SDN 0003.24 – Surface Treatments

1. Introduction

This design note summarizes the surface treatments that will be applied to GNIRS parts, and the justification for the choices.

Basically, surfaces divide into two categories: low emissivity (reflective) surfaces and high emissivity (absorbing) surfaces.

· The bench exteriors and the shields, as well as parts attached to the shields and the bench outside, should be reflective. This is in order to minimize heat flow (by radiation) into the interior during normal operation.

· The bench interior and all parts inside the bench (mechanisms, optics mounts, etc.) should be absorbing. This is done for main two reasons: to help minimize scattered light, and to improve radiative coupling of the mechanisms to the structure during cool-down.

· Critical interfaces should not be treated at all. These comprise thermal interfaces which must conduct heat, and precision mechanical interfaces (i.e. mechanism interfaces but not cable clamps).

A discussion of the properties of surface treatments is provided in the Appendices and references.

2. Reflective Surfaces

The reflective surfaces can be left “as is” after machining and cleaning, with the exception of the shield components, which should be polished (and cleaned). It is recognized that the bench will be more emissive than it would be if polished, but it “sees” only the relatively cold driven (active) shield, so this is not a concern. Polishing the bench, which has a complex structure, would be time-consuming (and expensive). One consequence of this decision is that the design should ensure that the driven shield is kept at a temperature close to that of the bench, to minimize heat transfer from it to the bench.

Note that mounting plates and other parts that have surfaces on the outside of the bench as well as on the inside should be left bare on the outside surface and should be given an absorbing treatment on the inside surface.

3. Absorbing Surfaces

There are three types of absorbing surfaces: the insides of the major bench components, light baffles, and all other parts inside the bench, which include mechanisms and optics mounts.

The different surfaces will be treated differently. The choices reflect trade-offs between absorbing properties, durability, and ease of application (and rework), as well as the functions of the different surfaces.

3.1 Bench Interior

The bench interior will be painted with Aeroglaze Z306, except for interfaces. This should be done as late as possible, preferably after a fit check of all interfaces, in order to minimize rework.

This paint, properly applied, is reasonably durable, and has low reflectivity. There are many vendors capable of applying it, including at least one local vendor. Higher performance paints provide at most a factor of 2 improvement, and are more fragile and more difficult (and expensive) to apply. Black anodize has a reflectivity which is roughly 2x worse than Aeroglaze Z306; this factor alone does not rule it out, but there are also concerns about being able to mask off or re-machine all the critical interfaces on the bench, as well as the bench exterior. Areas that are re-worked after painting can be re-painted with care; this is not practical with anodizing.

The interior surface of large mounting plates should also be treated in this way; smaller parts (such as motor drive interfaces) can be treated as “mechanism parts” (3.2) on their interior surfaces.

3.2 Mechanisms, Optics Mounts and Similar Parts

These parts will be smooth, hard, black anodized. The process used at NOAO has reflectivity around 10%, which is adequate for locations other than baffles (see below). The mechanisms will receive enough handling that a cleanable surface is required. Also, some of the surfaces (slider, filter wheels and filter wheel baffles) need to be both hard and smooth, and therefore require anodizing rather than painting.

There are a few exceptions to this: critical interfaces (thermal or mechanical) should be masked or machined, and thin flexures should not be anodized.

It should be noted that the decision is to use smooth anodize rather than rough anodize (produced by bead-blasting before anodizing). The rough surface does not appear to increase absorption, as it scatters light instead of reflecting it specularly. For this reason, the performance is not much better for these applications and the additional steps are not therefore justified. A rough surface may enhance baffle performance, but it is not our intent to use anodized surfaces for baffles (see below).

Screws and other fasteners will not generally receive any special treatment, as they have small surface areas and should not be located in critical areas. The exceptions are for baffles (discussed below)

3.3 Light Baffles

The key to light control in the instrument is proper placement of baffles. Just trying to get everything as black as possible without further thought is unlikely to result in adequate performance. Baffles have been placed in the design as it progressed. Precisely because the baffles are critical to performance, it is important to ensure that they are as black as possible.

Different baffles will receive different treatment. Specifically:

· Most critical baffles. These are the slit mask and the Offner secondary baffle. Both should be treated with an ultra-black coating, such as Martin Black. Spares should be produced for at least the Offner secondary.

· Camera baffles. The baffles in the camera turret and the focus mechanism need to be black, but because of their proximity to the detector, paint presents higher risk than elsewhere in the instrument. Black Delrin is a reasonable compromise in this regard, since it is almost as black in the near-infrared as Z306. The concern in this area is not so much out-gassing as fragility, where mishandling might lead to paint flecks on the array.

· All other baffles. These will be treated with Aeroglaze Z306.

Screws represent a special case. Ideally, screws will lie well away from the defining edge of the baffle, or will be fasten from the back (without a through hole). In these cases, no special treatment is needed. If space constraints require exposed screw heads in critical areas, then the best solution is probably to paint the baffle after it is attached to the next level of assembly, thus painting the screws as well. Note that this implies re-painting the baffle if it is ever removed. Note that screws are stainless steel in order to minimize corrosion.

Appendix A – NOAO Black Coating Specifications

There are several relevant documents. General specifications for black surfaces are discussed in NOAO Technical Memo UNI.8001.0006. As discussed below (Appendix B) the information for reflectivity of black anodize is probably incorrect.

The specification for smooth hard black anodize is 89-NOAO-4205-0003. The specification for Aeroglaze Z306 is UNI.8001.0007.

Appendix B – Properties of Black Anodized Surfaces

The literature on the reflectivity on black anodized surfaces is not conclusive, and the quoted results in the Handbook of Optics are in fact contradictory. This must in part be due to the fact that a black anodized surface can be produced in a variety of ways, and may thus have a range of properties. The main concern (from our point of view) is a peak in reflectivity around 1 mm that is seen in some data.

Dick Joyce has made a series of measurements of reflectivity of black anodized surfaces. Three surfaces were investigated: rough hard black anodize, smooth hard black anodize (on a polished surface) and hard black anodize on Alumiplate. The reflectometer operates with a beam at about f/20, so only specular (or quasi-specular) reflection in measured. Dick’s measurement of smooth hard black anodize out to 2.5 mm is reproduced below, and shows an average specular reflectivity of 3-4%. There is no sign of a peak near 1 mm (or anywhere else). The data out to 5 mm are similar.

Visual examination of the samples shows that both the rough anodize and the anodized Alumiplate have much more scattering than the smooth anodized surface (which has much more specular reflection). The surfaces of all the samples are not so rough that multiple reflections are likely, so it is plausible that the rough and smooth anodized surfaces actually have similar absorption.

The anodized Alumiplate does appear generally blacker (visually) than the rough anodized sample, so it is possible that it has slightly higher absorption. If it is rougher and produces more large-angle scattering, though, the Alumiplate may not in fact be any better. The purpose of the original investigation was to show that anodized Alumiplate was not significantly worse than standard anodized surfaces, and not to explore its use as an especially black coating.

The Handbook of Optics (chapter 37) quotes an average absorption (emissivity) of ~88% for anodized aluminum. It is possible that the NOAO standard process produces slightly better performance, but the difference must in large part be due to scattered light “missed” by the reflectivity measurements. (It is also the case that the reflectivity of anodized surfaces appears to increase for grazing incidence – which case the reflectometer measurements will also be an underestimate.) For the purposes of our design, we assume that the figure of 88% is correct, but one should recognize that actual performance might be somewhat better.

By way of contrast, the average absorption of Aeroglaze Z306 is quoted as 96% from 0.3 to 2.4 mm and 91% from 5 to 35 mm. Data suggest a peak in the reflectivity approaching 10% in the 4-6 mm region (see Handbook of Optics).

If you have any questions or suggestions regarding this website, please contact Melissa Bowersock.

NOAO Intranet Services

NOAO Copyright
Statement

National Optical Astronomy Observatories, 950 North Cherry Avenue, P.O. Box 26732, Tucson, Arizona 85726,
Phone: (520) 318-8000, Fax: (520) 318-8360