Optics: Lenses, Spectacles, Telescopes, Microscopes

Difficulty: a simple magnifier is moderate handwork; repeatable spectacles and imaging-quality microscope or telescope optics require demanding fabrication, measurement, mounting, alignment, and testing. Prerequisites: suitable transparent glass or tested optical crystal, controlled abrasives and silica dust, and pottery or metal for laps. Instruments also need precision mounts and measurement.

Why it matters

Lenses are the highest-leverage objects a time traveler can make, for one reason: they extend human senses beyond what the unaided body can do.

• Spectacles give a skilled craftsman or scholar an extra 20+ productive years. Presbyopia hits everyone around 45, and in a world without reading glasses, that's when scribes, jewelers, and surgeons stopped working. The invention of spectacles in 13th-century Italy is plausibly one of the largest single boosts to intellectual productivity in history.
• Microscopes make microorganisms directly observable. That supports germ theory, but a view of animalcules alone does not show that a particular organism causes a disease; controlled epidemiology, transmission evidence, cultures or staining, and contamination controls are separate parts of that case.
• Telescopes overturn cosmology, and are immediately practical for navigation and for seeing who's coming over the hill.

Start with the glass

Everything depends on the blank. Go back to glass and make the best you can:

• Bubble-free — hold the melt long at high temperature to fine out the seeds.
• Suitable transmission and color — selected by test; a magnet removes only free magnetic grains from sand and manganese does not repair every impurity.
• Homogeneous — minimal striae (streaks of differing refractive index). Furnace mixing and fining must be designed to reduce rather than introduce gradients, bubbles, or contamination. Polishing cannot remove a bulk defect.
• Well annealed — internal stress causes birefringence and, eventually, cracking.

Cast blanks slightly thicker and larger than the finished lens. For a first attempt, small is much easier: a 25–40 mm blank is far more forgiving than a big one.

Rock crystal (natural quartz) is a genuine alternative and was used historically — the Visby lenses and early "reading stones" were beryl and quartz. It's harder to grind but naturally free of bubbles and striae. If your glass is poor, find good quartz.

Grinding: the process

The working principle is controlled abrasion between a workpiece and a tool of opposite curvature. Tool/work diameter, stroke length and pattern, overhang, rotation, pressure, support, and wear determine whether the surface approaches a useful sphere or develops zones and edge error. It does not converge automatically; measure curvature and figure repeatedly and use corrective strokes.

Make a lap

A lap (or tool) is a dish of the opposite curvature to your lens: concave to grind a convex lens, convex to grind a concave one. Make it from cast iron, bronze, hard fired ceramic, or hardwood faced with pitch. Rough it to approximately the right curve.

Choosing the curve: the focal length of a thin lens follows

1/f = (n − 1) × (1/R₁ − 1/R₂)

For typical glass, n ≈ 1.5, so n − 1 ≈ 0.5. For a plano-convex lens (one flat face), that simplifies to f ≈ 2R — the focal length is about twice the radius of curvature. So a lens ground on a lap with a 200 mm radius has roughly a 400 mm focal length. That one rule of thumb is enough to design most of what follows.

Grind

1. Rough grinding — coarse abrasive (sand, crushed glass, emery) and water between blank and lap. Work with a randomized stroke: overhanging strokes across the center, rotating both the lens and yourself around the lap. Never repeat the same path — regular strokes create regular errors.
2. Fine grinding — progress through finer and finer abrasive grades. Wash everything — hands, lap, lens, bench — completely between grades. One stray coarse grain undoes an hour of fine work. This is the discipline that separates success from endless frustration.
3. Test by holding the lens up and checking the grey surface for pits from the previous grade. Move on only when the whole surface is uniformly fine.

Abrasives, coarse to fine: sand → crushed quartz → emery (natural corundum, sorted by settling time in water — coarse settles fast, fine settles slow, so decanting gives you graded grits for free) → fine emery → tin oxide or rouge (iron oxide, made by calcining iron sulfate — see sulfur).

Polish

Switch to a pitch lap: coat the tool with pine pitch softened with tallow, press the lens onto it while warm to bed it to the exact curve, and cut channels across the pitch to carry away slurry. Polish with rouge and water.

Pitch is a slow-flowing solid — it conforms to the surface under sustained pressure but acts rigid under quick strokes. That's precisely why it polishes without destroying the figure. This is still how precision optics are polished today.

Polishing takes hours and the surface goes from grey to clear gradually. Be patient; rushing gives a surface that looks clear but scatters light.

Testing without instruments

• Focal length: project a distant terrestrial object onto a nonflammable screen and measure the distance. Never look at the Sun through a lens or telescope, and do not focus sunlight onto paper or skin; concentrated light can cause permanent eye injury or fire.

• Whole-optic image test: look at a distant point light through the mounted lens. A clean compact image is encouraging, but flare, streaks, or haze can come from surface roughness, figure, aberration, bulk striae, dirt, mounting, alignment, or the source itself. This test does not identify surface quality by itself.

• Figure: float a straight edge across the surface against the light, or compare against a known-good test plate. If two ground surfaces of matching curvature are pressed together with a film of liquid, they'll show interference fringes — this is the principle behind Newton-ring and optical contact tests.

  Read them cautiously: use a known reference surface, clean controlled contact and support, and preferably monochromatic light of known wavelength. Fringes map the difference between the two adjacent surfaces, not the test piece alone. Straight fringes on nominal flats commonly indicate a wedge; curved or circular fringes show changing separation, which may include radius mismatch, local figure error, dirt, support distortion, or contact pressure. One fringe interval represents about half a wavelength of gap change in reflected light. A sparse or absent pattern is not, by itself, proof of a perfect surface, and the sign of an error cannot be inferred from one universal "press and watch" rule.

What to build

Spectacles (start here)

Convex lenses correct hyperopia and presbyopia (near vision — the common age-related loss). Concave lenses correct myopia (distance vision), which came later historically because concave surfaces are harder to grind.

Optical power in diopters = 1/focal length in metres. A +2.00 lens has a 500 mm focal length. Make a graded set — +1.00 through +4.00 — mount them in wood, horn, bone, or wire frames. A low-power reading-lens trial can help a person with uncomplicated presbyopia, but it is not complete optometry: each eye may differ, astigmatism and eye disease require more than spherical power, and poor centration or fit can cause strain and falls. Screen for pain, redness, sudden or one-sided change, field loss, flashes/floaters, injury, or persistent poor vision and refer to skilled eye care. Record each eye, working distance, lens power, centration, and outcome rather than selling one age-based answer to everyone.

Magnifier and simple microscope

A single strongly curved lens is a magnifier. Magnification ≈ 250 mm / f, so a lens with a 10 mm focal length magnifies about 25×.

Van Leeuwenhoek's approach used very small single lenses: melt the end of a fine glass rod in a flame until it forms a tiny bead. A small sphere is a very short-focal-length, surprisingly good lens. Mount a 1–2 mm bead in a hole in a metal plate with a specimen on an adjustable pin in front of it. He reached roughly 200–270× with his best instruments and reported bacteria in 1676. Forming a bead may be quick; obtaining useful high magnification also needs aperture control, a well-formed lens, rigid mounting, fine focus, illumination, specimen preparation, and practiced observation. Treat Leeuwenhoek's results as an expert historical upper bound, not an afternoon guarantee.

Compound microscope

Objective lens (short focal length) forms a magnified real image inside a tube; eyepiece magnifies that image again. Total magnification ≈ (tube length / f_obj) × (250 / f_eye).

Harder than it sounds, because aberrations compound. Early compound scopes were worse than Leeuwenhoek's single beads. Do the simple one first.

Telescope

• Galilean (1609): long-focus convex objective, short-focus concave eyepiece placed inside the focus. Upright image, narrow field. Simple.
• Keplerian: convex objective, convex eyepiece beyond the focus. Inverted image, much wider field, and it allows a crosshair reticle at the focal plane — which is what makes it a measuring instrument, not just a viewing one. Magnification = f_objective / f_eyepiece.
• Newtonian reflector: a concave mirror instead of a lens, with a small flat diagonal bringing the image out the side. Mirrors have **no chromatic aberration, but the primary must be accurately figured—often toward a paraboloid rather than merely polished spherical—and the diagonal must be optically flat. The instrument also needs stable support and collimation. Historically speculum metal (see copper and bronze, ~30% tin, takes a brilliant polish but tarnishes and needs repolishing). Better: grind glass and use a validated chemical silvering process in a controlled laboratory. Silver nitrate is an oxidizer that burns/stains tissue, its preparation can involve nitric acid and toxic gas, and the waste is regulated metal contamination; see saltpeter and silver.

The aberration problem

• Chromatic aberration — glass bends blue light more than red, so a simple lens focuses colors at different points and everything is fringed in purple. Early astronomers fought this with absurdly long telescopes (Huygens used 50-metre focal lengths) because longer focus means milder color error. The real fix is the achromatic doublet (Hall/Dollond, 1730s): pair a convex lens of ordinary crown glass with a concave lens of flint glass — glass made with lead oxide, which has much higher dispersion relative to its refractive index. The two color errors cancel while the focusing power doesn't. This is a beautiful piece of physics, but it requires measured refractive index and dispersion, calculated powers and radii, homogeneous glass, matched centering, spacing, and testing—not merely one crown and one lead-glass lens. Lead-glass production issues are in glass. (Handle lead with the cautions in silver; lens glass isn't a food-contact use, so the risk is in the making, not the using.)

  Cemented and deliberately air-spaced achromats are both valid designs. Cement only when the calculated design has matching internal radii and the adhesive is optically clear, compatible, stable, clean, and bubble-free. An air gap adds surfaces but can be an intentional optical variable, not automatically a flaw.

• Spherical aberration — a sphere isn't the ideal focusing shape; edge rays focus short. Fix by stopping down (an aperture mask over the lens edge — free, and it also improves contrast), by using long focal ratios, or eventually by figuring a paraboloid for a mirror.

• Fix the easy ones first: stop the lens down and use a long focal ratio. Both cost nothing and cure most of what a beginner's telescope suffers from.

Other optics worth knowing

• Camera obscura — a pinhole (or a lens) in a darkened room projects an inverted image of the outside. Needs no glass at all with a pinhole. Used for drawing aids, and for safely observing solar eclipses.
• Burning glass — a large convex lens focusing sunlight reaches genuinely high temperatures and was used in real 18th-century chemistry. Free heat with no fuel.
• Mirrors — polished bronze or speculum metal is the ancient answer. Glass mirrors backed with a tin–mercury amalgam (the Venetian method) are far better but mercury is seriously toxic — Venetian mirror workers were poisoned routinely. Use the silver-nitrate modern controlled silvering avoids mercury but still needs trained chemical handling, ventilation, compatible reagents, and captured silver waste. It is not a domestic "safer" recipe merely because nitric acid is available.
• Prisms — split white light into a spectrum. Newton's demonstration that white light is composite is one of the cleanest and most convincing experiments ever devised, and you can do it with one prism and a dark room.
• Telescopic surveying instruments — a theodolite needs a telescope and reticle on controlled horizontal and vertical axes, angular circles or arcs, leveling and centering/plumbing, clamps and fine motion, and calibration by reversal. A telescope on one graduated circle is not yet a theodolite.

Failure modes

Sources and limits

NIST's work on optical and stylus surface measurement is a useful reminder that even modern surface methods have range-dependent disagreement. Reading a focal length, fringe, or magnification is not by itself a complete accuracy statement. Optical performance also depends on bulk material, figure, roughness, coating, mounting, alignment, illumination, detector or eye, and environmental conditions; see the repository-wide source policy. NIST's optical-flat calibration study also shows that quantitative fringe interpretation compares surface profiles under controlled illumination, support, and repeated measurement.

Era notes

Burning glasses and reading stones are ancient — the Nimrud lens is ~2700 years old and there are Roman references to using a glass globe of water to magnify. Spectacles appear in northern Italy around 1286; the telescope in the Netherlands in 1608; Leeuwenhoek's microscopes in the 1670s; the achromatic doublet in the 1730s.

The striking thing is how late all this is. Romans had glass, abrasives, and metalworking sufficient to make spectacles, and simply never made the conceptual connection. There is a plausible 1500-year gap between when spectacles could have been made and when they were. That gap is your opportunity, and it's probably the largest gap in this repository between what a society could have built with what it already had, and what it actually built.