►
Description
Engineering custom astronomical accessories using technologies popularized by the maker movement.
Julien Lecomte
In this presentation, I will attempt to inspire and encourage all of you to try out and apply some of the technologies popularized by the maker movement, such as 3D modeling and 3D printing, in combination with optics, electronics, and software programming, to create custom astronomical accessories that can enhance your own equipment. At the same time, I will also cover a few important tenets of the maker movement, which are collaboration, information sharing, and open-source. Finally, I will talk about a few of the accessories that I have recently created and how they have contributed to increasing my enjoyment of the hobby.
A
C
Perfect
all
right,
so
thank
you,
everyone
for
having
me
by
the
way
I
used
to
be
a
an
sja
member
a
long
time
ago,
over
20
years
ago,
in
the
early
2000s
and
I
used
to
attend
their
your
general,
I
think
it
was
called
the
general
meeting
every
month
where
they
had
a
usually
a
scientist
presenting.
I
don't
know
if
you
guys
still
do
that,
but
I
really
enjoyed
doing
those.
C
C
So,
in
this
talk,
my
goal
really
is
to
inspire
you
to
maybe
take
the
plunge.
We
with
some
of
these
technologies
or,
if
you've
already
taken
the
plunge.
Maybe
it
will
give
you
new
ideas,
and
you
know
stimulation
is
always
a
good
thing
with
this
kind
of
thing.
So,
let's
get
started
there.
You
go
all
right
a
little
bit
about
me.
First
of
all,
so
my
name
is
julian.
I
grew
up.
C
I
was
born,
and
I
grew
up
in
france
in
the
french
alps,
just
across
the
border,
from
francesco
and
about
40
kilometers
south
of
geneva
and
about
20
years
ago,
a
little
over
20
years
ago.
I
moved
to
california
right
after
I
graduated
from
university
in
france,
and
this
is
where
I've
worked
since
so
I'm
a
software
engineer
by
trade,
which
is
not
surprising
here
in
silicon
valley
and
that
gif
is
an
accurate
depiction
of
how
I
spend
most
of
my
days.
C
I
love
to
make
stuff,
and
these
are
some
of
the
pieces-
the
woodworking
pieces
that
I've
made
over
the
years.
So
two
of
these
actually
were
featured
in
fine
woodworking
magazine.
If
you're
familiar
with
the
with
the
magazine
that
you
can
see
on
the
on
the
stands
so
yeah,
I
love
making
things
out
of
wood.
C
C
I
even
designed
my
own
telescope,
a
16-inch
f4
telescope
with
a
lot
of
guidance
from
albert
high.
Maybe
some
of
you
know
the
name,
fantastic
guy,
who
actually
wrote
a
book
that
was
published
with
wilman
bell.
I
believe
and
really
a
great
engineer
and
somebody
who
was
very
generous
with
this
time
to
help
me.
So
you
can
see
here
a
depiction
of
the
the
telescope
before
I
built
it.
This
was
done.
C
C
So
my
interest
in
astronomy
dates
back
to
about
1990,
so
we're
talking
like
over
30
years
ago.
I
was
11
years
old
and
I
remember
I
was
bored
on
a
summer
afternoon
and
I
went
to
my
dad's
office
and
was
looking
at
the
books
in
his
bookcase
and
one
of
them
caught
my
attention,
and
that
was
a
1950
edition
of
this
book.
C
Astronomy,
populaire
by
kenny
fleming,
and
so
I
basically
took
it
from
my
dad's
bookcase,
with
his
permission,
took
it
to
my
room
and
started
reading
it,
and
I
just
became
really
fascinated
with
with
astronomy
and
I've
been
ever
since
so
through.
The
years
mostly
have
been
a
visual
observer
and
I
was
very
active
back
in
2008
to
2012.
C
During
that
time
I
completed
the
herschel
400
astronomical
league
project
and
I
did
about
70
of
the
herschel
2
kind
of
lost
steam
towards
the
end,
but
I've
always
loved,
doing
visual
observing,
and
so
I
became
interested
in
astrophotography
only
a
couple
of
years
ago.
So
what
happened?
Is
you
probably
all
remember?
C
C
C
Through
the
years
my
equipment
has
changed
a
lot
so
back
in
1990,
I
started
with
a
60
millimeter
refractor,
and
then
I
upgraded
to
a
small
newtonian
telescope
that
was
on
on
a
wobbly
equator
equatorial
mount
1997.
I
was
able
to
save
up
enough
money
to
buy
a
need.
Eight
inch,
lx50,
with
which
I
was
able
to
do
a
little
bit
of
planetary,
imaging
using
a
webcam,
and
I
had
built
a
little
adapter
for
it.
I
got
decent
results,
then
2002
after
I
moved
to
the
us.
A
C
C
2010
I
made
the
that
16-inch
dobson
telescope.
I
showed
you
earlier
and
then
just
recently
a
couple
of
years
ago,
I
purchased
a
130
millimeter,
so
5
inch
triplet
refractor
on
a
an
ioptron
cem70
mount.
So
that's
what
I
currently
use
for
astrophotography
all
right.
So
what
does
it
mean
to
be
a
maker?
So,
first
of
all,
I'm
not
a
huge
fan
of
the
term
maker.
C
I
mean
nothing
wrong
with
it,
but
you
know
maker
tinker.
I
don't
know
I
like
the
term
engineer
and
you
don't
have
to
be
an
engineer
by
trade.
You
can
just
have
an
engineering
mindset
a
curious
mind,
and
so
I
I
think
that
represents
me
and
I'll
also,
I'm
sure,
represents
a
lot
of
you
because,
as
a
metro,
astronomy,
especially
astrophotography,
is
a
technical
hobby
and
you
have
to
have
a
bit
of
an
engineering
mindset
in
order
to
troubleshoot
the
issues
that
we
all
encounter
during
a
night
of
astrophotography.
C
So
why
make
things
in
amateur
astronomy?
Well,
in
my
case
there
were,
there
were
several
factors.
The
first
one
was
the
price
of
some
of
the
equipment
that
I
needed.
For
example,
I
needed
a
an
s
compatible
flat
panel
and
the
the
ones
that
you
could
find
on
the
market
really
were
outrageously
expensive.
C
Oh
okay,
that's
fine
I'll
show
them
at
the
end:
okay,
yeah.
So
anyway,
the
the
price
of
that
unit
on
the
market
was
really
really
high,
and
so
I
decided
to
to
build
my
own.
It
turned
out
to
be
actually
better
than
what
I
could
find
and
cheaper
and
in
some
cases
there
are
things
that
you
just
cannot
find
on
the
market
at
all.
Either
the
availability
is
very
low
or
it's
no
longer
available,
or
it's
not
exactly
what
you
what
you
need.
C
C
C
So,
let's
talk
a
little
bit
about
the
tools
of
the
trade
and
the
first
step,
we'll
talk
about
3d,
modeling
software,
and
I'm
only
going
to
name
a
small
number.
I'm
sure
you
guys
might
use
others.
I
didn't
want
to
talk
about
sketchup
or
other,
but
these,
I
think,
are
the
most
popular
ones.
So
fusion
360
by
autodesk
is
probably
the
most
popular
3d
modeling
software.
C
C
I
wish
they
offered
a
license
at
a
lower
price
for
hobbyists
and
without
any
restrictions.
So,
but
anyway,
fusion
360,
I
think,
is
the
golf
gold
standard
and
it's
a
really
good
software
tinkercad,
I
believe,
is
also
by
autodesk.
That's
a
completely
web-based
application
and
it
works
really
well
for
simple
models
so
something
to
think
keep
in
mind
and
it's
free.
C
I
decided
to
use
freecad
because
to
me
learning
3d
3d
how
to
use
a
3d
modeling
software
is
a
very
time
intensive
activity.
It
takes
a
lot
of
time
to
master
that
kind
of
software,
and
so
I
I
wanted
to
invest
my
time
into
something
that
would
not
disappear
from
underneath
me
and
freecad
is
free
software,
it's
open
source
and
there's
a
strong
community
behind
it,
so
it
will
live
on
and
it's
been
around
for
20
years.
C
It's
not
the
greatest
there's
a
lot
of
bugs
and
there's
things
that
I
don't
like
about
it,
but
but
for
the
most
part
it
actually
works
really
well.
Openscad
is
another
one:
that's
free
and
open
source
and
some
people
really
really
like
it.
I
decided
to
use
freecad,
but
openscad
is
also
another
good
option.
C
3D
printers.
Let
me
talk
briefly
about
my
3d
printer,
but
before
we
get
into
that,
let
me
explain
why
I
got
into
3d
printing,
so
I
actually
got
into
3d
printing
earlier
this
year,
so
I'm
by
no
means
I'm
not
an
expert
at
all.
I
actually
consider
myself
a
beginner,
and
I
remember
about
a
year
ago,
at
castar.
C
So
I
ended
up
investing
into
a
budget
printer.
So
this
is
a
creality
ender
3v2,
which
costs.
I
think
I
paid
250
for
it
on
amazon
and
it
comes
as
a
kit
and
you
just
assemble
it
there's
a
lot
of.
There
are
a
lot
of
youtube
videos
that
show
you
how
to
assemble
it
correctly,
which
is
very
important,
and
then
it
actually
works
out
of
the
box
really
well
I
mean
my
first
print
was
amazing,
and
so
I
printed
one
of
those
calibration
cubes
and
it
just
came
out
perfect.
So
I
thought
wow.
C
C
Is
that
little
probe
here
that
sits
on
the
side
next
to
the
extruder
on
x,
in
the
nozzle,
and
that
is
really
important,
because
by
default,
this
unit
uses
micro
switches
for
all
three
axes
in
x
and
y.
It's
not
a
big
deal,
but
the
the
accuracy
of
the
the
end
stop
on
the
z-axis
has
to
be
really
high,
and
these
micro
switches
just
are
not
accurate
enough.
C
C
Is
that
little
thing
here
and
that
makes
it
really
easy
to
get
the
parts
off
of
the
printer,
especially
large
parts,
and
then
a
free
upgrade
was
to
also
update
the
firmware
on
this
thing
and
the
gyros
firmware
is
a
really
nice
piece
of
software,
so
I
really
recommend
it.
But
overall,
I
think
I
paid
about
350,
which
you
know
it's
not
super
high
and
it's
a
really
really
well.
C
It's
it
works,
amazingly
well
for
the
price,
I
think
a
lot
of
other
small
parts
and
things
are
needed
so
metric
taps
to
so.
Usually
you
try
to
avoid
making
threads
in
plastic
because
they
don't
hold
very
well,
but
in
some
cases
it
can
be
helpful.
So
I
have
a
set
of
these.
That's
really
cheap
and
then
you
know
a
bunch
of
nuts
and
bolts
and
threaded
inserts.
C
C
So
I
actually
bought
an
arduino
starter
kit
again
off
of
amazon,
for
I
think
19.
It
was
really
cheap
and
I
I
recommend
it
to
adults.
I
recommend
it
to
as
a
gift
to
teenagers.
If
you
don't,
if
you
have
a
teenager
and
you
don't
know
what
to
give
them
for
christmas,
an
arduino
starter
kit
is
an
amazing
gift,
especially
if
your
kid
is
technically
inclined.
C
So
so
that's
one
thing,
and
then
you
know
some
equipment
for
soldering,
including
a
soldering
pump,
really
important.
Sometimes
it'll
it'll
save
your
bacon,
a
multimeter
with
an
oscilloscope,
whether
it's
a
separate
unit
or
included
in
the
multimeter,
especially
when
you
have
to
debug
pwm
signal.
That's
really
important
to
to
have
an
oscilloscope
pcbs
I
like
these,
which
also
you
can
buy
pretty
much
everywhere,
and
then
you
just
solder
your
components.
C
C
Usually,
the
circuits
that
I
create
are
relatively
simple
and
I
only
need
to
make
one,
but
if
you
need
to
make
a
bunch
and
they're
a
little
complex,
maybe
create
it
in
software
and
then
have
a
company
make
them
for
you
electrical
wires
and
stripper
who
name
a
man
who
doesn't
like
to
have
a
stripper
in
this
house
right.
So
so.
This
is
a
really
important
tool
and,
of
course,
miscellaneous
components
and
connectors.
And
let's
talk
about
the
software,
so
windows
is
really
important
in
astrophotography.
C
Unfortunately,
but
you
know
it
is
what
it
is
like.
The
asco
platform
is,
the
ascom
classic,
I
should
say,
is
only
for
windows
and
so
all
of
the
drivers,
all
of
the
devices
they're
all
on
windows.
So
it
is
an
unfortunate
reality,
but
you
know
it's
not
that
bad
either
microsoft,
visual
studio
is
actually
surprisingly
good,
I'm
afraid
to
say
I
remember
when
I
when
I
started
my
career
here
in
the
silicon
valley.
C
I
used
to
use
microsoft,
visual
c
plus
plus,
and
it
was
pretty
terrible
at
the
time,
although
it
was
already
the
best,
but
it
was
pretty
terrible
and
just
recently
when
I
started
using
microsoft,
visual
studio
again
on
windows,
11,
so
visual
studio
2022,
I
was
actually
surprised.
It's
really
good,
and
so
they
made
a
lot
of
progress.
C
Okay,
let
me
cover
a
core
principle
of
the
maker
movement,
and
so
I
hope
that
if
you
guys
build
your
own
devices-
and
you
know
think
that
you
came
up
with
something
unique
sharing-
is
a
really
important
aspect
of
what
we
do.
What
is
what
I,
what
what
I
do
so
share
your
work
with
others.
The
first
thing
I
want
to
mention
is
people
often
don't
share
enough.
I
remember
when
I
was
working
on
my
automated
telescope
cover.
C
I
did
some
research
on
youtube
and
other
sources,
and
I
I
was
able
to
find
people
sharing
their
their
devices,
but
almost
no
explanation.
They
weren't
really
sharing
any
of
their
code
or
3d
files
and
people
in
the
common
threads
kept
asking
hey.
Can
you
share
your
your
electronic
circuit
schematics,
and
can
you
share
your
code?
Put
it
on
github
whatever
and
a
lot
of
people
they
either
don't
or
or
maybe
they
created
something
unique
and
they
just
keep
it
for
themselves,
which
is
kind
of
a
shame
so
sharing.
C
If
you
think
it's
just
a
proof
of
concept,
even
if
you
think
that
your
code
is
embarrassing-
and
sometimes
I
think
my
code
is
embarrassing-
or
maybe
sometimes
it
doesn't
work-
maybe
there's
things
that
still
you
know
maybe
there's
a
few
corner
cases.
It
still
has
value
to
share
that
code,
and
some
people
might
find
it
stimulating
might
come
up
with
new
ideas
might
contribute
some
of
their
time
to
make
your
software
better
document.
C
Your
work,
it's
one
thing
to
put
your
code
up
on
github,
but
it's
much
more
important
to
actually
document
it,
ideally
in
the
sim
code
repository,
for
example,
in
a
readme
file,
so
that
it's
easy
for
people
looking
at
the
code
to
go
through
the
instructions
to
know
how
to
compile
how
to
install
the
software,
how
to
use
it,
make
sure
that
you
use
a
permissive
license.
So
I
all
of
my
software
is
published
under
the
mit
license,
which
allows
people
to
use
it
in.
C
C
All
right,
so
yes,
the
mit
license,
is
one
of
these.
Those
licenses
that
allow
you
to
publish
your
code,
it
can
be
used
in
commercial
products
and
that's
perfectly
fine
by
me.
You
may
want
to
avoid
gpl
licenses
because
of
I'm
not
again,
I'm
not
a
legal
expert,
but
I
know
enough
about
gpl
that
a
lot
of
people
want
to
stay
away
from
it
and
finally
avoid
patents.
So
here's
why
I'm
putting
this
in
this
slide.
I
recently
watched
a
video
of
a
guy
who
came
up
with
something
that
was
really
neat.
C
C
My
reaction
was
to
roll
my
eyes
and
said
you
know
what
we're
not
really
in
this
hobby,
for
you
know,
you're
not
going
to
become
a
millionaire
of
a
device
to
take
flats
with
a
celestron
rasa.
Let's
be
real
so
to
me,
there's
a
lot
more
value
in
just
giving
your
work
for
free
and
and
just
let
others
enjoy
it
and
maybe
even
make
it
they
will
help
you
make
it
better.
C
So
you
know,
please
don't
avoid
patents
all
right,
and
then
you
know
it's
one
thing
to
to
do
your
work
and
publish
it,
but
it's
important
to
evangelize
it
as
well.
So
you
know
start
a
youtube
channel
or
a
website
or
blog
post
about
your
work
on
social
media
or
in
amateur
astronomy
forums.
So
I,
like
cloudy
nights
in
france,
there's
a
very
famous
forum
called
astro
stuff,
but
there's
many
others.
C
A
tech
talk
with
your
local
astronomy
club,
like
sjaa,
I
think,
is
a
great
venue
to
share
some
of
your
ideas
and
and
share
some
of
your
experiences
with
like-minded
folks
star
parties
are
a
great
place
to
see
equipment
in
action
and
get
new
ideas.
I
remember
I
mean
one
thing
that
I
love
to
do
when
I
go
to
star
parties
like
castor
or
gssb
is
to
or
even
to
whenever
I
go
out
and
meet
up
with
friends
in
a
dark
side.
C
Also
like
one
one
guy
had
a
mount
that
he
had
made
out
of
wood
for
his
binoculars,
and
I
thought
that
was
a
really
neat
piece
of
engineering,
something
that
you
know.
I
took
some
photos
and
I
wanted
to
eventually
replicate
it.
So
yeah
stop.
Parties
are
a
great
way
to
to
see
equipment
in
action.
Maybe
even
try
it
out
present
at
conferences
or
star
parties,
and
then
one
thing
that's
important,
and
that's
always
something
that
I
do
is
focus
on
the
process,
not
the
product.
C
A
lot
of
people
will
ask
you
for
your
stl
files
and
the
code
that's
ready
to
go,
as
in
you,
you'll
be
providing
them
with
a
product.
No,
no,
my
point
of
view
is
you
give
them
the
tools
so
that
they
can
build
the
device
on
their
own,
maybe
give
it
their
own
twist?
Maybe
simplify
it,
make
it
more
complicated,
but
you
don't
give
them
a
product.
You
give
them
a
process
that
they
can
follow,
understand
and
then
build
the
product
on
their
own.
C
C
So
starting
now,
if
you
have
a
question,
you
can
interrupt
me
and
we'll
take
a
break
and
I'll
answer
your
question
and
then
we
will
continue.
Okay,
so
like,
like
I
mentioned
earlier,
some
of
the
some
of
the
projects
that
I've
done
are
not
hardware
projects.
They
can
be
just
pure
software
projects,
and
so
this
is
something
that
I
did
a
while
back,
which
allows
me
to
so
there's
always
the
question
of
okay.
C
So
I
wanted
to
have
something
that
was
more
formal
and
I
created
a
piece
of
software
using
some
something
called
called
gnu
octave,
which
is,
you
might
be
maybe
familiar
with
matlab,
so
mathematical
simulation
software
gnu
octave
is
a
free
version
of
matlab
free
as
in
free
software,
and
it's
actually
really
good.
It
can
do
probably
I
don't
know
70
of
what
matlab
can
do
and
and
it's
actually
compatible
with
matlab
in
in
many
ways.
So
I
created
this
thing,
and
this
was
the
output
for
different
set
of
input
parameters.
C
So
on
the
left,
you
can
see
graphs
for
a
4
inch,
refractor
f7
with
a
zw,
asi
1600
and
then
on
the
right
in
celestron
925
at
f7.
So
with
the
reducer
with
acw
is
signed
294
and
then
you
have
a
single
star
and
then
double
stars,
a
double
star
that
is
separated
by
2.5
arc.
Second,
and
then
I
also
have
a,
I
also
have
a
seeing
that
is
set
to
what's
average
in
my
in
my
location.
C
So
if
I
measure
the
fwhm
of
stars
in
my
exposure
when
I
take
them
from
my
backyard,
I
usually
get
about
between
two
and
two
and
a
half
arc.
Second,
so
that's
representative
of
my
seeing
conditions,
but
that's
also
one
of
the
input
parameters
in
the
software,
and
so
this
is
actually
what
you
get,
and
you
can
see
that
you
know.
Unsurprisingly,
the
the
larger
telescope
with
the
smaller
pixels,
give
you
higher
resolution
than
the
the
smaller
telescope
with
the
bigger
pixels.
So
again,.
B
C
Surprising,
but
this
software
allows
you
to
play
with
a
few
parameters
and
try
different
things,
and
I
thought
that
was
really
interesting.
One
thing
that
this
taught
me
is
that
seeing
is
really
the
limiting
factor
in
when
it
comes
to
deep
sky
imaging
I
mean,
I'm
sure
all
of
you
guys
know
that,
because
this
is
the
imaging
special
interest
group,
but
that's
something
that
I
didn't
know
again.
I
started
astrophotography
only
two
years
ago
and
I
realized
that
you
know
unless
I
moved
to
an
area
worth
seeing
a
lot
better.
It
probably
doesn't.
C
C
Okay,
so
I
invite
you
to
check
out
this
github
repo
here
the
link
is
at
the
bottom
and
then
there's
a
lot
of
explanations
in
the
in
the
readme
file
and
to
show
you
how
to
use
it
all
right.
So
this
was
my
first
hardware
project
for
my
telescope
and
that's
a
lost
mandy
dovetail
clamp.
So
let
me
explain.
I
have
two
two
pieces
of
equipment
that
ride
on
top
of
my
telescope.
C
C
C
So
I
have
a
lost
mandy
dovetail
on
the
top
of
my
telescope
here
in
black,
and
I
made
these
little
things
here
and,
and
they
have
all
the
holes
necessary
to
attach
the
the
computer
and
the
pocket
power
box
advance
right.
So
you
can
see
what
it
looks
like.
I
even
made
these
knobs
on
my
3d
printer
and
these
knobs
are
actually
made
by.
C
I
went
to
home
depot
and
I
purchased
a
a
long
bolt
with
a
an
x
head
and
and
then
I
basically
made
the
designed
in
in
3d
modeling
software
designed
the
the
head
here
and
pressed
the
the
bolt
against
it
so
yeah.
This
is
kind
of
how
I
made
it
relatively
simple,
but
you
have
like
two
parts
to
it
and
there's
a
little
spring
here.
C
C
So
what
I
did
is
I
waited
until
the
the
the
print
head
was
off
into
the
corner
and
then
I
basically
just
dropped
the
that
square
nuts
into
that
hole
and
then
it
just
prints
on
top
of
it,
so
it
becomes
kind
of
imprisoned
inside
the
the
part
yeah.
So
that's
that
was
kind
of
a
fun
project
to
design
and
to
print,
and
I
was
encouraged
after
I
made
this
gave
me
a
nice
clean
solution
to
to
attach
my
equipment.
C
C
So
I
wanted
to
make
this
ascom
telescope
cover
and
it
took
me
a
little
while
to
figure
out
how
to
make
it.
I
saw
a
few
on
youtube
and
unfortunately,
people
weren't
again
weren't
sharing
anything
which
was
kind
of
a.
It
was
a
little
sad.
So
it
took
me
quite
a
while
to
reverse
engineer
how
to
build
that,
but
in
the
end
it
was
actually
not
that
difficult.
C
D
C
C
So
let
me
show
you,
I
think,
okay,
so
let's
talk
about
the
electronics
very
simple,
so
there's
a
microcontroller
with
whoops,
sorry
microcontroller
with
a
usbc
port.
I
got
the
seed
window
xiao
because
it's
a
very,
very
affordable,
microcontroller,
five
dollars
a
piece.
So
you
know
I
bought
10
of
them.
They
come
from
china.
C
It
takes
a
couple
of
months
to
get
them
and
they're
really
small
they're
the
size
of
my
thumb,
and
I
got
a
servo
which
is
able
to
turn
270
degrees
and
then
because
I
use
a
12
volts
power
source
for
everything
I
needed
to
reduce
that
to
six
volts,
I
believe
for
the
servo,
so
I
used
a
buck
converter.
So
in
this
case
it's
called
the
mini
360,
which
is
a
an
adjustable
buck,
converter
and
again
super
tiny
component,
and
you
can
see
the
the
finished
circuit
on
the
right.
C
C
C
Actually,
it's
not
a
microcontroller
is
not
at
all
like
a
computer,
so
a
macro
controller,
the
the
main
difference
to
simplify
things,
the
main
difference
between
a
microcontroller
and
a
computer
is
that
a
microcontroller
when
you
power
it
up
immediately
it's
available
right,
whereas
a
computer
there's
an
operating
system,
it
takes
a
little
while
to
boot
up.
So
if
you,
if
you
look
at
the
raspberry
pi
for
example,
and
also
a
raspberry
pi,
is
a
micro
computer
right,
it's
also
a
lot
more
expensive.
A
raspberry
pi
is
quite
a
bit
more
expensive.
C
I
don't
know
how
much
they
cost
nowadays,
but
a
micro
controller
is
it's
just
basically
a
small
chip
that
you
can
program
and
we
use
the
arduino
software
platform.
You
know
when
you
hear
the
word
arduino.
It
doesn't
refer
to
a
hardware
platform.
It
doesn't
refer
to
a
microcomputer
microcontroller.
It's
actually
a
software
platform
that
allows
you
to
that
basically
gives
us
gives
you
sane
apis
to
program
your
microcontroller
now.
C
Yeah,
I
wouldn't
call
it
an
operating
system,
although
it's
possible
to
run
an
operating
system
on
or
even
like
a
super
simple
operating
system
on
a
microcontroller.
But
basically
what
I
do
is
ahead
of
time.
I
flash
the
firmware,
so
I
develop
the
firmware
in
in
using
the
arduino
ide
and
that's
done
in
c,
plus
plus,
and
then
you
compile
it.
You
connect
it
to
the
you
to
the
usbc
connector
here
to
your
computer
and
you
upload
the
firmware
to
the
microcontroller.
C
A
C
C
C
C
A
C
C
Is
it
it?
You
know
if
you
use
a
an
operating
system
like
a
modern
operating
system,
when
you
try
to
write
to
a
part
of
the
memory
that
you
don't
have
access
to?
There
is
a
protection
that
is
set
up
at
the
cpu
level
by
the
os
remember
here.
There
is
no
such
api
at
the
at
the
processing
unit
level
and
there's
no
opening
system,
so
you
can
literally
do
whatever
you
want
and
what
happens
if
you
do
something
crazy,
I
don't
know.
Actually
that's
a
good
question.
B
D
A
You
could
conceivably
jump
to
a
location
where
you're
executing
trash
and
it
will
you
know
not
like
if
you
corrupted
a
stack
and
then
it
would,
you
would
think
it
would
go
away
forever.
So
another
approach
to
this
would
be.
You
know
like
some
kind
of
execute
to
completion
and
you
know
restart
every
you
know
millisecond
or
10
milliseconds,
but
whatever
I.
E
B
E
D
E
C
That's
that's
a
good
question
as
well
great
see.
This
is
why
I
like
these
questions
so
to
integrate
with
imaging
software.
You
don't
really
have
the
choice
you
have
to
use
ascom,
so
I
created
an
ascom
driver
for
it,
and
the
s-com
driver
is
very
simple
in
this
case.
It
is,
I
use
the
iswitch
v2
interface,
so
it
is
seen
by
the
imaging
software.
In
my
case,
I
use
nina,
but
you
could
technically
use
anything
you
want.
C
It
is
seen
as
a
switch
and
so
as
part
of
my
sequence,
one
of
the
very
first
instruction
is
to
turn
the
switch
on
which
basically
opens
the
cover,
okay
and
so
in
this
driver.
What
I
do
is
all
this
driver
does
really
is
connect
to
the
connect
to
the
the
device
to
the
microcontroller
and
then
send
it
commands.
So,
for
example,
if
we
look
at
let's
see
if
we
look
at
that
switch
here,
if
you
pass
through,
it
will
call
open
cover
and
open
cover.
E
D
One
of
the
things
that
you
can
do
with
the
arduino
is
you
know
if,
if
you
can,
if
somebody
has
reverse
engineered
the
serial
protocol
that
something
uses
right,
then
you
can
reuse
their
driver
right
and
you
just
have
to
have
the
arduino
respond
to
the
to
the
serial
command.
So,
like
you
know,
we've
done
focusers
and
rotators
and
and
other
things
the
flat
panel
right,
using
pre-existing
drivers
for
a
commercial
product
or
other.
C
C
C
C
C
I
wanted
something
that
was
not
too
difficult
to
build
relatively
affordable,
that
had
good
uniformity
and
that,
of
course,
I
could
control
using
the
ascom
protocol
and
also
I
wanted
something
that
was
not
too
big.
That
was
just
the
right
size
for
my
telescope
right
I
mean
because
some
people
have
come
up
with
solutions
that
involve
okay.
You
buy
one
of
those
led
tracing
panel
off
of
amazon
and
you
open
it
up.
C
C
C
C
And
I
put
a
lot
of
comments
in
my
code,
but
the
math
is
actually
quite
complicated
to
determine
the
the
level
I
mean,
there's
a
lot
of
things
to
take
into
account.
So
so
I
came
up
with
the
with
this
thing
and
there
you
go
and
then
the
the
results
is
rather
surprising.
The
result
is
that
it's
actually
not
uniform
at
all.
C
According
to
my
simulator,
and
I
actually
built
a
prototype
version
of
that
and
I
was
able
to
confirm
not
not
the
exact
number,
but
I
was
able
to
confirm
that
it's
actually
not
uniform
at
all.
So
at
that
point
I
started
doing
some
research.
Okay.
So
how
do?
How
do
you
make
a
perfectly
uniform
panel?
C
Okay
and
the
idea
behind
the
light
getting
played
is
that
the
the
light
that
comes
from
those
led
strips
reflect
within
tiny
reflect
against
tiny
cavities
that
are
inside
that
plastic
sheet
and
then
the
plastic
sheets
kind
of
glows
uniformly.
Okay,
so
that's
kind
of
the
idea
there's
many
ways
it
can
be
done.
I
don't
actually
know
how
tv
manufacturers
do
it.
I
happen
to
work
for
samsung,
but
I've
never
dared
to
ask
but,
like
I
know
some
some
people,
some
companies,
they
they
create
acrylic
sheets-
that
contain
tiny
particles.
C
C
So
let
me
show
you
what
I
did
here
on
the
left
side.
You
can
see
the
led
strip
all
right,
and
this
thing
here
is
actually
the
light
getting
plates
underneath,
of
course,
there's
a
white
sheet
of
paper.
Basically,
so
where
did
I
get?
The
light
getting
plates
well,
I
went
to
my
local
store
and
I
bought
a
sitting
light:
okay,
a
12-inch,
seating
light,
which
I
think
I
purchased
for
like
18
and
took
it
apart
and
inside
of
it.
There
were
quite
a
few
things
that
I
could
scavenge.
There
was
an
led
strip.
C
It
was
kind
of
a
matte
white
and
then
there
was
that
light
getting
plate,
which
is
a
piece
of
plastic
that
is
just
maybe
like
three
or
four
millimeters
thick
and
if
you
let
it
along
the
side
along
the
perimeter,
you'll
see
that
it
starts
glowing
uniformly.
It's
a
beautiful
thing.
So
the
problem
is.
It
was
much
too
big
for
me,
so
I
what
I
did
is
I
created
a
template
and
then
I
routed
the
the
part
I
have
a
router
table
so
that
it
would
fit
exactly
my
telescope,
and
so
this
is.
C
This
was
the
result,
and
on
top
of
that,
I
added
a
an
acrylic
diffuser.
So
again
an
acrylic
sheet,
that's
semi-transparent
or
semi-opaque
and
again
on
the
router
table.
I
cut
it
in
a
circle
and
just
fit
it
on
top
and
then
there's
a
another
3d
printed
components
that
goes
on
top
of
it
with
screws
and
it
presses
everything
together
and
by
the
way
there
is
literally
no
space
between
the
led
lights
and
each
photo
site
and
the
the
side
of
the
light
getting
plates.
So
that
was
also
important.
C
C
B
B
C
C
C
There
is
a
mini
360
buck
converter,
because
the
led
strip
that
I
used
technically,
it's
a
12
volt,
it's
ready
for
12
volts,
but
if
you
you
can't
power
it
at
12
volt.
If
you
do,
you
get
a
lot
of
flickering
and
when
I
first
saw
that
I
was
like
oh,
I
must
have
a
bad
connection.
No,
it
turns
out
that
I
was
actually
at
the
limit
of
what
the
the
strip
was
able
to.
So
I
had
to
lower
it
to
about
nine
volts.
Also,
a
12
volt
was
way
too
bright.
C
There
is
a
power
mosfet,
which
is
used
to
turn
on
and
off
the
the
led
strip,
and
this
is
controlled
by
a
pwm
signal
that
it
comes
from
the
one
of
the
pins
of
the
microcontroller
okay,
so
here
is
3.3,
volts
and,
and
it
controls
a
system
that
takes
nine
volts
and
a
much
higher
current.
This
is
one
amp
current
to
power
the
strip,
whereas
a
pin
from
the
microcontroller
can
source
only
about
20
milliamps.
C
C
C
Yes
and
the
way
you
control
the
intensity
is
by
changing
the
duty
cycle
of
the
pwm
signal
right.
So
pwm
signal
pwm
stands
for
pulse
wave
modulation,
so
you
can
imagine,
as
it
looks
like
a
square
wave.
Okay
and
the
duty
cycle
is
the
ratio
between
when
the
the
signal
is
high
versus
the
signal
is
low,
okay,
so
duty
cycle
of
100.
Basically,
the
signal
is
always
high
zero
percent.
The
signal
is
always
low
at
50
percent.
C
E
C
Powering
an
led
strip
or
flat
panel
using
a
pwm
signal
can
cause
some
issues.
So
one
of
the
problems
that
I
have
with
the
very
first
version
of
this
project
is
that
I
was
using
the
default
frequency
default.
Pv
frequency
of
this
microcontroller,
which
I
think
was
700
hertz,
and
so
I
remember
taking
my
first
flat
using
my
luminance
filter
and
it
was
a
very
short
exposure
like
0.1
second,
and
I
saw
horrible
artifacts
and
I
was
like
no
man.
C
C
Goes
back
to
one
of
the
one
of
the
core
tenets
of
the
macro
movement
is
to
tinker
and
iterate
right.
You
make
a
first
version.
You
test
it
most
of
the
times
you're,
going
to
see
some
problems,
you're
going
to
go
back
sometimes
to
the
drawing
board
or
you're,
going
to
make
some
tweaks
experiment
some
more
so
that
iterative
process.
I
personally
find
it
very
interesting
and
that's
one
of
the
things
that
I
like
about
about
doing.
This
is
just
it's
intellectually
challenging.
C
Okay,
so
I
actually
evaluated
the
uniformity
of
the
flat
panel
and
the
way
I
did
that
is
in
the
dark.
In
my
garage
I
put
the
flat
panel
and
I
put
my
a
dslr
camera,
and
I
took
a
photo
of
the
of
the
panel
and
I
made
sure
that
this
was
like
about
thirty
percent
adu,
and
then
I
looked
at
the
resultant
pics
inside
and
you
can
see
that
it's
just
like
one.
The
the
spread
of
the
histogram
is
extremely
low,
extremely
small.
A
A
C
If
you
guys
have
an
observatory
in
a
large
telescope,
I
would
say
the
best
option
is
probably
something
I've
seen
on
cloudy
nights,
where
people
put
their
flat
panel
vertically
along
the
side
of
their
observatory
and
when
their
telescope
is
in
the
zero
position.
It
just
points
straight
towards
the
flat
panel,
and
then
you
just
turn
it
on
and
you
can
take
your
flat.
So
I
would
actually
go
this
way
if
I
had
a
bigger
telescope
and
and
something
in
an
observatory.
C
C
So
yeah
just
to
go
back
to
your
question,
the
the
software
can
actually
dim
the
panel
automatically
and
it's
a
really
powerful
thing,
because
you
can
have
very
short
exposures
for
all
of
your
filters
right.
I
mean
before
with
my
led
tracing
panel,
even
on
the
highest
highest
brightness
on
narrowband
filters,
the
exposure
tended
to
be
a
little
long
and
then
I
had
to
manually
pause
between
filters.
It
was
kind
of
a
pain.
Now
it's
completely
automated.
The
software
does
everything
on
its
own.
It
changes
the
brightness
of
the
panel
between
filters.
C
It
does
everything
I
just
press
a
button
and
I
get
all
my
flats
within
a
minute
or
two.
I
don't
even
have
to
think
about
it
anymore.
It's
amazing
all
right.
Let's
talk
about
the
oag
focuser
that
I
built
recently,
so
this
is
probably
the
project
that
I'm
the
most
proud
of,
and
the
reason
is
because
I
think
I
came
up
with
something
that
is
unique.
That
nobody's
come
up
with
before.
C
C
So
I
was
really
excited,
but
then
I
realized
another
problem,
which
is
that
when
I
switch
filters
and
my
system
switches
filters
on
its
own
in
the
middle
of
the
night,
when,
when
I'm
sleeping,
the
focus
of
the
of
the
guide,
camera
is
changes,
and
sometimes
it
changes
to
to
an
extent
that
phd2
actually
cannot
find
a
guide
star.
As
a
result,
and
it
starts
complaining,
I
can't
find
a
guest
star
so.
C
Because
of
filter
offsets
I'll
cover
the
details,
but
let
me
show
you
what
what
I
first
saw
and
it'll
be
very
clear.
So
here
I'm
using
now
on
the
left
on
the
right
is
phd2
you'll
see
that
pause
here.
So
I
have
the
I'm
connected
to.
So
I
have
the
the
luminance
filter
selected
and
you
see
that
all
of
the
stars
look
nice
and
sharp
in
the
guide
camera
and
I'm
going
to
change,
fill
to
the
red
filter.
C
So
I
would
basically
image
with
only
one
filter
throughout
the
night
right
and
then
you
know
in
the
evening
before
I
go
to
bed,
I
would
focus
the
the
the
guide
camera
and
then
I
would
let
the
telescope
do
its
thing.
But
since
I
switched
to
a
monochrome
camera
now
now
I
have
seven
filters,
and
so
this
is
no
longer
an
option.
C
So
the
problem
really
is
all
right.
Let's
see
all
right,
so
the
problem
is
caused
by
by
filter
offsets
right
so
because
of
filter
offsets,
we
have
to
change
the
focus
of
the
main
imaging
camera
when
we
switch
filters
and
the
problem
is
that
your
off
axis
guider
is
located
before
the
filter
wheel
right.
C
C
But
the
question
is
by
how
much
I
mean
you
basically
have
to
counteract
how
much
you've
moved
the
main
imaging
camera,
if,
hopefully,
that
makes
sense.
So
the
question
is
how
many
steps
right.
So
we
we
know
the
offset
for
each
filter,
because
that
is
something
that
you
have
to
measure
independently.
C
That's
part
of
the
calibration
process,
and
so
basically,
it's
theoretically
theoretically
possible
to
compute
the
right
amount
that
you
have
to
move
the
focus
of
your
get
camera
to
compensate
for
that
filter,
offset
okay
and
so
the
way
I
implemented
it
is
that
I
that
was
actually
an
idea
that
someone
gave
me
on
the
nina
discord
server
to
create
a
virtual
filter,
wheel,
ascum
driver
okay.
So
when
I
connect
what
in
nina
what
I
do
is
I
actually
connect
to
a
virtual
filter
wheel
as
com
driver
and
within
the
that
aspect
driver.
C
C
C
So
at
first
I
experimented
with
a
build
system
and
here
is
a
stepper
motor
and
it
kind
of
worked,
but
it
would
occasionally
slip.
So
I
didn't
really
like
this
solution
too
much.
So
by
the
way,
this
thing
is
just
press
fitted
onto
the
helical
focus
or
there's
no
glue
or
anything.
It's
printed
with
such
accuracy
that
it's
exactly
the
right
size
and
you
press
it
onto
the
you
press,
fit
it
onto
the
outer
perimeter
of
the
the
focuser
so
anyway.
C
So
this
didn't
work
too
well,
so
I
gave
up
on
this
then
I
saw
someone
on
youtube.
Do
3d
printed
gears,
and
so
I
I
tried
it
and
it
worked
amazingly
well.
So
again
this
is
press
fit
onto
the
helical
focuser.
This
is
also
press
fit
onto
the
axle
of
the
of
the
stepper
motor.
There
is
no
glue
involved
and
this
worked
really
really
really
well.
C
C
C
I
actually
did
not
know
how
stepper
motors
worked
when
I
started
this
project,
so
I
had
to
do
a
lot
of
research
and
it's
actually
not
that
difficult,
and
there
are
a
lot
of
resources
on
the
internet
to
explain
how
they
work
and
how
to
control
them.
But
basically
the
idea
is,
you
send
signals
that
are
high
or
low,
and
you
and
you
make
sure
you
make
sure
that
they
follow
a
certain
sequence
on
each
channel
and
that
basically
makes
the
stepping
motor
move
the
stepper
motor
has
to
this.
C
This
specific
unit
is
controlled
by
a
12
volts
power
supply
and
it
can
take
like
0.3
0.4
amps
when
it
moves
so
again,
not
something
that
an
arduino
can
control
directly.
So
there's
a
uln
2003,
transistor
array
that
allows
me
to
control
this
and
the
leds
with
a
resistor.
These
are
just
there
for
debugging.
C
So
this
is
what
it
looks
like
on
a
breadboard,
and
I
insist
when
you
do
a
project
that
involves
electronics.
Breadboards
are
amazing.
They
really
allow
you
to
fine-tune
your
design
test.
Everything
before
you
commit
to
a
to
soldering
components
onto
a
pcb
trust.
Me,
that's
a
really
important
part,
and
with
this
I
was
actually
able
to
you're
able
to
develop
the
firmware
and
then
test
everything
properly
and
really
recommended
step
this.
This
is
the
final
circuit,
so
you
don't
see
the
leds,
because
it's
not
very
useful.
C
This
is
where
the
the
stepper
motor
connects,
and
this
is
the
12
volt
dc
input,
and
you
can
see
on
the
other
side,
all
the
connections
and
then
inside
the
enclosure
on
the
right.
That's
what
it
looks
like
so
what's
really
unique
about
this
device
is
actually
the
software,
the
software
that
does
that
computation
to
automatically
compensate
for
the
filter
offset
at
the
the
level
of
the
guide
camera.
C
So
when
you
open
the
when
you,
after
connecting
to
the
device
to
the
virtual
filter,
wheel
device,
you
can
open
the
settings
dialog-
and
this
is
what
you
see
so
here-
is
the
real
filter
wheel
device
that
it
will
forward.
Requests
to
these
are
the
filter,
offsets
for
each
filter,
and
then
these
are
settings
like
okay.
So
this
is
the
focuser
that
it
needs
to
control,
backlash
compensation,
overshoot
steps
and
then
the
ratio.
C
So
here
I
connect
to-
and
somebody
was
asking
me
earlier:
what
does
it
look
like
when
you
create
your
own
ascom
driver
and
it
actually
looks
like
this,
so
it's
kind
of
small-
I
don't
know
if
you
can
see,
but
basically
it
says:
dark
sky,
geeks,
filter,
wheel
proxy
for
oeg
focuser,
so
it
just
shows
up
here,
but
the
earlier.
I
think
we
were
talking
about
the
the
telescope
cover
which
I
implemented
as
a
switch.
C
C
C
All
right,
so
what
I'm
going
to
do
now
is
change
to
the
from
the
luminance
filter
to
the
rate
filter,
which
is
what
I
did
in
the
previous
video
and
then
you're
going
to
see
what
happens
it's
going
to
get
blurry
and
then
immediately
after
it's
going
to
correct
for
it
and
there
you
go
so
that
was
like
the
moment,
which
is
oh.
I
can
actually
deal
with
the
the
filter
changes
on
my
oeg
in
software
and
it
actually
works
really
well.
C
So
one
thing:
this
is
not
something
that
you
necessarily
if
you
have
a
a
a
mirrored
telescope
and
or
if
you
have
a
super
high
quality
refractor
like
a
takahashi
or
astrophysics,
you
may
not
have
this
problem.
The
reason
I
have
this
problem
is
because
I
have
a
cheap
refractor,
which
is
not
a
really
truly
aprochromatic
and
that's
and
also
I
I
use
filters
from
different
manufacturers.
C
C
C
His
name
is
christian
bill,
a
french
amateur
astronomer
and
I
printed
the
the
device
myself
and
I
assembled
it
and
tuned
it,
but
I
did
not
design
it.
So
that's
really
important
so
sol
x
is
the
solar
explorer
and
star
x
is
an
extension
of
so
x
that
allows
you
to
use
it
for
stellar
spectroscopy.
So
solix
is
a
spectrohidograph
okay.
So
it's
a
high
resolution
spectrograph,
which
allows
you
to
take
images
of
the
sun
in
multiple
wavelengths.
So
how
does
it
work?
C
C
While
you
scan
the
surface
of
the
sun,
okay
and
then
you
take
that
movie,
you
feed
it
to
a
software
called
int
inti
and
it's
able
to-
and
you
select,
is
very
specific
wavelength.
So
on
the
left,
it's
hr
file
on
the
right,
it's
calcium,
I
believe
calcium,
k
or
calcium
h,
and
it's
able
to
create
or
recreate
an
image
of
the
sun
in
that
wavelength.
C
C
C
So
it's
it's
kind
of
a
fascinating
system
which
honestly
I
I
had
no
idea
that
it
even
existed
until
recently,
but
it's
it's
pretty
amazing,
so
full
disclosure.
I
do
not
do
this.
I
went
directly
to
to
the
star
x
module,
so
I
only
use
mine
for
stator
spectroscopy,
but
part
of
the
reason
is
because
I
don't
have
a
short
focal
length
refractor
and
I'm
not
super
interested
in
solar,
solar
imaging,
but
it's
possible
to
use
this
project
for
really
high
quality.
C
C
Okay,
so
here
is
the
famous
h
alpha
ray
h,
alpha
line,
that's
in
the
red,
of
course,
calcium,
h
and
k,
which
are
in
the
bluest
part
of
the
visible
spectrum
manganese,
which
is
in
the
green
part
of
the
spectrum.
So
that's
actually
a
triplet,
the
famous
sodium
doublet
line,
which
are
separated
by
six
angstrom,
that's
in
the
yellow
part
of
the
spectrum,
and
then,
if
you
go
towards
the
infrared,
so
I
think
it's
like
a
715
nanometers.
C
You
start
seeing
what's
called
telerik
lines.
I
believe
so.
These
are
from
a
molecular
oxygen
from
earth's
atmosphere
and
you
can
see
these
bends,
which
are
actually
really
beautiful.
To
see
and
yeah
those
are
mostly
in
the
infrared
okay,
so
that
was
kind
of
my
very
first
experience
with
this
3d
printed
spectrograph.
C
C
It's
it's
quite
involved
and
I
had
to
reprint
several
parts
several
times
like,
for
example,
the
body
here
is
made
of
two
parts
and
it's
pretty
thick
and
one
thing
that
happened
again.
I
printed
this
with
petg.
One
thing
that
happened
is
that
when
the
parts
cooled
they
shrunk
ever
so
slightly
and
they
were
supposed
to
be
25
millimeters
thick
and
they
turned
out
to
be
24.6
or
24.7.
C
You
would
think.
Oh,
it's
kind
of
a
small
difference.
The
problem
is:
when
you
put
it
together,
then
the
hole
that
these
are
supposed
to
go
into
is
no
longer
the
right
size,
and
so
it
didn't
fit.
So
I
had
to
literally
reprint
them
and
print
them
by
scaling
the
y-axis
just
a
little
bit.
So
there's
a
lot
of
tricks
like
this,
which
makes
this
project
kind
of
difficult
to
print.
C
I
know
in
france
about
200
people
built
this
project
and
almost
all
of
them
had
them
printed
by
a
company
that
actually
is
run
by
an
amateur
astronomer
who
has
a
bunch
of
printers
set
up
in
his
in
his
house
and
and
he's
got
everything
like
dialed
in
perfectly
so
so
that's
that's
an
option,
but
I
would
not
have
any
random
company
printed
for
you.
It
just
just
wouldn't
work
so,
but
it
is
a
fun
project.
It
requires
a
lot
of
patience.
C
It's
not
crazy,
complicated
either,
but
it
does
require
some
experience
with
3d
printing
by
the
way
very
quickly.
How
does
it
work
I
should
have?
I
should
have
talked
about
that
here.
This
is
the
under
here
is
the
diffraction
grading?
Okay,
the
camera
goes
here,
the
main
imaging
camera,
the
light
from
the
telescope
comes
from
here,
and
there
is
a
slit
which
is
located
here.
C
There
are
a
couple
of
other
things
here,
there's
a
mirror
which
allows
you
to
have
the
image
of
the
of
the
slit
reflected
towards
a
guide
camera,
but
basically
that's
that's
how
it
works,
and
I
reckon
if
you're
interested
in
this
project.
I
recommend
that
you
check
this
page
here.
That's
very
well
explained
how
the
whole
thing
works.
C
C
C
So
soft,
there's
software
that
exists
that
allows
you
to
get
a
polynomial
dispersion
or
just
pulling
putting
on
your
rule
of
dispersion
of
your
spectrograph
and
then
from
there.
It's
able
to
know
exactly
which
pixel
corresponds
to
which
wavelength
you
also
have
to
generate
fat
fields
so
to
correct
for
a
number
of
defects,
including
dust
on
dust
on
the
sensor,
is
one
but
there's
also
there's
also
like
strange
things
that
happens
and
I'll
show.
C
C
Actually
I
show
it
better
here,
so
there's
a
little
neon
bulb,
and
so
this
is
connected
to
110
volts
little
resistor
here
to
limit
the
current,
and
then
you
have
like
little
plastic
fiber
optics,
which
come
out
of
this
device,
and
then
I
printed
a
ring
which
I
attached
at
the
top
of
my
telescope,
and
you
can
see
you
see
the
little
orange
light
that
comes
out
of
here.
It's
lit
up
by
this
little
tiny
bulb,
okay,
and
so
what
you
do.
C
C
Okay
and
it's
you
inject
it
while
you're
capturing
the
image
of
a
spectrum
of
a
star.
So
here's
what
it
looks
like
this
is
a
60
second
exposure
on
a
famous
bee
star,
called
shell
yak
or
bit
bitter
lyra,
and
you
can
see
this
is
around
eight-
shall
fine!
You
can
see
that
the
hl5
is
in
emission,
which
is
typical
of
b
stars
if
you
stretch
this
image.
This
is
what
you
see.
So
this
is
the
spectrum
of
that
star
right,
and
you
see
these
lines.
These
lines
actually
come
from
that
light.
C
C
It's
able
to
know
where
the
background
sky
is
so
it
can
subtract
the
background
sky.
It's
able
to
find
these
rays
in
some
cases
it
can
identify
them
automatically
or
you
can
tell
them
okay,
this
number
of
pixels
this
wavelength
and
then
from
there
it's
able
to
calculate
the
coefficient
of
that
polynomial
of
the
law
of
dispersion
for
your
spectrograph
and
from
there
you
can
generate
this,
which
is
a
spectral
profile.
C
Okay,
and
by
the
way,
this
is
really
interesting.
So
in
really
with
such
a
high
resolution,
you
can
see
changes
on
some
targets
in
a
matter
of
less
than
an
hour.
I
mean,
if
you
look
at
a
star
with
a
regular
telescope
and
you
take
a
photo
or
you
look
at
it
with
an
eyepiece
you're,
going
to
see
a
dot
right
and
even
with
the
gem
swept
space
telescope.
C
All
you
can
see
is
the
dot
okay,
it's
going
to
have
some
color,
and
maybe
the
brightness
can
change
a
little
bit,
but
you're
not
going
to
get
a
lot
of
details
with
a
spectrograph.
On
the
other
hand,
you're
gonna
see
a
lot
of
changes,
and
then
you
like
nothing,
nothing
stays
static.
Things
can
change
very,
very
quickly
like
in
some
cases,
even
like
half
an
hour.
You
can
see
changes,
so
this
is
actually
a
really
interesting
case.
C
So
the
b
stars
they
have
a
big,
a
christian
disc,
which
emits
in
the
hydrogen
lines
and
helium.
In
this
case,
this
is
a
helium
and
the
profile
of
the
h
alpha
line
here
you
can
actually
measure
the
speed,
the
rotation
speed
of
the
accretion
disk.
You
can
see
changes
in
the
shape
of
the
accretion
disk,
there's
a
lot
of
really
interesting
studies
that
are
being
done
on
b
stars.
So
this
is
a
kind
of
a
fascinating
field.
By
the
way
there
is
something
called
stellar
topography,
which
it's
kind
of
an
expert
topic.
C
I
don't
know
if
I'll
ever
get
to
to
do
it
with
my
equipment,
but
some
people
when
definitely
professionals,
I've
been
able
to
create
maps
of
stellar
spots
on
stars
by
looking
at
their
spectra
spectra.
Basically,
what
you
do
is
you
take
a
mini
spectra
of
that
star,
so
it
works
with
stars
for
which
the
their
axis
of
rotation
is
more
or
less
vertical.
Compared
to
like
our
line
of
sight,
perpendicular
to
our
layer
sites
and
the
stars
have
to
move
really
quickly
and
they
have
to
have
large
stereo
spots.
C
If
you,
if
you
have
a
star
that
meets
those
conditions,
you
will
see
changes
very
quickly.
In
the
in
say,
the
h
alpha
line
and
those
changes
are
caused
by
the
doppler
effect
and
if
you
acquire
enough
data
using
software
that
sort
of
reverses
all
of
those
all
the
spectral
data
professionals
have
been
able
to
generate
maps
of
sterospots
on
other
stars.
So
it's
called
stellar
tomography
and
it's
kind
of
fascinating
that
it's
possible
to
do
that
with
even
amateur
equipment.
C
So
this
is
kind
of
an
interesting
thing.
Again,
you
can
see
the
the
neon
lines
here,
but
this
is
m57.
So
again
you
have
a
slit
which
crosses
the
the
nebula
and
what
you
see
here
is
not
an
image
of
the
nebula.
The
fact
that
there's
a
specific
width
is
representative
of
the
doppler
effect.
You
can
actually
measure
very,
very
precisely
the
the
rate
of
expansion
of
m57
and
probably
amounts
to
like
50
kilometers
per
second
or
something
after
that,
and
you
can
see
so
this
is
hydrogen.
C
These
are
nitrogen
quote-unquote
forbidden
lines.
This
is
helium
here.
So
there's
a
lot
of
things
we
can
learn
from
a
spectrum.
This
is
kind
of
a
more
exotic
object
called
ss433.
Some
of
you
might
have
heard
of
it.
It's
a
micro
quasar
located
in
our
galaxy.
This
is
a
magnitude.
13
object,
doing
a
higher
magnitude.
13
object
in
high
resolution
is
kind
of
more
than
pushing
the
limits
of
my
equipment,
but
it
this
shows
that
it's
possible
and
you
can
see
again
the
h
alpha
line
emission.
C
C
So,
like
I
said
I,
I
can
use
the
spectrograph
in
real
resolution
mode,
so
in
high
resolution
mode,
I,
the
resolution,
is
23
000
in
my
case,
in
low
resolution
mode
is
900,
so
this
is
an
image
of
the
solar
spectrum
and
you
can
see.
I
also
added
the
neon
lines
here
and
I
put
some
of
the
main
lines.
So
calcium,
h
and
k
h,
beta
h,
alpha
h,
alpha
here,
sodium
magnesium,
h,
gamma
is
visible,
there's
some
iron
lines
as
well.
So
this
is
pretty
much
the
entire
visible
spectrum.
C
And
then
I
had
some
fun
with
discharge
lamps,
so
this
is
the
the
kind
of
neon
bulbs
that
I
use
for
calibration.
This
little
bulb
here
is
comes
from
a
ballast
starter
for
fluorescent
lamps
that
actually
contains
a
small
bulb
that
itself
contains
noble
gases
and
those
are
fun
to
analyze
using
a
spectrograph.
C
C
That's
one
of
the
reasons
why
I
did
the
analysis
here
by
the
way
these
are
2d
spectra
which
are
generated
from
calibrated
spectral
profiles
and
colorized
in
software
and
then
a
few
other
things
there's
an
led
panel.
You
can
see
here
there's
so.
First
of
all,
a
couple
of
things
are
interesting,
an
led
even
those
the
leds.
You
see
in
the
street
street
lights,
they
can
look
sort
of
yellow.
C
They
still
emit
a
lot
of
energy
in
the
blue,
okay,
and
this
is
one
of
the
reasons
why
I
cannot
use
an
led
to
create
a
flat
field
with
my
spectrograph,
because
it's
not
uniform
enough,
but
you
can
see
that
there's
some
variation
here.
You
see
those
oscillations,
I
don't
know
if
you
can
see
very
well
on
the
zoom,
but
that's
a
pretty
common
problem
and
flat
fields
are
able
to
correct
for
this,
and
this
is
caused
by
reflection
within
the
camera
sensor.
C
I
believe
between
the
camera
sensor
and
the
protective
window.
That's
in
the
front,
so
there's
a
little
bit
of
interference
that
happens.
This
is
an
income
incandescent
bulb.
So
that's
that's
actually
black
body
spectrum
and
you
can
see
that
it
emits
a
lot
in
the
infrared
very
little
in
the
blue.
So
that's
to
be
expected,
and
this
is
a
cfl
bulb,
which
is
what
I
have
in
my
garage.
C
So
I
just
pointed
the
spectrograph
to
the
ceiling,
and
this
is
what
you
see
and
you
can
see
like
lines
for
mercury.
For
example,
mercury,
vapor,
okay,
one
last
thing
improvements,
so
I
I
have
a
lot
of
ideas
for
future
projects
and
there's
one
thing
I
want
to
improve
and
it's
my
cabling,
so
you
probably
noticed
that
all
of
my
projects
use
two
cables
right:
one,
usb
cable
and
one
power
cable.
C
So
let's
talk
about
cabling.
This
is
what
I
think
my
telescope
cabling
looks
like,
but
the
reality
is
this
is
what
it
actually
looks
like
and
I'm
sure
I'm
sure
some
of
you
can
relate
to
this.
So
it
can
become
quite
a
you
know,
quite
messy
very
quickly,
so
the
solution
is
to
actually
live
in
the
21st
century
and
what
I
want
to
do
in
my
future
projects
is
to
leverage
ble,
which
is
bluetooth,
low
energy,
for
communication
right
so
to
send
commands
from
the
computer
to
the
device.
C
You
can
do
it
over
the
air.
You
don't
need
a
cable
for
that.
In
most
cases
an
on-board
battery
should
be
it
should
be
able
to
power.
The
microcontroller
and
the
components
that
you're
driving
and
you
know
ideally
you'd-
be
able
to
recharge
the
battery
in
place
and
maybe
provide
battery
status
again
api.
C
So
you
can
control
that
remotely
and
have
an
on
off
switch
on
a
device
with
an
led
indicator,
and
so
I
want
to
test
this
out
soon
on
a
new
spectroscopy
calibration
device
that
I
want
to
make.
That's
super
amazing.
That's
going
to
involve
a
seed.
We
know
xiao
microcontroller
with
ble
support
and
battery
and
lithium
battery
charging
support.
C
It
will
involve
a
so.
I
need
to
go
from
3.7
volts
to
110
ac,
so
there's
going
to
be
a
little
circuit.
That
does
that
which
I
already
tested
and
it
works
well,
there's
going
to
be
a
solid
state
relay
to
turn
on
and
off
the
bulb.
There's
a
lot
of
things
involved.
It's
going
to
be
a
lot
of
fun
and
if
that
works
well,
I'm
actually
planning
to
improve
some
of
the
devices
that
I
showed
you
tonight
to
basically
get
rid
of
the
cables.
C
C
C
E
C
That's
where
the
spectrum
goes:
yeah,
yeah,
that's
a
great
question.
So
actually
what
you
saw
here,
I'm
going
to
stop
using
I'm
actually
going
to
use
only
one
there
is
one
small
caveat.
Is
the
positioning
of
that
light.
It
has
to
be
you,
so
you
still
put
it
in
the
front
of
the
telescope
right
and
the
spectrograph
sits
at
the
end,
but
you
have
to
make
sure
that
that
bulb
is
aligned
with
the
slit
of
the
spectrograph,
because
if
you
don't
you're
going
to
get
a
calibration
bias,
so
they're
having
four.
C
C
Oh
so
someone
asked
about
zigbee
great
question,
so
I
actually
have
no
experience
working
with
zigbee,
but
I
don't
know
that's
something
that
I've
heard
about.
I've
never
actually
done
a
lot
of
research,
but
bluetooth
and
bluetooth.
Low
energy
for
low
bandwidth
communication
seems
like
it's
most
devices
nowadays
support
that.
So
it
seems
like
it's
not
going
anywhere.
C
Didn't
this
blow
your
mind
that
you
can
do
like
literally
image
a
black
hole
from
your
backyard
in
san
jose
or
or
sea
changes
in
star
accretion
disks
within
an
hour
of
time
I
mean
like
to
me:
this
is
the
real
power
of
spectroscopy
is.
Is
you
can
do
like
incredible
things
if
you,
if
you
know
how
to
interpret
the
the
spectrum.
A
C
Correct
so
this
little
thing
here
is
actually
the
absorption,
so
there
is
probably
there
may
be
so
this
looks
like
a
p
signee
profile
by
the
way
and
there's
probably
some
absorption
and
it's
probably
not
quite
centered-
on
the
helium
line,
so
the
there
might
be
like
a
sterile
wind
at
the
same
time.
So
I.
A
C
It
is
but
there's
why
would
it
go
in
absorption
here,
like
the
continuum?
Is
here
right?
So
there's
an
absorption
there's,
an
emission.
Why
do
you
have
absorption
is
because
and
and
if
it's
if
the
absorption
line
is
shifted
towards
the
the
blue,
it
means
that
you
have
a
sterile
wind,
so
it's
a
little
complicated
to
to
visualize,
but
that's
it's
called
p
cygni
profile,
it's
very
common
and
yeah.
It's
sometimes
there's
multiple
phenomena.
C
A
C
A
A
C
Consider
subscribing
to
my
youtube
channel.
This
is
kind
of
a
shameless
plug,
but
basically
I'll
be
talking
about.
I
really
like
youtube
as
a
platform
that
allows
me
to
share
some
of
the
the
things
that
I
create
and,
and
also,
of
course,
github
where
you
have
all
of
the
code
and
the
documentation,
but
yeah
subscribe
to
my
channel
and
you'll
see
that
I'll
be
posting
more
and
more
about
spectroscopy.
As
I
progress
along
my
journey.