PETER DE BAAN 06/24/14
“HOW TO MAKE YOUR OWN TELESCOPE MIRROR BY
HAND”
A To
make a telescope mirror from a flat disk (of glass), the flat surface has to be
changed to a spherical / parabolic surface, because a flat surface can only
reflect light in the same opposing angle as it comes to the surface, but for a enlarged
view to be seen, the light has to come to a small focal point, at our eye or
camera lens, therefore the surface is changed by grinding and polishing to the desired shape to produce that “focal-point”. By first rough grinding to an approximate shape as calculated, then by finer grinding to the proper spherical
shape, which will have a Focal point at the Radius of Curvature, (the center of
a ball .) when light is produced at that focal point it will be reflected from
the whole mirror surface back to that same point (if properly polished) If the
R.O.C. Point is more than 15 times the diameter of the mirror, the mirror may
be considered finished and does not need to be parabolized, only Aluminum or
Silver coated for better reflection, because at that distance the Focal point
from near-by and the focal point from “infinity” are nearly the same.
To make the focal
point closer to the mirror, like 12 times (or less) the mirror diameter. than
the focal point of the outer zones (“rings”) of the mirror will not be
coinciding with the mirror center. The remedy is to change the mirror-center
section zones to a shorter reflective distance and tapering-angling the
difference to the outer edge, which will not be “corrected”, until all zones will again have one focal point, this is called Parabolizing, measuring is done
with the zonal measuring on the Foucault testing, and ultimately with a “star
test”, or “auto-collocation” because the differences in surface depth/angle we
can only measure with light-waves.
The measuring on the Foucault tester is done at the Radius of Curvature
of the spherical mirror shape, which is
twice the distance of the parabolic focal point
and will show a “doughnut” shape when viewed face-on, showing that it is
not a sphere any more, the real shape is a smooth parabolic shaped surface. which can be tested and seen
with a “double-reflection” Auto Columation. (which is a indoor star-test)
B Making your telescope mirror-disk by hand.
Start by deciding what size your telescope will be
and what material to use for your mirror.
You can purchase the glass disk or “trepan” it yourselves.
If you have a drill-press you can quite easily “trepan” two disks, 6” (min.) x
3/8” from plain glass with a 6” diamond edged tool. Or you can make the trepan
cutter yourself: cut with a saber-saw or wood cutting router from ¾” plywood a
circular disk of the mirror size, mount around the edge a 2” wide strip of
galvanized iron, cut 4 to 6 diagonal reversed “teeth” in the cutting edge and mount a
center “driver” floor-flange for mounting on the drill, lay the glass sheet in
a water bath-tray, (block all edges from moving) start the slow turning, add grinding
carbon dioxide powder on the cutting line and some cooling water and/or glycerin, keep an eye on the cutting process.
For the material choice, plain window glass
(soda-lime glass) and “Pyrex” are the main choices and both can produce high
quality mirror. The size of the mirror-telescope and focal length is a personal
choice, for reason where and how it is to be used, and how the user is able to
handle and use it.
Advice for the first
time telescope-mirror maker: make your mirror no larger than 6” (152 mm ) diameter
and focal length of 36” (914mm F6,) or 48” (1219 mm F8)
minimal 3/8” thick of
soda lime glass for a Newtonian reflector telescope.
One Way to calculate the grinding-depth of the concave-sagitta-dip
is (measuring with a 4” home-made dial
indicator-sphero-meter) to divide #
1. with the desired focal length of the telescope mirror (1/36 =
0.0278” or .70 mm) or (when sunlight is
available you can dip the ground disk in water and the whet surface can reflect
the sunlight and show the focal length (the sun size reflection focal point is
about ½” diam)
MIRROR MAKING with circular strokes
When ready to make a
concave surface on the mirror you will need to have the mirror-blanc and
grinding tools and grinding grit, use also a equal size(10% larger or smaller
works fine) glass tool-disk, and grind-polish with the mirror on top, making circular strokes with (1/3 min.
of mirror overhanging, during hogging) this produces a concave spherical sagitta
surface. (causes less fatigue, and can be used throughout the process, it simplifies even the
parabola-correction)
Use the circular stroke with M.O.T.
(mirror on top) for all grinding and polishing and even parabolizing, only changing the overhang
and the center-weight on the mirror for the different functions, here is a film
showing circular stroke polishing
When grinding sounds
become muffled it is time to replenish the grit and water, after a few times
replenishing, clear off the waste and do it again until you have reached the
calculated sagitta (#1/planned foc. Length, as 1/35=0.028) using a 4”
spherometer, This home-made 4" diameter Indicator spherometer on 3 ballpoint pen legs works on all size mirrors and focal-lengths) or to measure the focal length divide 1/reading as1/0.020=50). When Sagitta is reached you can start fine grinding with the same circular stroke, with only 20% overhang, finer grinding
grit and no extra weight, polishing is done also with this stroke using a
pitch-lap for the polishing
“Hogging ” 12” or
larger mirrors use a sub-diameter
tool, lay the mirror face up, on a larger ¾” octagonal or round sealed plywood board, (with a pivot center hole), to hold
the mirror in place with 3 screwed-on clips and lay the board on a level work
place on a short pivot and a anti-slip mat. Sprinkle some 80 grit Carborundum (Silicone
Carbide) powder and little water on the mirror surface and start grinding with a
tool (a cast iron pipe-cap, SMALLER THAN THE MIRROR RADIUS) doing strokes all
over the surface onto, but not over the edge. (bevel the sharp mirror edge before the
changing to avoid edge spalding) lenthen work the full size tool with the
same circular stroke with 30% overhang thereby changing the F.L. (this change
need to be done including the mirror
center to avoid making 2 different spheres) and avoiding T.D.E. Test for good sphere with the pencil or Sharpie- ink-test. While grinding
with a full size tool. Use a pencil,
circle around the mirror and a cross over center, a few light weight small
circular strokes with #5 Al Ox should wipe the pencil mark from the mirror,
showing perfect contact and sphere.
1A TOOL MAKING
When you used a sub size tool for hogging. You need
to make a full size fine grinding tool. Lay the mirror face up on a level surface (the tool is cast on the mirror face,
protected by thin plastic sheeting),Grease the top side of the plastic sheet,
and place a “wall” around the plastic covered mirror about 1” higher than the
mirror, place 1” glass or tile pieces on the grease, (to keep the tiles from moving),
mix “Dental-stone” or “Hydro-stone” (from
Tile-supply) cement powder with water to a thick liquid without lumps,
pour careful into the tool-form ,minimum ½ to ¾” thick,without stirring or shaking, lay a precut
backing board (with 3 screws penetrating trough the board for better bonding) in
the soft mix.
Clean all cement mixing equipment and wait until tool is
hardened (allow 15 min,) remove the mould ring, clean the tool edge, rinse the
tool, pour-sprinkle some #120 Silicone-carbide-grit (or recycled grid) and
water on the tool. Start grinding with maximum
20% overhang to keep the sagitta constant (no need to wait for
comlpete drying of the tool )
THE “FINE” GRINDING:
Use a full-size tool, small weight, M.O.T., using # 120 grit (or recycled grit) , 15 to 20%
maximum diameter overhang, creating a sphere with circular
strokes for 15 or more minutes, clear the mirror, check the surface for extra
large pits with a magnifier, if none are found, check with a 4” sphero meter restart
with # 220 grit / 320 going trough the Carbo-grit sizes, then change to Aluminum
Oxide micron-sizes from 25 to 5 micron, apply 5 micron only once and when more
liquid is needed use only water (and possible little tooth-paste) no pressure
on mirror, just moving the mirror to avoid fine scratches and produce a very
fine prepolish, check sagitta depth, and spherical surface with pencil or Sharpy pen test, which
is important before beginning polishing, next look over the horizontal
mirror surface, if it reflects the distance power-pole/flag-pole or other view,
it has a good pre-polish.
Very
carefull clean all working parts and-working
area before and during pitch-lap handling, to avoid time-consuming scratches in
the polished surface
Making the Pitch-lap
Lay the mirror face up on level surface, place the vinyl
mould face up on the mirror, spray little dish-soap water on for mould release,
support mould corners, so that the
mould rest complete on the mirror, melt the pitch slowly to MAX 400 F, (with
A thermostat controlled heater) in old coffee-can, do not let the pitch
boil, and do not add any softener, (a hard pitch-lap does not change shape
and works longer) stir until no lumps are left, pour onto the mould over the
mirror area only, place the pre-cut backer-board onto the hot pitch. (I use
usually Corian type counter-top material for pitch-lap backing),
When pitch is “set” (hardened), place mirror with mould and
pitch in clean-cold water, when cooled remove the pitch-lap careful from mould,
clean edges from excess pitch (use hammer and chisel carefully) test lap
contact, if improvement is needed place the mirror face up, cover with a
“screen” type (fiber-glass), spray little detergent water on for release, warm
the lap face with a heath lamp, when face is slightly soft, place face-down on
the screen-covered mirror, ADD a weight on it (hand pressing is uneven) for few
minutes, start polishing with the fine shaped good matching lap
2 PITCH-LAP POLISHING:
Lay the pitch-lap on the turn-pin, poor some liquid Cerium
Oxide on the lap and start with mirror
on top polishing mirror, after 20 or 30 minutes polishing with the circular stroke using max. 15 - 20%
mirror-diameter overhang and little weight, for the Cerium works with chemical
reaction, not pressure. Clean up the mirror and see the first polish, with sunshine you can measure your mirror’s focal-length (with artificial light you
can measure the Radius of Curvature,
which is dubble the focal length) and see whether the whole mirror has polish, keep polishing for a few hours, than see
with a laser at 45 degree angle, how good the polished surface is coming.
Reflection on the polished surface should be nearly none. If the mirror gets
“sticking” instead of sliding add more Cerium-Oxide If that does not
solve the problem, place the pitch-lap a few minutes under a heat-lamp until surface
is slightly softened and repress with the FG screen.When surface is near non
reflective(0.5%) test the Radius Of Curvature with the Foucalt knife edge for a
“null” blanc-out (indicating a sphere) and with the Ronci screen test to see the mirror-surface quality, (when
Ronci lines are straight from edge to edge you have a good sphere as shown with
the “null”,a Turn Down Edge (lines curving near the edge)causes much longer parabola
labor and can be avoided by polishing longer with 15 - 20% overhang.
ABOUT Turn.Down or Up Edge.
The turn down edge
name is somewhat of a misnomer, it is the name of the uncompleted strip
on the mirror edge, caused by insufficient/incorrect grinding and/or polishing,
complete to the mirror edge.
One T.D.E. is caused by changing M.O.T. to T.O.T., even for short time.
That change causes a reverse in the
grinding as well as polishing.
Do change that way only to make the focal-length
longer, and for that the whole surface needs to change, to the mirror center,
the effect begins at the mirror edge and produces first “two sphere-portions”. In
reality the outside begins changing towards convexity in stead of
concavete. By Foucault testing the area
involved can, even with a nice polish, null all focal-reflection as with convex
surface.
3 Parabola-correction with circular stroke
It is important to have a good smooth
surface spherical mirror with
no T.D.Edge when starting
the parabola correction. Use the same
full size hard micro-faced (made with the Fiberglass screen) repressed
pitch-lap (for we do want to
change the mirror shape, not the lap), done again with M.O.T. CIRCULAR HAND STROKES, overhanging
mirror by 30 to 40% avereage 35%, little extra center weight (the
mirror-center going over the lap edge must wear down mostly) and Cerium Oxide.
After 75 slow circular strokes test with Foucault to see if there is any surface
correction change. (With circular
strokes the first correction to see is a faint donut hole depression at mirror
center, the next to see is with the Rounchi screen slightly centered curved
lines instead of straight.) the adjacent zones will follow all to the rim with
the continual use of the circular stroke/weight /overhang. I mark the back of
the mirror with a sharpy marker,avoiding going over the edge Test often, for you do not want to OVER
CORRECT at all. When all zones are corrected, end the parabola strokes and
improve the surface with some very light-short overhang polishing strokes to
have a
higher
RMC surface avereage
(Note: the 1/3 to 35% overhang
at hogging will produce a sphere.
the 35% avereage overhang
during Parabola correction will change the sphere to a parabola.)
3 A. PARABOLIZING MIRRORS ON THE Mirror.O.Matic-TURN-TABLE
Mirrors larger than 12…..14” to be corrected on Mirror-O-Matic machine with sub sizes Tool On Top, slow 5 rpm
turntable, stroke arm 20….50 p/m start
with a small 1/3 size lap going edge to edge but not over, for 5 to 10 min, test mirror when first of
doughnut shows, continue with same lap, simular strokes not over edge and
little or no weight, for 10 minutes, keep checking every 60 to 100 strokes. On the Foucault with zone screen from Cauder and
Figure XP program numbers
3 B. Making a
mirror on a powered SPIN-TABLE
THE SPIN-TABLE IS THE MACHINE-MODE OF THE
CIRCULAR STROKE
The spin-table machine turning 29
to 30 rpm, is more accurate than the circular hand-stroke. Mirrors to 12” diam
are best done with mirror on top. On full size tools (The mirror is “spinning
freely while dragging on the powered base ) acting like the circular stroke,
hogging with 35% overhang, fine grinding
and polishing with max 15 - 20% overhang, avoiding Fl change. Parabola
correction with 35% overhang, test often
to avoid hole at center (caused by the spinning accuratie
(NOTE: TO AVOID THIS LET THE”PIN” MAKE SMALL SIDE-WAY STROKES possible
with the M.O.M, do some circular stroking by hand
without the Pin
All work for mirror making can be
done with this spinning from hogging to parabolizing.
measure regular for sagitta
depth, when that point is reached, change to finer grit, following with the
same changes as done by hand-fine grinding.
3 C Larger mirrors are made on
the Spin-table with Tool on top. this is
working upside down : REVERSED IN
EFFECT, AND NEEDS A PRESSURE ON THE TOOL EDGE WHERE THE MIRROR-CENTER “HOGGING”
OR PARABOLIZING CORRECTION TAKES PLACE
For working larger mirrors use sub-dia-meter
tools on top, (minimum 70% mirror-size), during hogging, and fine grinding and polishing, when ready
for polishing use a fine pressed hard lap troughout, little or no pin-weight, maximum
20% overhang, continual supplying Cerium-water, regular testing until ready for
Parabola correction Change to minimum 70% pitchlap, 40% “overhang”(= 40%
clear mirror) , little or no pin-weight, BUT
EXTRA WEIGHT ON TOOL EDGE, ( I MOUNTED A ROLLER-SKATE WHEEL ON THE”PIN”SIDE-ARM
ABOVE THE LAP-EDGE CARRYING A FEW LB LOAD), makes smooth spinning (LESS than ½
table rpm) and continual cerium-water dripping, regular testing the zones,
until edge zone nulls, test
often paying attention to “hole” in center of mirror, if that happens, remove
the large lap and pin. Use a 1/3
pitch-lap, center over center, edge to
edge strokes, by hand on the turning mirror until the hole is gone.
TESTING THE PARABOLOID
MIRROR
THE FOUCAULT TESTER
The first mirror test for a good sphere is with a
“null” test and straight Rounce screen lines with knife edge viewing or with
Camera on TV screen.
For Parabola testing use also the Foucault-tester now
with a full mirror-size “Cauder-screen” (calculated for Figure XP
program) on the mirror face with cut-outs to show the parabolic zones of the
mirror, each zone having a different focal-length for correct “nulling”, all
zones are of equal area calculated
MAKING THE CAUDER
SCREEN: Cut a thin cardboard disk the size of the mirror, Calculate the total mirror surface area, (r*r*3.1415)
divide area by the number of zones you planned for, calculate the center-zone
area and radius (zone 1), zone 2 radius will be calculated from twice
the center-area, etc. Draw the different
zone radius on the screen-board and
cut some (25 deg. sect) in the left and
right side of each zone area. using the zone
edge information for Figure XP program .
When all zones are on the Foucalt knive edge corrected,”NULLED”
to mach the programmed zones on the Radius of Curvature. from the Figure XP program shows a good parabola, we can do the Rounce
test which should now show the curved parabola lines. (Note: without Rounce screen a parabola will
show a “doughnut” but any doughnut is not a parabola but shows we do not have a
sphere anymore.)
Next do a Star-test. or (when an optical flat larger than the mirror size is available a test by
AUTO COLLIMATION.
We set up another stand on the test bench for the Auto-collimation test, which is done at the
mirror Focal length which is ½ the R.O.C
and uses double reflection, producing a artificial star test. We use on this
stand a coated “Optical-flat” mirror of same or larger size as the mirror with a
central LED light, which shines at the
mirror and a CAMERA looking at the mirror, the two mirrors are facing each
other and needs to be aligned by adjusting the mirror stand XYZ axes first, so
that the single reflection from the mirror-center points at the camera beyond
the flat, next we align both mirrors so that the (fainter) complete mirror is
projected trough the camera, by minutely changing the focal length we find the
exact focal length by the NULL and the
quality of the parabola. (note the Focal.Length is distance from test mirror to
camera or knife-edge. not from mirror to
opt.flat)
Other test are also available.
For mirrors larger than the Opt.Flat mirror, the Ross Null
test, which uses a lense instead of the optical flat mirror.
Next step is to get the mirror Aluminized or
Silvered.
take
equaly more time for shorter mirrors and larger mirrors.
The
final quality of the mirror is found after Parabolizing, not after spherical
polishing, so end up to do some very lightweight polishing, with less overhang
for a fine surface without changing your parabobic shape.
Only
smoother with a higher RMC surface avereage
