Potential servicing missions for the Webb

[su27k]

I'll just say this: If god forbid something happened during deployment, or the telescope runs out of fuel early than expected, you can bet NASA will suddenly come up with a way to save it by servicing, all the "But it can't do it because xyz" will magically disappear, and everybody who objected to servicing in this thread will pretend it was never an issue.

[Eric Hedman]

I'll just say this: If god forbid something happened during deployment, or the telescope runs out of fuel early than expected, you can bet NASA will suddenly come up with a way to save it by servicing, all the "But it can't do it because xyz" will magically disappear, and everybody who objected to servicing in this thread will pretend it was never an issue.

Do you have any idea of how long it would take to design a mission and a spacecraft to do a repair if it was possible?  You would need to build and test the hardware.  By the time you had it ready to launch you're probably talking 7 to 10 years if not more.  It would be a better use of money to work on the next telescope.

[DanClemmensen]

I'll just say this: If god forbid something happened during deployment, or the telescope runs out of fuel early than expected, you can bet NASA will suddenly come up with a way to save it by servicing, all the "But it can't do it because xyz" will magically disappear, and everybody who objected to servicing in this thread will pretend it was never an issue.

Do you have any idea of how long it would take to design a mission and a spacecraft to do a repair if it was possible?  You would need to build and test the hardware.  By the time you had it ready to launch you're probably talking 7 to 10 years if not more.  It would be a better use of money to work on the next telescope.

STS-49 salvaged Intelsat 603, which had failed to separate from the second stage of its launcher and was left in LEO. It took two years to plan the mission and build the needed hardware.
  https://en.wikipedia.org/wiki/STS-49
Hubble was launched with a defective optical system. While Hubble had been designed to be serviced by the shuttle, a fix for this defect was not an anticipated repair, so a repair mission had to be designed and implemented.
  https://en.wikipedia.org/wiki/Hubble_Space_Telescope#Flawed_mirror

Because Starship launches will be inexpensive and because Starship can reach any orbit, repairs will be more feasible in the future.

[Eric Hedman]

I'll just say this: If god forbid something happened during deployment, or the telescope runs out of fuel early than expected, you can bet NASA will suddenly come up with a way to save it by servicing, all the "But it can't do it because xyz" will magically disappear, and everybody who objected to servicing in this thread will pretend it was never an issue.

Do you have any idea of how long it would take to design a mission and a spacecraft to do a repair if it was possible?  You would need to build and test the hardware.  By the time you had it ready to launch you're probably talking 7 to 10 years if not more.  It would be a better use of money to work on the next telescope.

STS-49 salvaged Intelsat 603, which had failed to separate from the second stage of its launcher and was left in LEO. It took two years to plan the mission and build the needed hardware.
  https://en.wikipedia.org/wiki/STS-49
Hubble was launched with a defective optical system. While Hubble had been designed to be serviced by the shuttle, a fix for this defect was not an anticipated repair, so a repair mission had to be designed and implemented.
  https://en.wikipedia.org/wiki/Hubble_Space_Telescope#Flawed_mirror

Because Starship launches will be inexpensive and because Starship can reach any orbit, repairs will be more feasible in the future.

These were in LEO, not Sun Earth L2.  These were easily reachable with humans on the Shuttle.  There are thousands of possible failure modes for James Webb.  Most would more than likely require a human in the loop.  No space vehicle, not Orion and not Starship are going to go and do a mission anytime soon.  Orion is good for 21 days.  It takes a hell of lot longer than 21 days to get were the Webb telescope is going. The chance of being able to design and implement a successful unmanned mission is extremely low for most failure scenarios.  If the James Webb Telescope fails to deploy, it will be lost.

[DanClemmensen]

I'll just say this: If god forbid something happened during deployment, or the telescope runs out of fuel early than expected, you can bet NASA will suddenly come up with a way to save it by servicing, all the "But it can't do it because xyz" will magically disappear, and everybody who objected to servicing in this thread will pretend it was never an issue.

Do you have any idea of how long it would take to design a mission and a spacecraft to do a repair if it was possible?  You would need to build and test the hardware.  By the time you had it ready to launch you're probably talking 7 to 10 years if not more.  It would be a better use of money to work on the next telescope.

STS-49 salvaged Intelsat 603, which had failed to separate from the second stage of its launcher and was left in LEO. It took two years to plan the mission and build the needed hardware.
  https://en.wikipedia.org/wiki/STS-49
Hubble was launched with a defective optical system. While Hubble had been designed to be serviced by the shuttle, a fix for this defect was not an anticipated repair, so a repair mission had to be designed and implemented.
  https://en.wikipedia.org/wiki/Hubble_Space_Telescope#Flawed_mirror

Because Starship launches will be inexpensive and because Starship can reach any orbit, repairs will be more feasible in the future.

These were in LEO, not Sun Earth L2.  These were easily reachable with humans on the Shuttle.  There are thousands of possible failure modes for James Webb.  Most would more than likely require a human in the loop.  No space vehicle, not Orion and not Starship are going to go and do a mission anytime soon.  Orion is good for 21 days.  It takes a hell of lot longer than 21 days to get were the Webb telescope is going. The chance of being able to design and implement a successful unmanned mission is extremely low for most failure scenarios.  If the James Webb Telescope fails to deploy, it will be lost.

I agree that Orion cannot do this for both capability and cost reasons.

As I said: Starship will be able reach any orbit. It is not restricted to LEO. It can support a long-duration crewed mission: it is designed to carry crew to Mars and back. It  can carry as much repair equipment as needed. Intelsat 603 stayed in LEO unused and unusable for two years while the repair mission was negotiated and planned.

A crewed repair mission to Earth-sun L2 will be very expensive, but not as expensive as replacing JWST.  The trip would not happen before 2024, so I suspect it would include replenishment of consumables (helium and xenon?) and of course there is no guarantee of success.

[Lee Jay]

As I said: Starship will be able reach any orbit.

Its booster hasn't flown yet. Its spacecraft hasn't come close to being equipped with the needs of a crew yet. It may be so large that its propulsion will demolish the sun shield. It has no airlock. It has no arm. JWST doesn't have a docking or attachment point. Propulsion from Starship would likely contaminate JWST.

Do you have rational, economic, near-term solutions to those issues?

[DanClemmensen]

As I said: Starship will be able reach any orbit.

Its booster hasn't flown yet. Its spacecraft hasn't come close to being equipped with the needs of a crew yet. It may be so large that its propulsion will demolish the sun shield. It has no airlock. It has no arm. JWST doesn't have a docking or attachment point. Propulsion from Starship would likely contaminate JWST.

Do you have rational, economic, near-term solutions to those issues?

Nope. I'm not a space professional. I must trust that it will fly (NASA believes this: see the HLS contract), that it will be equipped for  crew (dearMoon, HLS, Elon's Mars plans), that it will have an airlock (HLS, Elon's Mars plans). I must also trust that a two-year effort by a rescue planning team (like the Intelsat 603 team) can solve the remaining problems if given, say, a $3 billion budget. Intelsat 603 had no docking or attachment point.

After about 5 minutes as a non-professional responding to a post on the Internet, my mission plan would be:
*arrive at a point 1000 meters away from target at 0 relative velocity using normal maneuvering thrusters and engines, with the trajectory computed to keep engines and thrusters pointed away from the target.
*use purpose-built inefficient wide-dispersion cold gas thrusters to maneuver to close proximity (10 meters?) at 0 relative velocity, taking 48 hours if needed.
*during multiple EVAs, crew will build purpose-designed scaffolding to connect the Starship to the JWST and act as the work platform.
*during further EVAs, crew undertakes repairs
*during further EVAs, crew detaches and retrieves scaffolding
*use the cold gas thrusters to very slowly move to 1000 meters
*go home on a trajectory that keeps the thrusters and engines pointed away from the target.


Do you believe that the US aerospace industry is incapable of planning and implementing this mission or planning a better one?

[M.E.T.]

As I said: Starship will be able reach any orbit.

Its booster hasn't flown yet. Its spacecraft hasn't come close to being equipped with the needs of a crew yet. It may be so large that its propulsion will demolish the sun shield. It has no airlock. It has no arm. JWST doesn't have a docking or attachment point. Propulsion from Starship would likely contaminate JWST.

Do you have rational, economic, near-term solutions to those issues?

Not commenting on the general viability of this plan, but I have to laugh at some of the specific objections raised to Starship in the above post.

No booster?
Cannot support crew?
No airlock?

That’s just lazy arguing. NASA’s entire Artemis program relies on the above three requirements as a given - among many other criteria.

Good time for a Joe Biden-style exasperated “Come on man!”

[Eric Hedman]

These were in LEO, not Sun Earth L2.  These were easily reachable with humans on the Shuttle.  There are thousands of possible failure modes for James Webb.  Most would more than likely require a human in the loop.  No space vehicle, not Orion and not Starship are going to go and do a mission anytime soon.  Orion is good for 21 days.  It takes a hell of lot longer than 21 days to get were the Webb telescope is going. The chance of being able to design and implement a successful unmanned mission is extremely low for most failure scenarios.  If the James Webb Telescope fails to deploy, it will be lost.

I agree that Orion cannot do this for both capability and cost reasons.

As I said: Starship will be able reach any orbit. It is not restricted to LEO. It can support a long-duration crewed mission: it is designed to carry crew to Mars and back. It  can carry as much repair equipment as needed. Intelsat 603 stayed in LEO unused and unusable for two years while the repair mission was negotiated and planned.

A crewed repair mission to Earth-sun L2 will be very expensive, but not as expensive as replacing JWST.  The trip would not happen before 2024, so I suspect it would include replenishment of consumables (helium and xenon?) and of course there is no guarantee of success.

NASA would be crazy to spend money on a mission using Starship until it is proven with refueling and return from high speed trajectories.  It also needs to prove that it's life support system is reliable for a many months long mission out to L2 and back.  In addition, there also may be no way to fully assess the problems with the Webb before they get there.  What if they come out to fix one problem and discover it isn't the only one stopping the telescope from working?  I think it would be highly unlikely any repair mission would be mounted.  It would be better to concentrate on the next telescope.

[Paul451]

NASA would be crazy to spend money on a mission using Starship until it is proven with refueling and return from high speed trajectories.

And yet that's exactly what they are doing with Artemis. Crazy!

[Edit: That said, potential repair missions on that scale seem off-topic for this thread. Even discussing a MEV life-extension seems to be pushing the edges too far. Creating a new topic for "JWST life-extension and repair mission concepts and speculation" is just a matter of clicking a button.]

[LouScheffer]

This is a thread about potential servicing missions for the Webb telescope.

Perhaps the mods can pick out the serving mission posts and move them here, after which I'll delete this sentence.

The arguments for are:
  * Webb is quite expensive, and has unique capabilities
  * L2 orbit maintenance and momentum dumping require fuel, but there are no other consumables.  So an MEV-1 type addition (taking over attitude control only) might work.
  * Extending life might be cheaper than a new mission

The arguments against are:
  * Webb is not designed for servicing of any kind.  No place to attach, add fuel, etc.
  * A servicing mission would be too expensive for the benefit gained
  * It would take too long to develop
  * A servicing mission would damage Webb (hurt sunscreen, contaminate optics, etc.)
  * We should spend the money on a new mission, not fixing an old one

And here is a spot for Webb design documents that might shed light on the issues:  More can be added as found...
   * The paper "Status of the JWST sunshield and spacecraft" shows that the fill and drain valves are exposed. See Figure attached.
   * ...

[etudiant]

This seems like a useful test case of the impact of much lower cost/lb orbital access offered by SH/SS.
Is it cheaper to do a custom replenishment mission or to just do a JW2 reflight, taking advantage of the vastly increased volume and lift offered to eliminate the expensive origami aspects of the original?

[LouScheffer]

Here are some possible cost comparisons:

The DARPA "Orbital Express" project developed a refueler spacecraft and a target spacecraft for about $215 million (DARPA budget documents have
 2002: $5M
 2003: $40M
 2004: $55M
 2005: $47M
 2006: $37M
 2007: $31M
Though other sources put the total at $300M.

MEV-1 (comsat life extension) charges $13M a year for services, and the spacecraft is expected to last 15 years.  So the cost must be less than $195M, including launch.

[Jim]

, taking advantage of the vastly increased volume and lift offered to eliminate the expensive origami aspects of the original?

Really doesn't change much.  Only eliminates the mirror wings.  Secondary mirror still would need a deployment.  And the sun shield would still require packaging and deployment.  And it would be basically a compete redesign.

And the "origami" wasn't the expensive part.

[Jim]

The sunshield makes JWST unapproachable.

.  This is only 100lb thrusters

[DanClemmensen]

These were in LEO, not Sun Earth L2.  These were easily reachable with humans on the Shuttle.  There are thousands of possible failure modes for James Webb.  Most would more than likely require a human in the loop.  No space vehicle, not Orion and not Starship are going to go and do a mission anytime soon.  Orion is good for 21 days.  It takes a hell of lot longer than 21 days to get were the Webb telescope is going. The chance of being able to design and implement a successful unmanned mission is extremely low for most failure scenarios.  If the James Webb Telescope fails to deploy, it will be lost.

I agree that Orion cannot do this for both capability and cost reasons.

As I said: Starship will be able reach any orbit. It is not restricted to LEO. It can support a long-duration crewed mission: it is designed to carry crew to Mars and back. It  can carry as much repair equipment as needed. Intelsat 603 stayed in LEO unused and unusable for two years while the repair mission was negotiated and planned.

A crewed repair mission to Earth-sun L2 will be very expensive, but not as expensive as replacing JWST.  The trip would not happen before 2024, so I suspect it would include replenishment of consumables (helium and xenon?) and of course there is no guarantee of success.

NASA would be crazy to spend money on a mission using Starship until it is proven with refueling and return from high speed trajectories.  It also needs to prove that it's life support system is reliable for a many months long mission out to L2 and back.  In addition, there also may be no way to fully assess the problems with the Webb before they get there.  What if they come out to fix one problem and discover it isn't the only one stopping the telescope from working?  I think it would be highly unlikely any repair mission would be mounted.  It would be better to concentrate on the next telescope.

(Replying by copying from the other thread. Apologies if this is not the correct way to do this)

SpaceX clearly already plans to develop and demonstrate refueling, return from high-speed trajectories, and long-term life support, so no new spending is needed there. But spending a small amount of money on planning prior to completion of these demonstrations makes a lot of sense: the plan would allow NASA to fire up the rescue project design and implementation quickly as soon as the last demonstration is complete. This is not "instead of" the next telescope. Instead, it is an attempt to salvage some value from the $9.6 billion already spent on JWST. Sure sunk costs are sunk (we spent $9.9 B, but the current value of an operational JWST might be only $3 B or whatever) and sure, the rescue mission might not work. But it might work, so spending $2 million on an initial concept plan might be money well spent. It would at least be able to identify obvious show-stoppers (no possible solution) or major challenges (might be solvable but need real, in-depth analysis.)

[su27k]

Some official responses wrt servicing:

Scientists, engineers push for servicing and assembly of future space observatories

While JWST is not designed for servicing — Grunsfeld said it might be possible, but risky, to do some kind of robotic refueling mission for the telescope about 10 years after launch

Zurbuchen Taking One Last Look At Jwst Servicing Compatiblity

At a meeting of the NASA Advisory Council (NAC) today, Zurbuchen was asked by NAC member Mike Gold if NASA was ensuring some degree of JWST compatibility for servicing.  Gold works for Maxar Technologies.  Its Space Systems Loral business unit is one of the leading companies in developing satellite servicing technologies.

<snip>

Satellite servicing advocates have pointed out for years that although systems do not exist today that could robotically service JWST at L2, they might in the future.  They urged NASA to at least put stickers on the telescope to allow a future robotic system to locate it.

Zurbuchen confirmed today that NASA already has taken some steps and now is looking to see if there is anything else that can be done.

So it doesn't look to me that servicing is completely ruled out by NASA.

[Hobbes-22]

The sunshield makes JWST unapproachable.

  This is only 100lb thrusters

It certainly makes the approach more difficult.
On the other hand:
- this was something they hadn't accounted for in this flyaround, and may have been the first time the phenomenon was seen.
- they managed to dock Apollo to Skylab several times without blowing the sunshield away
- JWST has several thruster sets operating in the vicinity of the sunshield. So running a thruster with its thrust axis parallel to the sunshield seems to be possible if designed correctly.

There are ways to reduce the chance of damaging the sunshield:
- approach slowly, to minimize thruster use near JWST. The servicing vehicle is unmanned, so no problem if it takes 3 months to dock.
- to slow down when flying toward JWST, use thrusters set at a large angle to the velocity vector and accept the cosine losses.
- use a telescoping docking mechanism to absorb the remaining approach speed
- use thrusters mounted on a long boom to put their exhaust cone outside of the sunshield

You'd need a purpose-built vehicle rather than a standard MEV. We spent $9B on JWST to get 10 years of observation time.  We've paid $500M per servicing mission to extend Hubble's life several times.

IOW there's a gap between difficult and impossible.

[LouScheffer]

As far as contaminating the optics, the sunshade should largely prevent this, as exhaust will travel in straight lines.  After all, Webb has thrusters on the warm side (though angled away from the sunshade.)

If this is not enough, hydrogen or helium cold-gas thrusters would work.  The JWST optics are at about 50K, and neither hydrogen or helium will condense at that temperature.

[woods170]

As far as contaminating the optics, the sunshade should largely prevent this, as exhaust will travel in straight lines.  After all, Webb has thrusters on the warm side (though angled away from the sunshade.)

If this is not enough, hydrogen or helium cold-gas thrusters would work.  The JWST optics are at about 50K, and neither hydrogen or helium will condense at that temperature.

It is not about contamination. It is about physically shredding or tearing one or more layers of the sunshield via interaction with the RCS thrusters of an approaching servicing spacecraft.