Beam-powered propulsion - keeping the beam focused

[mfb]

There is a technical article somewhere, but it is hard to find.

The data rate would be extremely low, but a single bit once in a while can be sufficient to get a picture over time.

 

[sophiecentaur]

over time.

For a hi res picture, you could end up having to wait as long as the actual journey took. But that would be planned and a decision made as to whether it's worth it.
There will be a finite beam width achievable and it strikes me that a vehicle with a sail that's of a comparable size, you get the best value. Why would they use such small craft, rather than a craft that's thousands of metres across?

 

[mfb]

A larger spacecraft will need more powerful lasers, and it is harder to align the spacecraft properly. Sending more smaller spacecraft is easier.

Even with a single bit per hour you get an interesting picture over 1 year (10,000 hours).

 

[sophiecentaur]

Sending more smaller spacecraft is easier.

Why? The same power is needed for one big one or many small ones (it's total mass in the end.)
If the many small craft could get close together at the end, they could make a larger transmitting system.

Even with a single bit per hour you get an interesting picture over 1 year (10,000 hours).

There we are then. To get a number of worthwhile images would take something like the total journey time. I think your 1bit per hour was a fair enough guess.

 

[mfb]

Why? The same power is needed for one big one or many small ones (it's total mass in the end.)

You can distribute the energy over a longer time, which means you need a lower peak power: A smaller amount of lasers and telescopes.

There we are then. To get a number of worthwhile images would take something like the total journey time. I think your 1bit per hour was a fair enough guess.

1 year is significantly shorter than the journey time.

 

[sophiecentaur]

1 year is significantly shorter than the journey time.

But you would want several images, wouldn't you?

You can distribute the energy over a longer time, which means you need a lower peak power: A smaller amount of lasers and telescopes.

When you put it that way, it's true but where is the problem with operating the beam for many days? (Particularly if the beam is produced out in space.) It strikes me that they want to do it that way and are making a virtue out of necessity.

 

[schplade]

Aiming telescopes is necessary both in space and on the ground, and it is easier on the ground (where you have a solid surface as reference). It is not an issue at all, every telescope does that routinely for operation. Forget spacecraft steering - the telescope has to aim. You need several laser "telescopes" anyway: Even a magic spacecraft with perfect steering from nowhere would still need multiple telescope beams that aim at the same spot.

Obviously you would need to aim the laser at the spacecraft to get things started, but from that point forward, couldn't the spacecraft make adjustments of its own to stay on the beam? It should be relatively easy for the spacecraft to detect when it's moving off the center of the beam. Perhaps it could use thrusters, or even adjust the sail somehow, to stay centered? It would need to make corrections fast, for sure. I'm not saying it would be easy by any means, but wafer-sized space probes present massive challenges too, especially when it comes to communication.

 

[GTOM]

Thanks for your information mfb.

Otherwise I'm skeptical about that grams weight probes idea. How their delicate electronics supposed to withstand cosmic rays?

A GW laser focused by 100m Overly Large Telescope. Any price estimation for a GW laser? I guess it has to produced by a really large facility.

 

[sophiecentaur]

Obviously you would need to aim the laser at the spacecraft to get things started, but from that point forward, couldn't the spacecraft make adjustments of its own to stay on the beam?

This worries me a lot. If the craft goes off course (due to some asymmetry in its construction or even just some dirt on one side, firstly it would be hard / impossible to see the tiny craft and several years(?) delay in the control loop would mean that the beam couldn't correct for it. Could you really work 'open loop' for twenty years?

You can distribute the energy over a longer time, which means you need a lower peak power: A smaller amount of lasers and telescopes.

I think I must be working with a model in my head that's not what's been published but the journalese has probably got in the way of clarity.
Is it assumed that the whole of the beam power would hit one tiny craft? How long would the beam be turned on for? You seemed, at one point, to suggest that the boost would be given in just one day. Radiation pressure needs to act for a long time to transfer any significant momentum which means months / years worth of drive.

 

[mfb]

But you would want several images, wouldn't you?

The 1 bit/hour was a random number, don't treat it like a proper estimate. It will depend on the spacecraft design, the available receivers and so on.

When you put it that way, it's true but where is the problem with operating the beam for many days? (Particularly if the beam is produced out in space.) It strikes me that they want to do it that way and are making a virtue out of necessity.

Acceleration goes down with increasing distance of the spacecraft as your beam spreads out over time. A longer, slower acceleration leads to a lower final velocity. Even if you can point the beam at the spacecraft for days (which would need space-based lasers, or way too many ground-based ones), it is not a practical solution to achieve high speeds. It would also reduce the launch rate a lot.

Obviously you would need to aim the laser at the spacecraft to get things started, but from that point forward, couldn't the spacecraft make adjustments of its own to stay on the beam? It should be relatively easy for the spacecraft to detect when it's moving off the center of the beam. Perhaps it could use thrusters, or even adjust the sail somehow, to stay centered? It would need to make corrections fast, for sure. I'm not saying it would be easy by any means, but wafer-sized space probes present massive challenges too, especially when it comes to communication.

The smallest current systems that can stabilize itself with thrusters are of the order of a kilogram. A gram-sized spacecraft , where most of the mass is the sail? No way.
It is trivial to adjust the direction of telescopes. It has to be done anyway. There is no point in adding complexity to the spacecraft that doesn't save anything elsewhere.

How their delicate electronics supposed to withstand cosmic rays?

There are electronic components with a very high radiation tolerance (>1 gigarad). Something that can survive the conditions in the LHC detectors can also survive interstellar space. It won't survive collisions with dust particles, but those are rare, and many probes can be sent to have some of them surviving.

A GW laser focused by 100m Overly Large Telescope. Any price estimation for a GW laser? I guess it has to produced by a really large facility.

We don't have such a laser yet, hard to estimate.

This worries me a lot. If the craft goes off course (due to some asymmetry in its construction or even just some dirt on one side, firstly it would be hard / impossible to see the tiny craft and several years(?) delay in the control loop would mean that the beam couldn't correct for it. Could you really work 'open loop' for twenty years?

You can easily see it during the acceleration phase (you illuminate it with a very bright laser...) and steer it with the beam, afterwards there is no way to steer it - it will just fly in a straight line.

I think I must be working with a model in my head that's not what's been published but the journalese has probably got in the way of clarity.
Is it assumed that the whole of the beam power would hit one tiny craft? How long would the beam be turned on for? You seemed, at one point, to suggest that the boost would be given in just one day. Radiation pressure needs to act for a long time to transfer any significant momentum which means months / years worth of drive.

The Breakthrough program uses 20 minutes of acceleration time as baseline. You cannot drive it for months or even years, there is no way to give it any relevant acceleration over the corresponding large distances. OWL can keep a beam nicely focused over 20 million kilometers, after that the intensity drops with 1/r². 20 minutes acceleration to 0.2 c lead to a final distance of 36 million km - that works. 1 day of acceleration would lead to a final distance of hundreds of millions of kilometers - that does not work. Months, years? No way.

 

[schplade]

The smallest current systems that can stabilize itself with thrusters are of the order of a kilogram. A gram-sized spacecraft , where most of the mass is the sail? No way.
It is trivial to adjust the direction of telescopes. It has to be done anyway. There is no point in adding complexity to the spacecraft that doesn't save anything elsewhere.

You're still thinking in terms of wafer-sized spacecraft . I was thinking of something that's more comparable in mass to our current space probes, pushed by an extremely powerful laser. Something that would take a long time to accelerate. We obviously can't aim a laser to hit a departing object that's several light-minutes away, unless we can use an amazingly accurate predictive algorithm. So I'm wondering if it's possible to build a sailcraft that could make its own adjustments and keep itself centered on a beam, riding that beam deep into space. If not, then blasting gram-sized spacecraft up to their top speed at 50,000 G's is probably the best we can ever do.

 

[sophiecentaur]

You can easily see it during the acceleration phase (you illuminate it with a very bright laser...) and steer it with the beam, afterwards there is no way to steer it - it will just fly in a straight line.

OK, that's clear now. But, unless there's some restriction on running time of the lasers, why not use them for a lot longer (even in daily doses) and give the same impulse to a much bigger craft? It would not be hard to pick up where you left off overnight. What a waste of a massive light source, if it only operates for 20 minutes per vehicle. Is there another possible use for it when it's not being used for propulsion?
I haven't grasped the advantage of using small craft rather than something 100 (10,000) times bigger. It sounds like standing up in a hammock to me.

 

[mfb]

You're still thinking in terms of wafer-sized spacecraft . I was thinking of something that's more comparable in mass to our current space probes, pushed by an extremely powerful laser. Something that would take a long time to accelerate. We obviously can't aim a laser to hit a departing object that's several light-minutes away, unless we can use an amazingly accurate predictive algorithm. So I'm wondering if it's possible to build a sailcraft that could make its own adjustments and keep itself centered on a beam, riding that beam deep into space. If not, then blasting gram-sized spacecraft up to their top speed at 50,000 G's is probably the best we can ever do.

We can easily aim at it. Aiming does not get notably harder with distance, as you are diffraction-limited anyway: The necessary angular accuracy does not increase. And it is within the capabilities of current telescopes.

OK, that's clear now. But, unless there's some restriction on running time of the lasers, why not use them for a lot longer (even in daily doses) and give the same impulse to a much bigger craft? It would not be hard to pick up where you left off overnight. What a waste of a massive light source, if it only operates for 20 minutes per vehicle. Is there another possible use for it when it's not being used for propulsion?

See the previous post. At the end of the 20 minutes the efficiency is down to ~1/4 the maximal value already, after an hour it would be less than 1%, after 10 hours it is below 0.01%. Yes you can keep shooting the laser at it, but the effect becomes completely negligible.

In terms of using the lasers: The project proposes to send thousands of small spacecraft , that takes a good fraction of a year, and you can target more than one star. You can burn space debris with the lasers. And you can use the telescopes as telescopes.

 

[GTOM]

We can easily aim at it. Aiming does not get notably harder with distance, as you are diffraction-limited anyway: The necessary angular accuracy does not increase. And it is within the capabilities of current telescopes.See the previous post. At the end of the 20 minutes the efficiency is down to ~1/4 the maximal value already, after an hour it would be less than 1%, after 10 hours it is below 0.01%. Yes you can keep shooting the laser at it, but the effect becomes completely negligible.

Sophiecentaur talked about a much bigger craft, could that laser be much more useful if pushing a ship with a km sail?
(It could deliver cargo inside solar system, or eject lots of tiny probes, and hope they withstand conditions better than the ones sent to Fukushima.)

 

[mfb]

A 1 square kilometer sail will keep the full efficiency longer - something like 200 million kilometers. At the same target velocity you can decrease acceleration by a factor 10 and increase time by the same factor.

Delivering cargo within the solar system is much easier as the velocities are orders of magnitude lower. You could probably just use sunlight with such a light sail.

 

[sophiecentaur]

At the end of the 20 minutes the efficiency is down to ~1/4 the maximal value already,

That will be because of spillage, I guess. The area of the nano sails will be a small part of the beam area so why not actually use an area that's equal to the beam area at the start (or even further out along the beam?). I seems that this project all hangs on an explosive (almost) start to the trip with nano sails, which is not what's usually expected with light sails. It is a bit hard to get rid of preconceived ideas about this method of propulsion. The acceleration involved must be phenomenal (around 10⁵ms^-2) to get to a speed approaching light speed in just 1200s. The whole idea seems to require some seriously 'adventurous' concepts, which are highly counter intuitive.

 

[sophiecentaur]

A 1 square kilometer sail will keep the full efficiency longer - something like 200 million kilometers. At the same target velocity you can decrease acceleration by a factor 10 and increase time by the same factor.

. . . . and increase the payload to something that approaches a credible value.

 

[mfb]

You need the high acceleration. If we fix the target speed and don't want to build telescopes beyond 100 meters in diameter (and tricky adaptive optics), the main acceleration process cannot take longer than 20 minutes. You can keep shooting the laser at it later, but it won't change the overall speed notably. Could be interesting for steering, but for the discussion here we can neglect it.

A smaller sail leads to a higher acceleration. The 1/4 is actually not very accurate, because it assumes a 100 meter beam diameter until the envelope has to increase. We can start with a much smaller cross section and therefore a higher acceleration value. But that only works with a smaller sail.

. . . . and increase the payload to something that approaches a credible value.

Only if you increase beam power by a similar factor.

 

[mheslep]

...What a waste of a massive light source, if it only operates for 20 minutes per vehicle. Is there another possible use for it when it's not being used for propulsion?

A similar comment could be made about the current chemical boosters and their mega-Newton thrust. The beam would necessarily be highly inefficient. I don't think it would valuable to use for anything except a launch (assuming viability, and I don't see a path to viability)

 

[GTOM]

A similar comment could be made about the current chemical boosters and their mega-Newton thrust. The beam would necessarily be highly inefficient. I don't think it would valuable to use for anything except a launch (assuming viability, and I don't see a path to viability)

It might could be used to power a lunar colony, without take a nuclear reactor there.

 

[mheslep]

It might could be used to power a lunar colony, without take a nuclear reactor there.

No, like chemical boosters, the beam system would be necessarily be designed to deliver very high bursts of power for a few minutes, not supply continuous power.

 

[GTOM]

No, like chemical boosters, the beam system would be necessarily be designed to deliver very high bursts of power for a few minutes, not supply continuous power.

Valid point. Still if it could power Earth-Moon vessels for example, than it could be frequently used.

 

[sophiecentaur]

A smaller sail leads to a higher acceleration.

I puzzled over this but I have to assume that you are assuming the proportions would need to be the same so area : mass = l²:l³
But why would that be necessary? You see, I cannot see how you can ever get over the spillage with a sail of just a few cm³ area. Surely a sail area that's comparable with the beam area has to get most momentum transfer ( = acceleration). The only reason that small sail = higher acceleration might be that mechanics don't scale as you'd think and small structures are stronger for a given shape. But the pressure over a large sail would / could be fairly uniform so stresses would still be low. This hasn't been justified in anything I have read.

 

[sophiecentaur]

A similar comment could be made about the current chemical boosters and their mega-Newton thrust.

A different comparison. The cost of a rocket is much less than a once in a lifetime telescope. The 30m telescope has a budget of £$1200M so bigger would be considerably more (third power law perhaps) and a rocket launch is upward of £$100M (figures trawled from all over) If the telescope were doing this job in its spare time then things would be different as it's highly re-usable.

 

[mfb]

I puzzled over this but I have to assume that you are assuming the proportions would need to be the same so area : mass = l²:l³
But why would that be necessary? You see, I cannot see how you can ever get over the spillage with a sail of just a few cm³ area. Surely a sail area that's comparable with the beam area has to get most momentum transfer ( = acceleration). The only reason that small sail = higher acceleration might be that mechanics don't scale as you'd think and small structures are stronger for a given shape. But the pressure over a large sail would / could be fairly uniform so stresses would still be low. This hasn't been justified in anything I have read.

A smaller sail means you can start with a more focused beam - a higher intensity. At the same mass per area ratio (=most of the mass is the sail), this leads to a higher acceleration.
You probably want the beam to be very uniform over the sail area to have a uniform acceleration. Which means the sail has to be smaller than the total beam width.

The cost of a rocket is much less than a once in a lifetime telescope.

It is not a once in a lifetime telescope if you use it for thousands of small spacecraft , and use it as regular telescope.

 

[schplade]

We can easily aim at it. Aiming does not get notably harder with distance, as you are diffraction-limited anyway: The necessary angular accuracy does not increase. And it is within the capabilities of current telescopes.

I might be misunderstanding something still. Let's say the spacecraft is 5 light-minutes away from the telescope/laser. You might be able to see it, but you won't see it at its current location. You'll see where it was 5 minutes ago, right? You can't know with certainty what its current location is, so if you want to aim a laser at it, you have to predict where it will be with great accuracy. Maybe it's possible to do that; it just seems outrageously difficult.

 

[GTOM]

I might be misunderstanding something still. Let's say the spacecraft is 5 light-minutes away from the telescope/laser. You might be able to see it, but you won't see it at its current location. You'll see where it was 5 minutes ago, right? You can't know with certainty what its current location is, so if you want to aim a laser at it, you have to predict where it will be with great accuracy. Maybe it's possible to do that; it just seems outrageously difficult.

If it don't make fast random movements, aiming will be still pretty accurate.

 

[sophiecentaur]

It is not a once in a lifetime telescope if you use it for thousands of small spacecraft , and use it as regular telescope.

Which is why I asked, further up, about other uses for such a telescope. Unfortunately, the word "telescope" is not really accurate because the transmitter would not be a good receiving system without serious modifications. I can't imagine it could be operated with a simple TX/RX switch. The use as a cargo propellor to the Moon would make sense, though.
But I still can't see the advantage of using such small craft when the beam could not be focussed on just one sail. Why not, at least, fill the beam with a single sail? One single craft would be much more use than many smaller ones (there can be no argument about that, surely). I can understand how a short sharp burst of acceleration makes sense, due to beam spreading at a great distance but spillage around a small sail would mean a huge wastage of hard-won propulsive energy. The Numbers Game is very relevant here and I haven't seen any actual numbers to dispel my intuitive objections. The project could be based on "We want to do it this way", and it wouldn't be the first such project.
Comms would be needlessly complicated in a case like that. Each craft would need its own channel and a dedicated receiver because of the necessary low bandwidth and the very long duration of a useful signal.

 

[sophiecentaur]

If it don't make fast random movements, aiming will be still pretty accurate.

How can you be sure that there would be no perturbations from, as yet unknown, sources. Just because the star appears at a fixed place in the sky, doesn't tell you about mechanical forces that could be operating way out there and a very tiny disturbance (a large, undiscovered Kuyper Belt object coming close, for instance) could deflect the craft. Open loop (ballistic) operation is a very risky way to go, imho. We have absolutely no way of being sure about this.

 

[mfb]

@sophiecentaur: Please read the proposal of the Breakthrough Starshot project. You keep proposing things that are completely impractical, things discussed in detail (why they don't work) by the Breakthrough project.

doesn't tell you about mechanical forces that could be operating way out there and a very tiny disturbance (a large, undiscovered Kuyper Belt object coming close, for instance) could deflect the craft

The spacecraft is fast. A near miss at Pluto would alter the velocity by something like 5 cm/s, or 30,000 km over 20 years. Completely negligible.
The probability of such a near miss (or impact) for a random object at a random location in the sky at the distance of Pluto is less than 10^-14. It is even smaller for objects further out. Negligible as well.
You could check the numbers before you invent issues that do not exist.

I might be misunderstanding something still. Let's say the spacecraft is 5 light-minutes away from the telescope/laser. You might be able to see it, but you won't see it at its current location. You'll see where it was 5 minutes ago, right? You can't know with certainty what its current location is, so if you want to aim a laser at it, you have to predict where it will be with great accuracy. Maybe it's possible to do that; it just seems outrageously difficult.

At the time it is 5 light minutes away, the laser won't be focused well any more and the acceleration (if present at all) is negligible.
The spacecraft has to have some control over its orientation in order to transmit data back to Earth once it reaches its destination, but that has to work with regular sunlight only, not in the intense acceleration phase.

 

[sophiecentaur]

@sophiecentaur: Please read the proposal of the Breakthrough Starshot project. You keep proposing things that are completely impractical, things discussed in detail (why they don't work) by the Breakthrough project.

It might have been helpful to post a link to the document if you have found it already. Pretty much all I have found is Pop Science with generalisations. I have seen an animation which shows a big array of Hollywood / James Bond style microwave dishes, which rather put me off reading much further. There are numbers quoted, such as that the nano craft could get to Mars in 2m. What could it do when it got there, apart from to keep on going. Likewise for Pluto etc. That sort of popular blurb proves nothing either way but it worries me that the drivers of the project are happy for it to go out with the publicity material. I guess they want backers and don't expect many potential backers to have much of an idea about the practicalities.
Meanwhile, if you could let me have a link to the large amount of information you are able to draw on, I should be very grateful.

You could check the numbers

Likewise, I would be grateful for some basics for working that out without starting from scratch. Or you could just show your workings?

 

[mfb]

I found it when the announcement was new, and read it, but didn't save the link. I don't think it is my task to find it again.

I guess they want backers and don't expect many potential backers to have much of an idea about the practicalities.

They have enough funding from billionaires to study the concept in more detail.

Going to Mars in 2 minutes (?) won't work, not even light is that fast.

Likewise, I would be grateful for some basics for working that out without starting from scratch. Or you could just show your workings?

Standard orbital mechanics, with some rough approximations. Pluto has a surface gravitational acceleration of 0.6 m/s² and a radius of 1200 km. The spacecraft will fly in a nearly straight line. As conservative estimate (leading to a larger deflection), assume the surface gravitational acceleration acts as perpendicular force over 5000 km of the spacecraft trajectory. At 20% the speed of light, or 60,000 km/s, that gives an acceleration over 1/12 s, for a speed change of 5 cm/s. Over 20 years this leads to a deflection of 30000 km.

The fraction of solid angle Pluto occupies in the sky is pi (radius of pluto)^2 / (4 pi (distance to pluto)^2).

Basic high school physics.

 

[sophiecentaur]

I don't think it is my task to find it again.

You told me I should read it so it is the least you could do for me.

Going to Mars in 2 minutes (?) won't work, not even light is that fast.

You are right - but I got it from one of those easily losable Google hits about eat topic.

for a speed change of 5 cm/s. Over 20 years this leads to a deflection of 30000 km.

Thanks.
I was also considering gas or dust clouds but I guess, if the star group is easily visible, there can't be much in the way.

 

[GTOM]

Too bad i can't found the comments of a phys.org article, but there was a discussion about the focusation problems of high-powered lasers, due to amplification process (multi mode lasers) or heat stress of lenses/mirrors or something like that.
Although in that case the laser equipment dotn have to be compact unlike a militaty application, so there can be a forest of single mode low powered lasers.