I cast a second 5” diameter specula. Beforehand I coated the mold with boron nitride. The coating reduces the puckering in the metal where it contacts the mold walls. A torch held on the pool of solidifying metal lead to an curious pattern. How it maps to sub-surface defects will be interesting.
The first 5” (127mm) diameter speculum mirror blank is in polish. I’ve continued to watch carefully and document the development of defects. First when I was sanded and now in polish.
Polishing is exhaustingly slow but there’s enough of a polish for the mirror to form an image with a soft halo of scattered light around it. Enough to measure the 50” focal length I printed into the tool1
Sanding through the blank hogging out the center I chased dendritic defects in the metal. Bronze is notoriously pit-ty. The de-gassing, de-oxidizing steps I used casting reduced but didn’t eliminate it.
Dendritic distribution of tiny pits. A new observation. Regular pitting is also present.
Sanding proceeds from left to right.
My observations have lead me to speculate that it was not fully degassed2. I turned that step up to eleven in the new casting. The deoxidizing step I left the same.
The first3 blank is by no means perfect but there is a large region, an annulus of metal nearly free of pitting. By masking off the center and a few lone spoilers I expect to see a noticeable reduction of scattered light.
The making of the 50mm mirror. Note the pits visible in the detail at the lower left
Speculum telescope mirror for use in a Gregorian telescope. The mirror was made by James Nasmyth between 1840 and 1850
Detail of the Nasmyth mirror. Note the extensive pitting.
My 5” speculum while being sanded flat. Note the absence of pitting
A region toward the center of the first 5” blank. This is typical pitting. The dark line is the edge of the counter reflected in the partially polished mirror blank. The cabinets are grey. I used them as a distributed light source to take this picture. The white lines are scratches that haven’t polished out yet.
1) I noticed the mostly hollow 3D printed tool deforming as I pressed the mirror blank down and back and forth over sandpaper glued to it. I reprinted it with 100% infill. It took 8 hours. The new 5in f/10 tool is 193 grams of solid white plastic. (6.8oz)
2) Hydrogen is called gas. With enough energy to overcome their political differences, hydrogen and oxygen form water. Within liquid metal they from steam. The steam then either dissociates again or leaves the melt.
Dissolved solids and gasses in liquids are pushed out as the liquid solidifies (freezes). The molecules line up against each other and squeeze out everything that doesn’t fit. Bronze is particularly soluble to oxygen and hydrogen. As the metal freezes, the gasses get pushed out of the forming grains. Limited by diffusion they aggregate between them. Where hydrogen runs into oxygen steam happens and as the metal is becoming a solid a bubble forms and stays put. Then when sanded through, the bubble becomes a pit. Without dissolved hydrogen, this can’t happen. Hydrogen can be removed by adding oxygen to the melt. The hydrogen oxidizes. It burns. To do this I stir in copper oxide. More than I need so the metal is saturated with dissolved oxygen. To deoxidize the molten metal I add phosphorous in the from of SF-1 brazing rod. Phosphorous oxide is liquid and lite and floats. Tin is added to compensate for all the extra copper.
3) Technically the first blank is actually the second but that disk of metal was wedged because I didn’t adequately level the mold before pouring the metal. I shattered that disk with a single hammer blow to the center (speculum is a brittle as glass) and melted the pieces for the next new “1st” casting.
