Subaru 8-meter Telescope Damaged by Leaking Coolant

[/caption]

A “serious hardware incident” has shut down the Subaru Telescope indefinitely. A leak allowed orange-colored coolant to spill over the primary mirror and into the main camera, as well as into other instruments and the structure of the telescope. The damage is still being assessed. During the clean-up and recovery of equipment, nighttime observations have been suspended, as well as daytime summit tours of the telescope.

An announcement posted on the Subaru telescope website said that operators detected an error signal while shutting down the observation system at the end of the night shift during the early morning of Saturday, July 2, 2011.

When engineers arrived to assess the situation, they found extensive leakage of coolant (ethylene glycol) over most of the entire telescope. The leak originated from the “top unit” of the telescope, which is located at the center of the top ring and includes the Subaru Prime Focus Camera (Suprime-Cam) and auxiliary optics.

Although they promptly shut off the supply of coolant, a significant amount of leakage had already occurred, from the top unit itself down to the tertiary mirror, the primary mirror and some of its actuators, the Faint Object Camera and Spectrograph (FOCAS, a Cassegrain instrument) and its auxiliary optics, and the telescope floor.

The engineers attempted to clean up and remove as much coolant as possible. However, such areas as optics, control circuits, and the inside of Suprime-Cam and FOCAS were inaccessible during the initial clean-up.
The coolant consists of a mixture of water and ethylene glycol, a liquid commonly used in a vehicle’s radiator for cooling. The coolant is not corrosive and does not damage the primary mirror, which has a foundation of glass.

The Subaru Telescope is located on the Mauna Kea on the Big Island of Hawaii, with offices in the town of Hilo. The Subaru website said they will post updates on the status of the telescope and its recovery.

Source: Subaru Telescope website

27 Replies to “Subaru 8-meter Telescope Damaged by Leaking Coolant”

  1. I sure hope they can fix this without major cost and time.
    Strange the designers never had these kind of failures built in in their design of the telescope. One should have thought that a potential leakage of the coolant fluid might cause extensive damage?

    1. A difficult proposal. Efficient cooling means bringing the cooling surfaces close/in contact with the active parts of the equipment; using liquid coolants means having sufficient flow sufficiently close.

      One way to cut down on non-avoidable risks is possibly to use heat pipes, minimizing active coolant volume. But they are problematic to work with, and the narrow work zone is probably not suitable for telescopes.

      Actually cooling/heating of equipment (and other stuff like vehicles and houses) are interesting areas where people can come up with innovative ideas. Some of them may actually work, too. =D

      Ultimately this is a design/know how/cost issue.

    2. A difficult proposal. Efficient cooling means bringing the cooling surfaces close/in contact with the active parts of the equipment; using liquid coolants means having sufficient flow sufficiently close.

      One way to cut down on non-avoidable risks is possibly to use heat pipes, minimizing active coolant volume. But they are problematic to work with, and the narrow work zone is probably not suitable for telescopes.

      Actually cooling/heating of equipment (and other stuff like vehicles and houses) are interesting areas where people can come up with innovative ideas. Some of them may actually work, too. =D

      Ultimately this is a design/know how/cost issue.

    3. I’m one of the operators of the Subaru Telescope, although at the time of this incident, I was 5,000 miles away on vacation (whew! alibi!).

      Obviously, any time you have coolant fluid for mechanical systems (and that’s how we use it – the detectors inside the instruments are cryogenically cooled with liquid nitrogen or closed-cycle helium, neither of which is much of a risk if they leak!) you want it to stay inside wherever you put it and not go all over the place. But of course, the cooling system is a requirement that we can’t really get away from, and the prime-focus instrument rotator is something we can’t exactly give up either – I believe Subaru’s unique among 8m+ telescopes (until LSST is built) in having that kind of capability.

      So the engineering really goes into designing every last detail of the telescope with limits defined in the control software, AND physical trip switches, AND physical stops. And when something manages to break despite all those checks… well, bad things happen. 🙁

      As far as I know, it was a malfunction of the rotator for the prime-focus unit, or the “cable wrap” that stores slack in cables and hoses for it. Why none of the safety measures prevented it, I don’t know, and I’m sure the engineers will be investigating that while they clean things up.

      Just my 2¢. 🙂

      1. Thank you Daniel! As (a very) amateur astronomer, I really appreciate it when a pro such as yourself gives us an insight into the workings of such an amazing behemoth of an instrument!

        I am glad to hear the ethylene glycol will not damage the primary, as that would obviously have been one heck of a re-aluminizing job with such a large surface area?

        Here’s hoping things can be salvaged OK soon.

        Thank you for the science and enlightenment you are helping provide humanity.

        Wezley

    1. Yes, most likely they were using the telescope to spy on al qaeda’s secret bases. They needed to be stopped.

  2. Hopefully this won’t happen to James Webb telescope once it is in orbit.

  3. In my experience of cooling liquids in instruments, “not corrosive” is more a general hope than something observed. For instance, using distilled water to make the liquid well specified and without salts makes water-based liquids very aggressive.

    Now as I understand it ethylene glycol makes up for much or all of that problems, so here’s hoping it will go well!

  4. Hopefully observations will resume soon, as time on 8-10 meter class scopes is pretty hard to come by.

    Somewhat related, a paper appeared on arXiv recently looking at the number of refereed papers produced by 8m+ ground-based telescopes from 2000 to 2009. The telescopes analyzed included Subaru, HET, Keck(2), Gemini(2) and VLT(4). It’s interesting to compare the scopes’ output (keeping in mind that some scopes have very specialized instrumentation not found at other observatories) and read some of the suggestions for future Extremely Large Telescopes and accompanying instrumentation soon to be built.

    “Paper Productivity of Ground-based Large Optical Telescopes from 2000 to 2009”: http://arxiv.org/abs/1107.0353

      1. I have to disagree 100% with your comment about arXiv. Most papers are posted on arXiv before they are published by scientific journals, however they are nearly always at the final stage prior to publication so they are usually the exact paper which is publsihed. In general this means they are peer reviewed and effectively published in the scientific literature. Therefore, arXiv is a fantastic source for good science papers, although you need to be a little bit careful because not all papers on there are at the stage I described above.

      2. I have to disagree 100% with your comment about arXiv. Most papers are posted on arXiv before they are published by scientific journals, however they are nearly always at the final stage prior to publication so they are usually the exact paper which is publsihed. In general this means they are peer reviewed and effectively published in the scientific literature. Therefore, arXiv is a fantastic source for good science papers, although you need to be a little bit careful because not all papers on there are at the stage I described above.

      3. Indeed I have found numerous good scientific papers on arXiv — my hard drive is bursting with them!

    1. Shouldn’t that be “Photo Productivity”, the final result be damned? 🙂

      These things are very good to work on, I have some similar experience from industry where a few large plants can be compared for productivity lessons learned. (One plant teach you more about *im*productivity does and don’ts.)

      I recognize the ramp-up and reliability issues. Both of those are supported well by having fast turn-around of observation-build or later observation-fix cycles, i.e. a divide and conquer approach. You _do not_ want the large scale approach that large scale projects invite!* Except for initial and overall planning, which actually works, believe it or not. =D

      ——
      * A successful manager once told me that “you don’t drive cars by looking in the back mirror”.

    2. Shouldn’t that be “Photo Productivity”, the final result be damned? 🙂

      These things are very good to work on, I have some similar experience from industry where a few large plants can be compared for productivity lessons learned. (One plant teach you more about *im*productivity does and don’ts.)

      I recognize the ramp-up and reliability issues. Both of those are supported well by having fast turn-around of observation-build or later observation-fix cycles, i.e. a divide and conquer approach. You _do not_ want the large scale approach that large scale projects invite!* Except for initial and overall planning, which actually works, believe it or not. =D

      ——
      * A successful manager once told me that “you don’t drive cars by looking in the back mirror”.

Comments are closed.