Astro890 L8: So you wanna build an instrument?
Paul Martini, May 14, 2008
Today we will cover the process of building an instrument. There are four broad steps to
go from an idea to a working instrument on a telescope. These are:
- Step I: Science Case, Conceptual Design, and Funding
- Step II: Requirements and the Detail Design
- Step III: Construction and Testing
- Step IV: Commissioning
Most astronomers who contribute to an instrument project will contribute at Step I,
particularly with writing the detailed science case, which often turns directly into part of
the funding proposal. The astronomer might also develop the conceptual design. The next
most common area where astronomers may encounter the instrument is at Step IV, the
commissioning phase. Those who truly consider themselves ‘instrumentalists’ will also
work on Steps II and III. A subset of instrumentalists may actually do all of this work too,
at least for small instruments, but for the most part instrumentalists will meet regularly
with (and usually manage) the engineers, vendors, machinists, and technicians who do
this work as well as determine the detailed requirements of the instrument, evaluate
tradeoffs in the detail design (e.g. between cost, schedule, functionality), and help with
testing before deployment.
Now we will look at these steps in more detail. Keep in mind that the extent to which one
needs to follow these steps is proportional to at least the square of the instrument’s cost.
This means that simple & cheap instruments generally do not have a big staff and do not
go through multiple design reviews, while complex instruments for large telescopes have
large staffs and many many reviews, often by external committees.
Step I: Science Case, Conceptual Design, and Funding
Successful instruments are the intersection of what you want and what is possible. To
determine what you really
need (not just want), you need to understand your science case
very well. To determine what is possible, you need to consider: 1) how much money you
have; 2) what is physically possible to build. Examples of the second include
considerations such as the materials you need must actually exist (no instrument can be
built of Unobtainium), the telescope can hold the weight of the instrument, the instrument
will fit on the telescope, the dispersing elements can be manufactured, etc. While a
particular science goal will ultimately drive the instrument, the instrument design that
emerges from Step I will usually be the result of several iterations between the science
case, funding realities, and the conceptual design.