1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Other Career in developing scientific instrumentation

  1. Feb 14, 2016 #1
    I've been considering a change in career direction recently, and have been thinking about a career in developing scientific instrumentation. I know that "scientific instrumentation" is vague and encompasses a lot, hence the reason I am posting -- to try sharpen my understanding of exactly what types of careers are out there.

    As a starting point: there exist teams responsible for building the Hubble space telescope, LISA, the LHC, etc. This is the type of "instrumentation" I am talking about -- large scale scientific experiments. Obviously, a huge number of people are involved in such projects. Can someone provide me with a broad categorization of the type of roles people play in such projects along with the type of preparation best suited for such roles (physics, EE, masters, phd, etc.) and where/for whom such people work (private companies, universities, etc.)?
  2. jcsd
  3. Feb 14, 2016 #2
    There are a lot of positions for elctro-optical engineers. Look into Thor Labs or Nufern to start off with. In terms of preparation I'd look into getting a PhD in Physics or EE with a specialty in Optics or something along that line. In terms of semiconductor production, you'd pretty much need a PhD to work there. If you wanna work on Hubble, try going for an EE position at NASA.
  4. Feb 16, 2016 #3


    User Avatar
    Science Advisor

    I design scientific instrumentation for a living. The group I work in has designed components for many detector systems including the LHC.

    We need a variety of skills to do our work successfully. In our group, about half the staff have PhDs (either in Electrical Engineering or Physics). The key is to get experience working on large system development as a graduate student. I worked in the group while I was still a student, then worked in industry for 8 years before I returned. Most of the other staff also worked in the group as a student or post-doc. So the point there is GET AN INTERNSHIP.

    I work for the Department of Energy at a National Lab. A lot of instrumentation work goes on there (in the USA). A lot of instrumentation work is also done in NASA (particularly at JPL). Scientific instrumentation is a niche area so there isn't a lot of industrial work in it, but defense projects can be similar. So, if you are a citizen of the US, you can do similar work for a defense contract such as Boeing or Northrup Grumman.

    From my perspective it is easiest to get into the field with a PhD in Electrical Engineering. That is because there is much more competition from industry for EE people capable of designing systems. Since the pay is much lower working for the government you will be competing with fewer people. We sometimes struggle getting top people because of the pay. On the physics side it is hard to get to work here because it is a very desirable job.

    To summarize, try really hard to get an internship at a DOE lab or NASA or defense contractor while a student. Even better if you can do it in grad school. Also, don't be a jack of all trades... be very good at one thing but be flexible as the need arises.

    Good luck!
  5. Feb 16, 2016 #4
    Thanks for the replies! analogdesign, can you describe what your job entails? What is involved in the projects that you and your coworkers work on? For instance, is it more common to design a specific circuit board, or to engineer how all the parts fit together to make the instrument function? How much of the work is theoretical modeling/designing and how much is hands on building/experimentation?
  6. Feb 17, 2016 #5


    User Avatar
    Science Advisor

    We work on all aspects of detector instrumentation from sensors to software, as we like to say. So to answer your question about whether it is more common to design a specific circuit board, or to engineer how all the parts fit together to make the instrument function I would say both.

    Basically, if an experiment can buy hardware and software off the shelf, they will. Therefore, anything we do is not available in the market. So, we have to do a good bit of theoretical modeling and design to see if we can actually achieve our goals. Then we see what we have to develop. I would say our work is mostly oriented to building instruments that are used to do science but often we have to do some research on some sticky problems.

    My specialty is custom integrated circuit design. We also design the boards the chips go on, the firmware for the FPGA (for example data acquisition and transport hardware) and the software to operate the instruments or components. We even develop sensors when necessary (for example photo diodes or CCDs). As you can see we are responsible for everything related to getting the instrument working, from designing it, to building it, to testing it, to getting it installed. It's a lot of fun.
  7. Mar 9, 2016 #6
    Thank you for your insight @analogdesign, and for any others who may share their experience.

    I'm studying electrical engineering with a concentration in circuits and systems or IC design (this is dependent on my study abroad application). I am highly interested in instrumentation and I was introduced to it in past material science internships. For this reason, I thought it would be a good idea to have a competitive background in chemistry compared to my peers, and I have been taking more advanced chemistry coursework for my technical electives. Examples of these courses are quantum mechanics in the physical chemistry series, organic, and inorganic chemistry. I will be taking molecular spectroscopy this coming quarter. I just completed a new course offered at my institution called "Open Source Hardware Movement", which we learn about it and look at prominent projects, and I was able to make a contribution to the project I was looking at more thoroughly; also, I'll be taking another newly available course called "Rapid Prototyping". I think this class will be fun, interesting, hands-on, and helpful for my goals. Now, I am having concerns now that I am hearing your advice "don't be a jack of all trades." Would it be better to more strictly take circuit coursework?

    Also, since you did the IC design depth, do you think this was a good way to go, or would you recommend another concentration? Are there any courses, concepts, or skills you wish you had achieved prior to your position? Common mistakes? The main downfall with my institution, in my opinion, is that we don't have an instrumentation depth nor a course covering concepts centralized about instrumentation.
  8. Mar 17, 2016 #7


    User Avatar
    Science Advisor

    Sorry it took me a while to get back to you. I enjoyed IC Design and if you like it too, it is a good concentration. I would say the best concentration is something you are passionate about, because if you love something you will probably be more effective at it.

    I never took any course on instrumentation. Reading out a pixel detector really isn't any different from a hard-disk read channel or a radio... a few of the blocks are different and the emphasis is different but it is very similar.

    Regarding the "jack of all trades" thing, it is a touchy subject. Companies want highly specialized employees, it works for them. The danger is what happens if your highly specialized skills go out of fashion?

    If you'll indulge a decent into nerdiness, I like to think of a skillset in engineering as a bandpass filter. You want to have a key skill that you are known for and you hang your hat on. For example, for me it would be "transistor-level analog IC design". Then you want to have abilities on either side of that level of abstraction but you also want to have a roll off (either 6 or 12 dB/octave). For example, I consider myself strong in Analog IC Design, less strong but competent in Digital IC Design, still competent in board-level Design, and kind of week but passable in Software. You can't be great in everything but for an instrumentation job it is best to be great in one thing and good in a lot of others. Does that make sense?
  9. Mar 18, 2016 #8
    That's a pretty good clarification. It makes good sense to me. Thank you for the reply! Any reply is helpful, and better than no reply ;).
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted

Similar Discussions: Career in developing scientific instrumentation
  1. Career Developer (Replies: 6)