,
so
there's
lcd
panel
next
usb
power,
yeah,
that's
it.
We
can
go
out
there
and
look
at
all
these
connections.
Yeah.
I
think
let's
do.
E
A
Okay,
so
it's
11
30..
Should
we
go
out
there
and
go
over
the
wiring
in
in
real
life,
and
then
we
can
do
that.
So.
A
H
A
J
A
I
think
we
decided
that
we're
gonna
focus
on
on
getting
a
lot
of
people
finished
on
a
printer
with
the
quality
control,
because
I
think
they
were
important
quality
control
points
that
we
didn't
have
yet
now
we
have
it
so
we'll
just
leave
it
to
do
it
that
today
and
start
on
tractor
tomorrow.
It
looks
like
in
all.
D
H
A
Yeah,
so
we
can
go
to
the
workshop
whoever's
ready.