With the launch of Planck and Herschel only two days away, excitement is reaching fever pitch. As the countdown inches slowly towards the moment of reckoning the tension mounts…

This post would have been a bit more exciting if all that had been true. Of course we now do have a definite launch window for Planck, 14th May 2009. The launch window opens at 14.12 BST and will remain open for about two hours. Let’s hope they manage to get the thing up in that time, otherwise there’ll be yet another substantial delay.

Planck will be launched with its sister-mission, Herschel.  They will both be carried by an Ariane 5 rocket from the European Space Agency’s launch site in Kourou, French Guyana. Within half an hour of launch, Planck and Herschel will separate and start on their journeys.  While both satellites are going to orbit the second Lagrangian Point (L2), they will have slightly different orbits.  It will take Planck around 6 weeks to get to L2, during which time it will start to cool down its cryogenic systems. Eventually it will be the coolest thing in space.

Of course that is all very exciting, but it would have been a lie to say that the excitement is mounting that much back here at home. Together with the fact that the undergraduate examination period is upon us, the department is extremely quiet and those that are most nervous have taken their jitters to South America. The fact is that most of the people directly involved with Planck or Herschel have actually been invited to the launch and have either already made their way there or have at least set out on their journeys to the jolly.

We do, however, plan to have a small function here to mark the  launch on Thursday with wine and nibbles and talks about the science. I hope it’s not tempting fate. I”m not exactly nervous myself, but probably will get butterflies as we watch the launch on the net. Still, there’ll be wine to steady our nerves…

I  remember very well the “launch”, in 1996, of a mission called Cluster which many of my colleagues at Queen Mary were heavily involved. This was the first flight of Ariane-5. Bugs in the software meant it lost control shortly after launch and the party very soon turned into a wake, although the resulting fireworks were quite spectacular.

Because the Ariane-5 vehicle was brand new, and somewhat untested, the European Space Agency had decided to take advantage of an offer to launch the mission without charge. This seemed like a good deal because the costs of putting an experiment in space are a sizeable fraction of the overall budget for such missions. It turned out, though, that the old expression was true. There’s no such thing as a free launch.

In fact, Cluster did eventually fly using flight spares and a launch on a Russian spacecraft. If Planck and Herschel go boom then there’s no way they can be replaced. It would be a terrible thing if this happened, for a large number of reasons, but Ariane-5 has launched many times since then, and I’m confident that both Planck and Herschel will soon be safely on their way to L2.

But don’t expect any science immediately, especially not from Planck. It will be years before the key science results emerge and, until then, the science team is sworn to secrecy….


17 Responses to “Planckety-Planck”

  1. I’m not a hardware guy, but surely most of the cost (apart from the launch)
    is in the development and test of the components, planning the experiment
    etc etc (and most of these are probably salary costs). How much more
    expensive would it have been to build two spacecraft, or how much more
    would it cost to build a new one if the launch ends in disaster?

    Even if a second launch is an additional cost, surely it would be essentially
    the only additional cost, if my assumption that building a duplicate would
    cost just a fraction of the total budget.

    What about insurance? Are launches insured? Since there aren’t that
    many launches, and not that many organisations who pay for launches,
    of course it is cheaper not to be insured and accept the odd crash. But
    such a crash hits those involved in that particular mission rather hard.
    With insurance, each mission would cost a bit more, but no-one would be
    really hard-hit (assuming, again, that the cost for a replacement
    spacecraft is small compared to the other costs).

  2. telescoper Says:

    The reason Planck and Herschel are sharing the launch is because of the cost. The cost of an Ariane-5 launch runs into hundreds of millions of euros at commercial rates, a significant fraction of the cost of the whole mission.

    Scientific missions like this are not insured. It would be too expensive, even if an underwriter could be found.

  3. For simplicity, assume one mission per launch.

    Say one launch in 10 fails. What options do we have?

    1) Do nothing: 1 mission in 10 fails to get launched and is destroyed.

    2) Insure each launch for (10% of launch cost + profit for underwriter): failed missions can be relaunched.

    Assuming that the cost of building a new spacecraft is negligible compared to the other costs (I don’t know if this is true), then the total cost per SUCCESSFUL mission is the same in both cases, except for the profit for the underwriter. The underwriter’s profit is the only additional cost, but in the case of insurance it means that no research group loses out big time if its mission fails to be launched successfully.

    Why aren’t missions insured? Is it because the underwriter’s PROFIT would be too large (say, more than a few per cent of the cost of a launch) or is it because, even if all the development and testing has been done, the costs of building a replacement spacecraft are too high? (Even in the latter case, why not insure the whole project? The failure rate for launches can’t be more than 10% or so; this is almost within the noise for the budget for a big project.)

  4. telescoper Says:

    I’m not an expert but I think these missions are not insured because the premiums are far too high to contemplate, probably because there are very few companies willing to underwrite them. I’d be interested to hear from anyone who knows more about it.

  5. Anton Garrett Says:

    I’m sure the cost of insurance is why not. The premium, based on rocket reliability, would be a significant proportion of the cost of making the instrument, and it would have to get past the sponsors each time.

    Let’s hope that (1) they make it into orbit safely, and (2) on the way they don’t bump into the Shuttle that is currently repairing and upgrading the Hubble space telescope. This is an important week for astronomy.

    Where is the live internet feed of the launch?


  6. Actually, WHATEVER the costs are and how they are distributed, and WHATEVER the reliability rate of rockets is, the cost per SUCCESSFUL mission would be about the same with or without insurance but without leaving those out in the cold who had the bad luck to have their mission fail. The only caveat is that this holds only as long as the underwriter’s PROFIT is not a substantial fraction of the entire premium.

    Of course, this only makes sense if one considers all missions together. For an individual mission, there is the balance between risk and cost. Of course, the whole idea of insurance is that all pay a premium and only those with bad luck collect. A government funding agency could take out insurance for about the same cost per successful mission and give folks the assurance that even if the rocket blows up their mission is not lost.

    Of course, individual project managers might want to opt for no insurance and higher risk. If the mission fails, it’s mostly students and postdocs who suffer most, since those making the decisions usually have permanent

    In Germany, large companies sometimes don’t have to have certain types of insurance, since due to shear size their accident rate will be about the average accident rate and by paying for accidents as they come they actually save money compared to insurance since no insurance company makes a profit. Similarly, a government funding agency could allocate, say, 10% less to all projects but re-fund them if the launch fails. On the other hand, such a decision would be made years after the original funding decision, and it might very well be that there are better ways to spend the money NOW. To have the best of both worlds, perhaps there should be some fast-track funding scheme for relaunching failed missions. For example, if Planck and Herschel crash, the best scientific case might be made for re-funding them, instead of funding some new project, especially since there is a new situation after the crash and the re-funding costs might be substantially less than the original costs.

    I think someone interested in psychology and game theory could have a field day with this.

  7. telescoper Says:

    You’re assuming that there are underwriters willing to take on the risk for a premium that’s less than prohibitive.

  8. Anton Garrett Says:

    Philip: I’m sure that insurance companies (if nobody else) have done detailed mathematical analyses of this sort. Such considerations were significant drivers of probability theory in the era after it was first quantified by Pascal and Fermat. I remember asking similar questions after an ordnance depot fire, when an MP asked if the government should have insurance against such events.

  9. It would be interesting to know how high the premium would actually be and what fraction of that is profit. Much big science is funded badly, cutting corners in the wrong places (remember the bad optics HST once had?).
    Maybe insurance would actually be affordable in some sense but most funding agencies opt for the “live fast, die young” risky approach.

    Leaving out underwriters for the moment—assume Planck is lost during the launch. We now have a new situation: no Planck. Should funding agencies fund a completely new project, or re-fund Planck, presumably for a fraction of the original cost? Presumably the decision should be based on science per money spent, which should be higher for a re-funded project than for a completely new one, everything else being equal.

  10. Hi Philip – what if all ten missions were funded by the same agency?

    We’re having a bit of a knees-up at Sussex too, hopefully linked up to the ESA live satellite feed, but apparently we get to pop open the champagne only “if all goes well”…

  11. “Hi Philip – what if all ten missions were funded by the same agency?”

    Suppose 1 mission in 10 fails. Say a mission costs a few hundred*. Whether I accept that one mission fails or re-fund it, to first order the cost per SUCCESSFUL mission is the same. (In the case of refunding, the total cost is more, but there is one more SUCCESSFUL mission.)

  12. “Hi Philip – what if all ten missions were funded by the same agency?”

    Suppose 1 mission in 10 fails. Say a mission costs a few hundred*. Whether I accept that one mission fails or re-fund it, to first order the cost per SUCCESSFUL mission is the same. (In the case of refunding, the total cost is more, but there is one more SUCCESSFUL mission.)

    * Reporter from MOJO magazine: How much are you worth?
    Gene Simmons: It’s a few hundred.
    Reporter: Million dollars?
    Simmons: What else?

  13. […] my paper to the journal and the ArXiv before the little shindig we’ve been planning for the Planck launch gets under way at 1pm. Business as usual so […]

  14. Anton Garrett Says:

    You can always use bookmakers as insurance agents…

  15. Pippa Goldschmidt Says:

    hi Peter
    Anything that is launched into outer space, and which the UK is involved with, should be insured – but this is a legal requirement to cover the health and safety aspects of the relevant risks (eg in case a bit of satellite falls onto someone’s head and they sue the UK Govt). Under UN treaties, the UK has liability for any damage caused in this way. See for fascinating info about the The Outer Space Act.
    How do I know this? When I left Imperial I joined the civil service and became responsible for this regulation …

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: