8 inch F3 binoscope (32x203)
Repurposing two Coulter Optical lensless Schmidt mirrors
Double your pleasure by doubling scopes and radiance
Mel Bartels, October 2025

Why

1. Access to lower magnifications than a monoscope. Put another way, the possibility of maximum etendue into both eyes, not just one.
2. More detail visible, both deep-sky and planetary/lunar.
3. Higher contrast images.
4. The milky smooth views. Unlike a monoscope's noisy background.
5. The comfort of two eyes.

For more comments, see my binoscope designer.

Lensless Schmidt cameras

In the lensless Schmidt camera design, coma is eliminated by replacing the Schmidt corrector with an aperture mask at the radius of curvature (twice the focal length), leaving only spherical aberration. Coulter Optical offered kits that included an 8 inch F2 spherical mirror and curved film holder. I built a camera from the kit. THe camera was so fast, collected so much light, that the film holder glowed in the daytime. Later I came into possession of a 2nd 8 inch lensless Schmidt mirror. I decided to turn the two mirrors into a F3 binoscope. The mirrors were wonderfully spherical. Check out the Ronchigram.

Here's a 1959 Sky and Telescope article on the lensless Schmidt camera.

Focal ratio considerations

My goal is the fastest speed possible constrained by the size of the secondaries and tertiaries and further constrained by eyepieces that, side by side, will fit our eye's IPD (InterPuppillary Distance). This maximizes etendue (light throughput which is aperture squared * field of view squared).

Dealing with coma:

1. The TV P2 CC calls for 5 inches more length which makes the secondary too large (unless the P2 is taken apart and a diagonal inserted in the middle)
2. The Baader CC is acceptable at F3.5 as well as the Houdini CC 20mm eyepiece which is advertised to F3.5
3. The new line of Houdini coma corrected eyepieces is reported to work with larger F3 reflectors.

Houdini 20mm coma corrected eyepiece

Critical to the fast binoscope design is a coma corrected eyepiece. My notes on the Houdini 20mm with built-in coma corrector:

• The Houdini is a marvelous eyepiece, working beautifully at F3 and acceptable at F2.7: stars are sharp to the edge.
• My eye fits over the Houdini eyepiece without my nose hitting the eyepiece; so no modification of the eyepiece to make room for my nose will be needed.
• The Houdini eyepiece has the same focus position as the ES25mm 100 deg eyepiece with a MPCC corrector and TV21mm Ethos 100 deg eyepiece with a MPCC corrector.
• I've previously measured how these eyepieces focus. The required focuser barrel's top is 1/8th inch inward of the focal plane.
• Hence the Houdini eyepiece's focal plane is 1/8th inch up from the shoulder bottom. I see the specs say 2mm, so I am within a mm.
• Consequently the Houdini's barrel bottom is 1.5 inches below the focal plane.
• Field stop is 30mm or 1.18 inches.
• Specs are here
• Why the 86 deg field and not a 100 deg field? See here.

Eyepiece study (note that the listed etendue values will be doubled since this is a binoscope)

Field quality at various focal ratios, using my 10.5 inch F2.7:

• At full aperture (F2.7):
  • Both the ES 25mm and the TV Ethos 21mm eyepieces looked great with the TV P2 CC
  • Both looked terrible (unable to focus the entire field at once) with the Badder CC (could not use the Nikon 17mm because the CC would not thread on)
  • The Houdini 20mm coma correcting eyepiece looked acceptable at the extreme edge
• Masking to 8 inches (F3.5):
  • Could hardly tell that the view was stopped down, both in stars and in nebulosity (NAN was the target) (though my eye's pupil was limiting in the full aperture F2.7)
  • Other eyepieces with the Baader CC were acceptable in that I could focus the entire field at once
• Masking to 9 inches (F3.2):
  • The TV Ethos 21mm with the MPCC showed minor defocus at the edge - probably acceptable
  • The ES 25mm with MPCC was a little worse - borderline
  • The Houdini 20mm coma correcting eyepiece looked excellent at the extreme edge

Conclusions:

1. The Baader CC is good enough for F3.5 and perhaps for F3.1; the issue is finding high etendue eyepieces that will fit people's IPDs.
2. The Houdini is a marvelous eyepiece for a binoscope.

The Baader MPCC's performance can be improved by making the primary mirror hyperbolic. See Rob Brown's analysis and his results from making and testing a 5 inch F3.5.

Expected performance (not that a binoscope can be directly compared to a monoscope):

• equivalent monoscope+binoviewer aperture is 11.20 (inches)
• equivalent monoscope aperture for stars ranges from 9.6 to 11.20 (inches)
• equivalent monoscope aperture for nebulosity ranges from 11.20 to 14.4 (inches)

At a 'shoot-out', Robert Asumendi's 8 inch binoscope came close to equalling my 25 inch F2.6 on large extended nebulae like the Veil.

Why 8 inch binoscope? Principally because I have these two 8 inch F2 blanks sitting on the shelf. But also that alignment is easier at lower magnifications and that increasingly large aperture steps need to be taken to gain a magnitude of light.

Contrast with binoculars:

1. 4 3/4 inch binoculars (20x, 3 deg FOV)
2. 5 inch binoculars (30x, 2.4 deg FOV)
3. 6 inch F5 binoculars by Airy Disk. They make their appearance 2:53 into the video.
4. 8 inch refracting binoscope by Altair Astro.

Mirror specs

The two mirrors have different thicknesses: 1 1/4 inches and 1 1/8 inches and weight 4 1/2 lbs and 4 lbs respectively. Surprisingly, the two 8 inch mirrors have slightly different sagitta. One mirror has a focal length exactly 16 inches and the other appears faster.

My goal is to slow the mirrors down to F3.5: FL=28 inches; RoC=56 inches.
8F2 sagitta is 0.25 inches, 8F3.5 sagitta is 0.143 inches, a difference of 0.107 inches (2.7 cubic inches of glass to remove).
Sagittal sensitivity: a 0.004 inches change in sagitta = ~0.1 change in focal ratio.

These will be my 8th and 9th fast mirrors. My previous experience with fast mirrors include:

1. 13.2 inch F3.0
2. 6 inch F2.8
3. 10.5 inch F2.7
4. 25 inch F2.6
5. 30 inch F2.7
6. A 2nd 30 inch F2.7
7. 20 inch F2.9 (laminated blank failed)

Grinding back the curve from F2 to F3.5

1st mirror

Numerous small chipping at edge: must keep a heavy bevel
Using a diamond ring tool of nearly the same diameter with 60-90 grit; accidentally flattened the edge too much
Switched to a 10.5 inch tool with FL=29 inches (F3.6 for the 8 inch) (TOT (Tool On Top) with strokes that took the mirror just over the tool's edge)
Several hours to finish first mirror

2nd mirror

I used the 10.5 inch tool throughout with 60-90 silicon carbide grit

Stopping to bevel frequently both mirrors

Making fine grinding and polishing tools

Densite with hexagonal tiles. I squeezed the tiles down onto the plastic wrap covering the mirror's face by tightly wrapping the cardboard ring.

Fine grinding

Fine grinding - 120 silicon carbide grit

The roughly 10.5 inch dia x 30 inch FL tool left the mirrors a little shallow
MOT (Mirror On Top) with 1/3 W strokes
Alternating mirrors every wet
~6 hrs grinding time; edge tiles are half ground down
Mirrors at F3.5

Fine grinding - 220 silicon carbide grit

TOT with 1/3 W strokes
Alternating mirrors every wet
~4-5 hrs

Fine grinding - 500 silicon carbide grit

TOT with 1/3 W strokes
Alternating mirrors every wet
~3 hrs

Fine grinding - 20 micron aluminum oxide finishing powder

TOT with 1/3 W strokes
Alternating mirrors every wet (10 min)
~5 hrs

Fine grinding - 9 micron aluminum oxide finishing powder

TOT with 1/3 W strokes
Alternating mirrors every wet (10 min)
~3 hrs

Fine grinding - 5 micron aluminum oxide finishing powder

TOT with 1/3 W strokes
Alternating mirrors every wet (10 min)
~2 hrs
The template fits nicely *** but measuring after flash polish shows that the mirrors' sagittae are 0.0014 inches deeper so F3.1 instead of the template's F3.5. The final aluminum oxide finishing powders seems to have shortened the radius of curvature.
You can see the 5 micron shine.

Polishing using full sized pitch lap on large polishing machine

Getting ready to pour the pitch lap on the Densite base (same material I used for the fine grinding tool). I duct tape a cardboard ring so that the ring peels away easily from the pitch by using quick jerking motions.

Pitch is Gugolz #55, shop temp is 60F. I poured the pitch to 5/8 inches thick.
Both mirrors slide smoothly indicating that they have identical curves.
Here the pitch lap is charged with cerium oxide. And showing the mirror on lap.

I mix one part cerium oxide with 10 parts water in a small squeeze bottle. I also have a water spray bottle that I've added a drop of Dawn Dishwashing soap.

My process to begin a polishing session.

1. I warm the pitch lap with the hot air gun for a minute.
2. I sqeeze a bit of the cerium oxide mixture onto the mirror's face.
3. I gently swirl the cerium oxide on the mirror's face with my fingers, often finding grit or contamination or large particles of cerium oxide that won't break down. I sweep off the offending particle with my finger, then wash my fingers under the shop sink. Then verify that there are no more large particles in the mixture on the mirror's face.
4. I spray a mist of water from my spray bottle onto the pitch lap.
5. I press the mirror with my hands for a minute.
6. I slide the mirror back and forth ensuring smooth contact with no hiccups.
7. I begin polishing.

Flash polish

MOT with 1/3 COC (Center Over Center) strokes.
10 second video of polishing machine in action. The machine strokes the mirror at half the rate I would by hand.
2 hrs (1 hr per mirror).
Mirrors show even shine from edge to center; no defects such as pits and scratches seen other than haze. Ronchi bands are straight with a 100 lines per inch grating.

Sagitta measures 0.157 inches (a 5/32 inch drill bit just fits) or F3.1, faster than the F3.5 that I had intended.

Polish out the mirrors

Continuing with MOT and 1/3 COC strokes.
5 hrs, swapping mirrors at the halfway mark. Shop temp 50F: stiffer cold pitch better for polishing out.
Both mirrors continue to fit against the pitch lap nicely, indicating that they are at the same radius of curvature.
Looks polished to the eye, laser shows considerable scatter diminishing slightly towards the mirror's center; no surface defects (scratches, pits); straight Ronchi bands, no TDE, no 'stig.

Continuing with MOT and 1/3 COC strokes.
5 hrs with swapping mirrors at the mid-point; shop 50F.
Usiung laser pointer, haze lessening: gone from center, more at the edge. ROC has shortened from 50 1/8 to 50 inches (F3.14 assuming clear diameter of 9.95 inch). Expect final F3.1. Ronchigram: straight bands, no TDE.

Shortening strokes to 1/4 COC, MOT, to bring more polishing action to the mirror's edge. Otherwise as above with shop temp at 50F.
6 hrs swapping mirrors at the mid-point.
Slight haze visible with laser pointer at mirrors' edge. Slightly oblate (anti-parabola).

Returning to 1/3 COC to flatten the oblate curve. MOT. Shop at 50F.
8 hrs swapping mirrors at the mid-point.
Very light sparse haze at extreme edge, visible with the laser pointer. Ronchi bands have straightened.

Continuing with 1/3 COC MOT.
6 hrs, swapping half way through.
Haze gone. Ronchi bands straight. Perhaps a very narrow thin TDE seen way outside ROC (could be grating diffraction)?

Time spent grinding and polishing both mirrors prior to parabolizing

Took about a month from start of rough grinding to end of polishing.

• Rough grinding by hand: 8 hrs
• Fine grinding by hand: 24 hrs
• Polishing by machine: 32 hrs. This machine's polishing is maybe half that of hand polishing. The rule of thumb is to expect an hour of hand polishing for each inch of aperture. So, 32 total hours, 16 hours on each mirror, the equivalent of 8 hrs by hand, which matches the rule of thumb.

Final state prior to parabolizing

I'll call the slightly thicker heavier mirror 'A' and the other mirror 'B'. The mirrors lost 1/16 inch thickness (approximately equal to the difference between F2 and F3.1 sagittae) so the 'A' mirror is now 1 3/16 inches thick and the 'B' mirror is now 1 1/16 inches thick.
Both mirrors' ROCs are 50 1/8th inches, unchanged from start of polishing. FL=25 1/16 inches and FR=3.13.

Both mirrors' RonchiGrams (RGs) are essentially the same. Note that all RGs are done with a 100 bars per inch grating.

Slight oblateness (anti-parabola) is to be expected after polishing. See my pitch lap calculator.

Tester stig

If the light source is offset diagonally from the grating, then the bands twist. Inside ROC they point to the light source, outside they rotate.

Parabolizing

Parabolic difference from a sphere is 12 waves.

Both mirrors have the exact same curve in that after pressing the pitch lap with one mirror, the other mirror slides across the lap perfectly.

Whatever I do to one mirror, I will do to the other, in order to keep them matched up.
And I'll use the polishing machine to slowly rotate the face up pitch lap, parabolizing the two mirrors by hand.

Two parabolizing approaches:

1. Standard MOT extreme W strokes.
   
2. MOT COC strokes on a parabolizing lap. Pictured here is my 6 ich F2.8, which has required a similar degree of parabolization. Note that this is a 7 inch oversized parabolizing lap for the 6 inch F2.8 mirror.
   

Parabolizing process: I'll use my process that I've developed over a number of fast mirrors. I'll gradually increase the parabolization of the mirror using my lightning quick Ronchi Calc to check overall progress, targeting any zonal issues. I'll use inside ROC Ronchigrams, later switching to outside ROC Ronchigrams as the mirrors' curve improves. Foremost I will keep the mirror's profile smooth. As I come closer I'll use my Ronchi Tape test for quick multi-figuring steps per session. As I near completion, I'll build a star test rig and add the star test to my evaluation. This way I'll finish the mirrors knowing exactly what I'll get at the eyepiece when the mirrors are silvered.

Began parabolizing with extreme W strokes. The water spray bottle has a few drops of Dawn dishwashing soap. I'm using generic or mixed cerium oxide from multiple sources. I use the hot air gun for a minute or two then push the mirror down on the pitch lap for a few seconds.

Initial calibration session #1 of 20 minutes (10 minutes each mirror) to gauge speed and parabolizing approach.
Straightened bands.

Session #2. Long wide Ws. 40 minutes.
Central depression has formed.

Session #3. Narrow Ws to widen the central depression. 1 hr.
Central depression widened. Surprised that I reached close to full parabolization this quickly, dismayed that the central depression hasn't widened to the edge yet to create a nice smooth curve.

Analysis shows high kink at 75% zone measuring ~1.5 waves off; central 25% a tiny bit overcorrected.

Looking outside ROC to better get at what zones are high and what zones are low. Here, the greatest issue is the 75% zone.

Session #4.

Several ways to fix a high zone.

1. Class Texereau method of TOT with the tool's edge on the mirror's 75% zone. This works but I don't like the possibility of scratches and sleeks that come with TOT.
2. Small sub-diameter tool polishing on the high zone. This has a strong tendency to leave a roughed, zoned surface.
3. Altering the pitch lap's contact where contact on the 75% zone is maximized and contact everywhere else is minimized. I've had good luck with this.

1. 1/4 CoC, to emphasize wearing down the 75% zone: 10 min
2. Extreme CoC, to smooth out the curve: 10 min
3. Extreme narrow Ws, to restore parabolization surely lost while working down the 75% zone: 5 min

Both mirror surfaces look defect free.

Session #5.

I switched to the Ronchi Band test, which is lightening quick and requires no analysis, just a glance at the band to mark lineup. I was able to do six sessions of five minutes each in under an hour. See my Ronchi Band Test discussion in my Ronchi Calc.

Since the 'B' mirror fell behind, I worked with it, 30 min, with Ronchi Tape Band checks every 5 min.
The bands fall a little short, indicating that the mirror is undercorrected. A quick matchup in the Ronchi Calc software shows about 90% corrected. Acceptable correction is 97% to 103%. The 'A' mirror looks similar.
Sadly, I also roughed up the surface. Contact was not perfect at first. I would do well to remember my rule, namely, to only work a session when motion is silky smooth with even resistance.
Overall, I did improve the 'B' mirror to match closely the 'A' mirror.

Session #6.

Working on the 'B' mirror. 45 minutes with quick Ronchi Tape checks every few minutes as warranted. Extreme W strokes with accentuated pressure at times over the lap's edge.
Radius of curvature is exactly 50 inches.
Much closer - the 'B' mirror is caught up to, if not slightly better, than the 'A' mirror - see the analysis - the 75% zone needs a touch more polishing.

Session #7.

Continuing with the 'B' mirror. My goal is to get the mirror close enough to justify a star test. Main defect is the high 75% zone.
30 min with checks every few minutes. Extreme W strokes along with arc strokes with accentuated pressure.
A hint undercorrected overall and at the extreme edge.

Session #8.

Improve the 'A' mirror so that it's ready for star testing.
Here's the starting Ronchigram analysis. As you can see, the mirror is undercorrected. Best fit parabolic correction is ~80%. Very close to the 'B' mirror at the start of session #6. Sessions #6 and #7 on the 'B' mirror comprised 1 hr 15 min of parabolizing time.

Way over parabolized in the first 15 minutes (120%); spent the next 15 minutes bringing the mirror back. 30 min total. Shop temp 60F. I warmed the pitch too much, softening it, so that the extreme W strokes caused the lap's edge to sink, losing contact there, causing over parabolization. I then used arc'd strokes accentuating pressure on the mirror's edge over the lap, to reduce parabolization.

Session #9.

Reduce the remaining over parabolization in the 'A' mirror.
4 min of arc'd strokes really flattened the mirror, reducing correction to 70% or so. At least the curve is smooth. Multiple sessions of approx 5 minutes each of extreme Ws brought back almost full correction. The 80% zone is lagging (high) just a tad. 56 min.

Session #10.
Lower the 80% zone. Which I did, but in the process I dug a low 50% zone. So after largely fixing that, I'm here. 47 min broken into 3-5 min quick checks.

Took two weeks to parabolize both mirrors.
10 sessions, 6 hrs 40 min total time for both mirrors.

Star testing: verify parabolizing.

Going into star testing, the 'A' mirror's 50% zone tests ever so slightly overcorrected (approx 1/20 wave, 0.5 M-L), the 'B' mirror tests nicely.
The mirrors:

For comparision, the finished Ronchigrams for my 6 inch F2.8, 10.5 inch F2.7, 13 inch F3.0 and 16 inch F2.9 mirrors. You can see that the 6 and 10.5 inch mirrors, tested with a 100 bars per inch grating, show slight deviations from straight. The 13.2 and 16 inchs tested with a coarser (but necessary because of the extreme parabolic deviation from spherical) show perfectly straight lines. The two 8 inch mirrors compare favorably to the 6 and 10.5 inch mirrors.

And comparing to my large thin fast Mirrors

First, build the star test rig

Initial star test results

In a month and a half, I ground, polished and parabolized both mirrors to matching focal lengths.

Designing the binoscope

Design criteria to consider (from my binoscope designer):

1. Optical path styles
2. Optical layout dimensions including secondary and tertiary sizes
3. IPD adjustment mechanism
4. Merging adjustment mechanism
5. Focusing: where focusing will occur and how
6. Altitude bearing placement
7. Tube assembly design

And here are the two designs together. Both have plusses and minuses.

What I like most and least of each design.

To be continued...

Retrospective

The good

1. Two mirrors came to the exact same focal length using a single tool alternating wets.
2. The Ronchi Tape Band test is lightening quick. I analyze the mirror and get it back on the lap in minutes. I spend half the time parabolizing and half the time getting ready to polish and cleaning and testing. I call this a parabolizing efficiency of 50%.
3. The Houdini coma correcting eyepieces are perfect for fast binocular telescopes.
4. Side by side binoculars have the same light throughput as reverse binoculars.

The bad

1. The full sized ring tool flattened the mirror's edge too much, leaving pits just inside the bevel that took extra time to grind through.
2. FR turned out faster than I had aimed for (F3.1 vs F3.5). Sagittae are 0.143 (F3.5) and 0.157 (F3.1), a difference of 0.014. To slow down to F3.5, I would need to stop down the mirror to 7.1 inches.
3. Warmer pitch being softer, I quickly overparabolized the 'A' mirror while using extreme W strokes. The pitch lap's outer zones were pressed down by the extreme strokes, leaving a high center on the pitch lap that quickly polished down the mirror's center.