Print through is the displacement (movement) of the unsupported face regions of a lightweight mirror as the polishing tool passes over them. As these areas bend, less polishing action is occuring. Conversely the area that has a rib behind it will bend less and will experience more polishing action. The resulting pattern, which mimics some or all of the rib patterns, is print through.

Print through should not be mistaken with common polishing artifacts. Annular ring zone(s) are most commonly caused from fixed-post polishing and/or sub-diameter tool work, where the tool's edges are allowed to repetitively hit the same radial position(s) on the optical surface. These types of artifacts will be present in a solid mirror also. Such errors are not print through.

The left image shows a different Dream mirror that was finished outside Dream. The Zygo color map shows rib correlation with some, but not all, of the artifacts seen in the interferometry map.

The level of print through in the Zygo data was substantially higher than Dream's engineering showed. It was later learned that the outside vendor was using 45-90 times the polishing pressure the mirror was designed for. The vendor's pressures were extremely high, even for solid mirrors. The outside vendor was told what polishing pressures the Dream mirror was designed for before the first mirror was cast.

This is a great example because it not only illustrates what actual print through looks like but it also illustrates how Dream's design & engineering can be disconnected from outside vendor processing. Examples like this, as well as many others over the years, have pushed Dream to become more vertical. Dream specializes in processing lightweight mirrors, with low MSF and an average of 6-9Å RMS surface roughness.

The vast majority of Dream's mirrors are designed for processing using conventional machines and polishing pressures, which we used while processing the 396mm f1.376 mirror. They are designed based on 0.25psi of polishing pressure, which equates to a 50 pound, full diameter tool on a 16" mirror. This should put the high pressures used by the outside vendor into perspective.

Click here to see additional interferometry of Dream mirrors that were processed correctly.

Simple designs, with little to no (real) mechanical engineering behind them, using simplistic repeating rib patterns, with fixed height & thickness ribs, often reduce design-related print through by using a thicker face. The main reason this and other simplistic, repeating patterns are so common is due to the fact that they are so quick, easy and cheap to produce, as well as making a traditional optician feel more at ease. Quick, easy and cheap are three adjectives that are never associated with a high performance product.

The face of a lightweight mirror is one of the largest drivers for the mass of the mirror. It is also one of the largest drivers for how quickly the mirror equalizes to ambient temperature. For a typical astronomical application, which is some upward-pointing angle for the primary mirror, the added mass of the thicker face displaces (sags) more during the gravity load case (real-world use of the mirror), compared to a thinner-faced mirror. So although a thicker face can reduce print through during polishing (all things being equal), the mechanical performance of the optical surface in final use, where it will spend 99.99% of its life, is made worse.

Traditional shops rarely test mirrors vertically, or in the actual mirror mount. Horizontally testing an optic puts the face (mostly) in shear, which means the face itself will displace less than the zenith-angle (vertical) case. Self-weight deflection that can be seen during vertical testing disappears during horizontal testing of a thick-faced "lightweight" mirror. When horizontal testing is combined with the lower polishing displacements of a thicker face, it gives the false impression that the optic is better off with a thicker face. Dream is an engineering firm that has never been bound by this type of narrow viewpoint. Dream pursues the performance of the mirror in final use, not the performance of the mirror on a test stand, in a temperature controlled room.

In the final instrument such a basic design using a thick face will weigh more, force other components to be heavier, will not be able to hold the figure as well (glass, especially when plate glass is used) and it will take longer to equalize. These are the same negative attributes of solid glass mirror technology, which has changed little over its more than 165 year history.

Dream's internal polishing has proven that print through, real and otherwise, can be caused by a dozen or more factors related to the polishing machine and tool. Better process control makes for a smoother surface on both lightweight and solid mirrors. Our thinnest face mirror to date (3mm thick for a 424mm physical OD mirror) has proven that Dream can provide our clients with the best of both words; a thinner face and a lack of print through. This provides a final mirror with unusually high performance. As one outside vendor stated, the figures of Dream's lightweight mirrors are just as stable as glass-ceramics and they can finished to the same high quality polish as well; L/20+ PV surface, L/125 RMS surface and 2Å RMS surface roughness. This level of quality was quantified using the second most powerful interferometer on the market, while vertically testing the mirror.

move on to page 2 of PRINT THROUGH