Solar sails, bits and pieces

[Enthalpy]

My dear visionary and megalomaniac human fellows,

you've probably heard about Solar sails, one attempt among many to exceed chemical rockets performance, these being inconveniently slow for hopping through our Solar system and beyond. Solar sails want to catch the light of our Sun to obtain a thrust which would indeed improve on rocket speed if the sail is big, the complete spacecraft light, and the push long enough.

 

http://en.wikipedia.org/wiki/Solar_sail
http://www.jspec.jaxa.jp/e/activity/ikaros.html
http://planetary.org/

 

As far from the Sun as Earth is, incoming light has a pressure of 4.5µPa and reflected light as much, so accelerating any significant spacecraft mass requires sail areas at least in the square hectometre range which, to keep their own mass low enough, must be few micrometres thin - the very reasons why we still don't use Solar sails as the main propulsion of every spacecraft.

 

Of course, I couldn't refrain from throwing a few thoughts at the engineering challenge.

 

=====================================================================

 

One standard design of Solar sail has few long booms (or masts) that hold the film at its apices. The stiff parts of the sail are less long for the same film area, hence less difficult to build light. At identical boom length, a square provides the maximum area-to-length ratio, letting call the design "square sail". A pentagon loses 5% ratio, a hexagon 15%, and the drop accelerates.

 

 

Though, if testing the deployment of a hectare-class sail on the ground, for which buildings limit the size, more sectors enable a bigger sail. I propose to try at a time just one sector between two booms. A continuous film can protect the payload against Sunlight, which is mandatory very near to the Sun; elsewhere we can split the film in sectors.

As a wind-free building, I take a roofed soccer or rugby stadium: 100m*70m of flat lawn, where we can float or hang the booms and the film for the test. This is bigger than all Solar sails built up to 2013, and the building is still decently common.

The best sector orientation on the lawn is (...with luck):

• Sector base parallel to lawn's diagonal for a triangular sail;
• Sector base parallel to lawn's length for a square, pentagon, hexagon;
• Sector side parallel to lawn's length for a heptagon, octagon;
• Sector height parallel to lawn's length for a nonagon to dodecagon and more;
• At some farther number, the sector's height is better parallel to the lawn's diagonal.

The resulting area varies irregularly with the number of apices. 5 is a good blend. 10 is a remote optimum, with much boom length.

 

 

Marc Schaefer, aka Enthalpy

 

=====================================================================

 

A nice innovation on the Solar sail Ikaros is that it uses Lcd surfaces as an attitude control. That is, by making these eccentric surfaces more or less reflective, radiation pressure is controlled there, which creates a tilting moment.

But Lcd have some drawbacks. They consume some electricity permanently, since the polarity must be reversed regularly to avoid wear-out. They are sensitive to sunlight, needing some protection.

My proposal is to replace them on Solar sails by a thin electrochemical cell, as the one I described in EP0564012:
http://www.freepatentsonline.com/EP0564012.html
http://www.freepatentsonline.com/EP0564012.pdf
which needs no organic material and should thus be more resistant to UV.

This patent has meanwhile been granted (and should hence be available in English) and abandoned, so its technology is public and free.

I never prototyped it. Making it work reversibly many times could require development.

I also considered it as a thermal control surface for spacecraft.

Marc Schaefer, aka Enthalpy

 

=====================================================================

 

Producing 3µm polyimide film is easier than I first thought. Just take 7.5µm film and make it thinner.

All right, all right, this needs further explanation.

You know the machines that metallize similar films? Their big vacuum vessel encloses two huge film rolls - the source and the destination - between which the film is moved in front of an aluminium vaporizer. Pump once, metallize kilometres of film.

Now, take such an existing machine, possibly a decommissioned one. Replace the vaporizing component by an etching unit - I suppose plasma etch would be fine, reasonably fast and wouldn't deteriorate the remaining 3µm thickness, as opposed to wet or gaseous etching. Adapt the pressure accordingly.

And then, add thickness sensors before and after etching and build a feedback loop to control etching speed. Here you get the necessary precision that prevented thinner films from being laminated.

As polyimide in this thickness range is semi-transparent to visible light, a dirt-cheap light attenuation sensor is enough. If needed, more sensors and etching units can be spread across the film's width to make the thickness uniform. You can also proceed in several steps, possibly by passing the film several times between the rolls.

Add a metallic roll to stabilize the film-to-plasma generator distance if it helps. Add light shields between the plasma and the sensors, add wavelength filters, modulate the sensors' light source to protect against parasitic light. Blah blah, you already guessed all this.

While this process may be too expensive for the most common uses on Earth (metallized polyester is used to wrap sweets) (and who needs 3µm film on Earth anyway) it looks really cheap for a Solar sail.

Marc Schaefer, aka Enthalpy

 

=====================================================================

 

An other way to make thin plastic film for the sail:

Take a varnish, lacquer or similar. Pour it over a denser liquid, so the varnish floats on it and spreads. Proper amounts achieve really thin films, if the varnish takes enough time to dry. The method was used to make wing films for ultra-small model aeroplanes. Much thinner than 25µm.

Polyimide varnish exists to coat high temperature transformer wires. Polyimide is dense, but some benign liquids are denser, like perfluorodecalin. To scale from 1dm² to 1km²... Pouring and pulling continously the film to a coil as it dries can be an element of answer.

Or maybe the varnish can be sprayed on an antiadhesive roll or film instead of poured on a liquid.

A few crossed carbon fibres glued on the film would usefully stop the propagation of slits.

I prefer the thinning machine already described, but it needs some investments.

Marc Schaefer, aka Enthalpy

[Moonguy]

Once again, a very nicely done piece!!

You can imagine how useful this is for operations at Mercury. Not only is Mercury the best environment for operating sails, it is the best for their fabrication as well. A facility on Mercury can mass produce ail material from local sources (including the polymides) and build fleets of sails for a fraction of the cost of identical Earth-launched counterparts.

Sails launching from Mercury's orbit have the benefit of at least six times the photon flux. As a result, a given payload mass launched from Mercury will develop the needed velocity much faster than the same mass being launched by the same sail from Earth's orbit. Alternately, much larger masses can be transported if we are ok with more cruise time.

It's the money involved that matters.

If you look at the equipment that would be needed to build sails, it is way less than that needed to build 777 aircraft or even cars. The only difficult task is stowing the 4 square kilometer sail for launch. If you are launching from a planetary surface with no atmosphere, the task is easier. If the sail is constructed in orbit by a self-positioning assembly jig, it is only really a matter of joining spars (and mast, if there is one) to a control/docking hub and then attaching the sail material rolled out from cassettes. Most of the jig would be open space-frame structure while only a few elements would be dense concentrations of machinery.

The only real limiter to solar sails is how much payload we can launch for a given mission from the planet's surface. Here again Mercury has the advantage.

What evolves from this is huge potential for economical mass transport all over the Solar System.

[Externet]

Has any solar sail/motor type of device has ever been built and worked on earth, experimentally ?

Sort of a Crooke's radiometer, with the radiation-pushed half of a spinning sail exposed to sunlight and the returning half shaded, all perhaps in high vacuum ?

 

If half the globe of a Crooke's radiometer is shaded, would such be an equivalent of a solar sail motor ?

[Moonguy]

Has any solar sail/motor type of device has ever been built and worked on earth, experimentally ?

Sort of a Crooke's radiometer, with the radiation-pushed half of a spinning sail exposed to sunlight and the returning half shaded, all perhaps in high vacuum ?

 

If half the globe of a Crooke's radiometer is shaded, would such be an equivalent of a solar sail motor ?

Japan's space agency, JAXA, has successfully flown a probe to Venus using a solar sail called IKAROS. Solar sails must now be regarded as a 'proven' technology - TRL 9 or 10. It remains to be seen how large they can be made. Since solar sails rely exclusively on solar photon pressure to operate, there is no 'motor' as we usually think of, such as an 'ion drive' would have.

A radiometer-based engine would be close to the elusive 'perpetual motion' machine. The problem is any mass of propellant would have to be expended at great velocity or be extremely large. There are mechanical limits to how fast a radiometer can spin with existing materials. Still, iI is an intriguing thought. . .

[CJMark5]

Hi i have a question what happens when the Solar Sail reachs the point between to stars does it stop or is there only one side that can be pushed.

[Moonguy]

Hi i have a question what happens when the Solar Sail reachs the point between to stars does it stop or is there only one side that can be pushed.

Picture aluminum foil that is very shiny (reflective) on one side but somewhat dull on the other. . .

Solar sails also have two sides. One is a reflector which can routinely be upwards of 90% reflective. A good mirror by any standard.

The other side is an 'emitter' surface that allows the sail to re-radiate most of the 10% of sunlight absorbed.

In between stars, the sail need only turn its reflector side towards a destination star and use its light to maneuver into the distant system. If it fails to do this, and the emitter side remains turned to the distant star, the sail will not be able to slow down as the emitter side is usually much less reflective. Solar sails cannot just stop at some balance point between stars as their excess velocity (as relates to the Sun) would keep them moving.

[Enthalpy]

Hi i have a question what happens when the Solar Sail reachs the point between to stars does it stop or is there only one side that can be pushed.

Radiation pressure gets really negligible, even on the huge area of a Solar sail, well before midway to the next star.

 

They're excellent candidates to Venus, Mercury, Sun-near missions, to the asteroids, possibly up to Saturn but that's harsh.

 

Any mission beyond involves accelerating close to the Sun and sailing thereafter on the acquired velocity. Sun-escape missions with Solar sails are perfectly possible, sensible and advantageous. It's easier than stopping at an outer planet.

All scenarios involve to pass close to the Sun and accelerate there. The most advanced scenarios first go farther from the Sun, plunge from there in a close Sun pass, to get the maximum kick.

Has any solar sail/motor type of device has ever been built and worked on earth, experimentally ? [...]

Several true Solar sails have already worked. They were all disapppointingly small: a few dam², while we need a few hm² to get good performance just for a 100kg probe. 1t would rather mean 1km², especially if a sail shall outperform my Solar thermal engine.

 

I'm not quite sure unfolding was tested on Earth... Even for the small sails up to now, I read mostly about testing in orbit.

 

My objective is to test the few hm² on Earth. I'll put more thoughts when I find time. Beyond this area, testing on Earth becomes more difficult.

[CJMark5]

hi i have a another question can light moved by magnetic forces.

[Enthalpy]

Simple answer: no.

 

Though, a few physicists try to detect a super-tiny effect in a mega-magnetic field. To my knowledge, no effect seen at all.

 

If light propagates in matter, the picture changes. Magnetic or electric fields can hae a strong influence on some material, including their optical properties. Banal observation at LCD screens.

 

Some Solar sails want to use a magnetic field, by a current loop that materializes only the edge of a disk hence is light. Though, these sails try only to catch the charged particles of the Solar wind, which provide a pressure 2 or 3 magnitudes fainter than the already tiny pressure by Sunlight. Is that any less difficult?

[Enthalpy]

The standard design with a few radial booms grants more mass to each boom and eases its conception... I too eventually grasped that, and here's a seemingly manageable boom design at rugby stadium scale, with five 85m long booms for 1.72hm² sail area, plus a fore and an aft half-booms.

Each boom consists of fourteen segments, 6.1m long plus overlap, which are concentric tubes that extend after separation from the launcher. The length fits in Falcon, Zenit, Soyuz fairings, Vega with some adaptation, and more. The angles between the extended booms result from shrouds whose nonparallel directions prevent collective buckling at the segments' joints.



The tubes have inner and outer faces of 110g/m² carbon fibre reinforced polymer (almost a papersheet, exists for hobbyists) on a 2mm core of 100kg/m³ balsa. Polymer foam would degas less. Could balsa be pyrolyzed and densified by a carburizing gas? That would achieve clean and possibly refractory sandwiches, not only for solar sails.

Central segments have D=0.30m, decreasing in 4.6mm radius steps to D=0.18m for the far segments. This leaves only 2.4mm radius clearance for tolerances, guidance and drive, for which cables and punched belts look possible. Segments weigh 2.1kg mean, 5+2/2 booms 180kg.

Taking for the Cfrp 1550kg/m³, longitudinal 98GPa and transverse 47GPa, the D=0.30m segment buckles at 48kN (oval) but 32kN (Euler) and the D=0.18m segment at 7kN. When testing one sector on Earth, the static compressive force is 1kN at D=0.30m. Look how nicely megalomaniac:



A 7.5µm film adds 200kg (including some fibres to stop tears), 25km shrouds 40kg, so the sail weighs 420kg or 24g/m². 480kg bus and payload get 61µm/s² when towing with 45° at Earth's distance from the Sun: reach Mercury in 5 years, take 5 more to bring samples back or to achieve a polar solar orbit.

Marc Schaefer, aka Enthalpy

[Enthalpy]

To communicate with Earth, a presently projected manned base on the Moon's far side needs a data relay. This can be a first application for a Solar sail on the scale already described here.

The relay would reside 2Mm North or South of the L2 Lagrange point, 65Mm beyond the Moon. The tilted sail provides over decades the permanent 46µm/s² thrust inaccessible to chemical propulsion. The relay is always clear from L2 where the Moon eclipses the transmission: better than the fuel-saving eight-shaped wobble around L2.



The thrust has always a component opposed to the Sun's direction, so the relay resides also before or after L2, maybe it circles around it in a month. The sail's orientation can rotate regularly or oscillate up to about twice a day. I won't refine this.

The 1.7hm² sail can have 5 booms and sectors as described on 25 August 2013. The booms described on 15 March 2015 weigh 180kg, shrouds 40kg, a 13µm film 320kg. At 60° angle of incidence, 41mN North or South allow a 900kg spacecraft, leaving 360kg for the bus and payload.

If compatible with the sail, the bus could resemble a geosynchronous satellite: one side facing Earth and Moon, with antennas on East and West sides, a rotating Solar panel at North or South side. Emitting 300W=+55dBm at 10GHz and 65Mm distance (-209dB), antennas of D=1.8m (44dBi) at the relay and D=1m (39dBi) on the Moon let receive -71dBm, enough for high definition videos. Interrupting the sail's metal film several times per microwave length makes it transparent.

If starting from slightly tilted Leo, the craft would take good 5 years to reach L2. Vega can put the 900kg on a higher orbit.

Marc Schaefer, aka Enthalpy

[Enthalpy]

2010 TK₇ is an Earth's Trojan, the first discovered and presently only known one
https://en.wikipedia.org/wiki/2010_TK7
http://communications.uwo.ca/trojanasteroid/downloads/pdfs/NatureLetterFinal3-preprint.pdf
orbiting the Sun in 1 year, 60° before Earth - but with 20.89° inclined orbit and with 0.809AU perihelion, 1.190AU aphelion (AU = Sun-Earth-distance ~149.6Gm).

A visit should bring interesting science, for instance to compare its composition with Earth. Alas, the inclined orbit is costly to reach. Wiki cites 9.4km/s from Earth's orbit to our Trojan's one; this depends on the script. Just tilting an elliptic 0.809/1.190AU orbit would cost 9.7km/s, ouch.

Chemical propulsion in a direct shot would divide the mass by roughly exp(3)=20 when arrived there, bad idea. Combining with Earth's escape as Oberth effect makes no miracle. Jupiter can tilt the orbit as it did for Ulysses, but going there costs time and 5.6km/s if not provided by Venus and Earth, braking at our Trojan too. I won't check if Earth and Venus flybys provide the inclination better than Jupiter does.

The sunheat engine and its 12.4km/s ejection speed would divide the mass by less than 3 at destination, less bad.

----------

With the size and booms described in this thread on Aug 25, 2013 and Mar 15, 2015, a solar sail tested on Earth brings a probe there in a decade, takes samples back, with no lost mass. We're getting somewhere.

Having 10 sectors tested individually on a rugby pitch, the sail is 3.25hm² big, and at 1AU and 45° to the light's direction, it pushes 105mN outwards and perpendicularly. The sail with booms, film and shrouds weighs 1100kg estimated as previously, the bus and science are granted 900kg, so the transverse acceleration is 52µm/s² at 1AU. Some unspecified launcher shall put the 2t above Earth's gravity; polar speed by Oberth effect would save delays.

Tilting the orbit can be done only around the apsis of the destination, so it takes twice as long: not 9500m/s by 52µm/s² but 11.6 years at 1AU.

This goes faster if nearer to the Sun. Orbital speeds increase as R^-0.5 but the sunlight's pressure hence acceleration as R^-2. Spiralling for 2 years, the craft can be at 0.763AU, where tilting takes 7.7years. The dumb sum would be the same, but as the sail tilts and lowers or raises the orbit at the same time, the optimized combination saves maybe 2 years for 10 years to our Trojan, less with the launcher's help.

----------

At 2010 TK₇, the probe makes pictures, remote analyses, and takes samples. Rather a small craft separated from the sail, probably the same the re-enters Earth's atmosphere. A hollow tethered harpoon pulls samples from many locations to the not landing craft? Someone else shall think at it.

The Earthlings' sail is as big as the target body. Delicately megalomaniac, isn't it?

----------

The return trip is easier because the capsule with the samples aerobrakes in Earth's atmosphere, so the transfer orbit can remain tilted. Raising the perihelion to about 1AU to slow the craft down to cross Earth after about 1 year would cost only 1.5km/s around the aphelion in an operation lasting 1.3 year, so the aphelion would be lowered a bit too. The apsis line outside the ecliptic plane is more complicated to imagine but costs as little. The return leg fits in 2 to 3 years.

The cost to the elliptic orbit is small and adds quadratically to the tilting cost, that's why I neglected it in the leg to our Trojan.

The capsule brakes from 15km/s in the atmosphere and lands with parachutes. Only the capsule limits the samples' mass. Check my lightweight boxes
http://www.scienceforums.net/topic/85103-mission-to-bring-back-moon-samples/?do=findComment&comment=823276
Marc Schaefer, aka Enthalpy

[Enthalpy]

Bringing samples back from Mercury looks feasible with the sizes and masses described here on Aug 25, 2013 and followings.

A sail of 10 sectors and 3.25hm² tilted by 45° to the Sun's direction pushes 105mN at 1AU (Sun-Earth distance). Weighing 1100kg plus 1300kg for the bus and the science, it accelerates by 44µm/s² here.
Starting just escaped from Earth, with 29.8km/s speed around the Sun, it takes 5.5 years to spiral in the ecliptic plane to 0.387AU, the mean Sun-Mercury distance. Mercury's orbit is tilted by 7° so its 47.4km/s have a 5.8km/s polar component. Combining this quadratically with half of the 17.6km/s spiral delta-V result in 19.3km/s need, or 10% more, lengthening the travel to 6.0 years. Mercury's elliptic orbit needs little performance more. The return leg takes as long.

The launcher injecting the craft towards Venus for a braking slingshot there would save much time, with the sail operating only near the Sun. Same for the return path. I won't detail this option now; later maybe perhaps, or not.

Capture by Mercury and orbit lowering is trivial. A low orbit with 4km/s escape and 2830m/s orbital speeds takes 1170m/s to join. 290µm/s² at 0.387AU acting 1/3 of the time achieve it in 138 days.

There, a 900kg prospector (or several smaller ones if possible) separates and lands on the night side. It gets electricity from liquid oxygen, hydrogen and fuel cells. 100W for 50 days (less than 88 days, one local night) at 50% power efficiency consume 64kg propellants. 100W shall feed the experiments and sampling activities, transmissions and housekeeping, and is enough to keep the probe lukewarm. The prospector uses also oxygen and hydrogen to descend to the surface and to climb to the orbit, all from the same superinsulated balloon tanks described elsewhere. Details later maybe, or not.

Mercury rotates by 1 turn in 58.6 days versus the distant stars, so in 50 days it turns by 180-26.4°. The prospector can't regain the same orbital plane unless we restrict the landing site to the equator. Though, the orbiting sail can change its plane meanwhile: turning 2830m/s by 26.4° costs 1305m/s which, at 470µm/s² (900kg less) acting 1/3 of the time, is achieved in 96 days. The prospector shall wait in orbit before descending or after climbing. Raising the apoapsis for the manoeuvre and lowering it afterwards may speed it up. More orbit changes permit several landing sites at different longitudes.

After the symmetric return leg, a capsule brings the samples to Earth's surface.

Marc Schaefer, aka Enthalpy

[Enthalpy]

More speeds and delays for the Mercury samples mission. Still not quite accurate.

A cheaper launcher like Vega-C can put the probe on a 400km Earth orbit (or higher for the drag), preferibly polar 6h-18h. The sail then adds 7675m/s in spiral to escape in some lengthy 7 years.

But if a mid-heavy launcher injects the probe in Venus transfer, the mission saves time. I have still no firm data about a Venus slingshot, so I take arbitrary 1.5km/s backwards from Venus' speed. The resulting trajectory is elliptic (but Mercury's orbit is too) with 0.66AU mean radius, from where the sail brakes to mean 0.39AU in only 1.8 year. The flyby can even more usefully tilt the orbit to Mercury's plane. This estimate in inaccurate, but 4 years saved on each leg is a lot - or the mission can be heavier and land on more sites.

I had computed an elliptic capture at Mercury, but the probe will rather spiral down. Twice the Delta-V but for 2/3 of the time, same duration hence.

A 1100kg D=203m sail isn't agile (2*10⁶kg*m²), even at Mercury (30.4µPa incoming pressure). Two 100m² on-off control surfaces at 45° provide 0.2N*m. The peak angular acceleration is  100nrad/s², so a sine oscillation of peak pi/4 takes a period of 5h (and 32h at Earth). Compare with 2h orbit around Mercury: the sail can only rotate regularly around itself, and the controls steer slowly this rotation or drift the orbital plane to follow the Sun's direction. On a 6h/18h orbit, the sail can stay at 45° to sunlight. Or the sail can rotate by 180° per equatorial orbit to push flat away from the Sun at one point and show its profile at the opposite point. Consequently, don't expect the sail to cast shadow on the bus.

The Sun's tidal force is strong on Mercury orbit and might let a sail's adequate orbital plane rotate to catch the prospector rising back from the surface. Or not. I didn't check.

From a Venus slingshot, the probe would reenter Earth's atmosphere at 11.5km/s while a spiral transfer without slingshot (+4 years) lets brake the probe to 7.9km/s around Earth (+7 years). The lighter heat shield doesn't justify it to my eyes.

Marc Schaefer, aka Enthalpy

[Enthalpy]

General figures about the thermal design for the Mercury samples mission.

Far from Mercury but at 0.307AU (Sun-Earth distances), sunlight is 14480W/m² strong. It heats to 300K a perpendicular second-surface mirror that absorbs a=0.03 of it and emits e=0.95 infrared from one side. Shaded areas are cold.

Staying close to Mercury's noon area at 600K (700K are reported) and parallel to it, an ideal flat area absorbing infrared from both faces but no sunlight attains 505K=+232°C, a conducting sphere too. The probe's outer faces should survive >600K but can't cool the equipment when facing Mercury.

A possible design puts several cooling faces at the probe body and makes a thermal contact (by a fluid or a deformation) with the coldest one at each time, when it doesn't face Mercury nor directly the Sun. At 60° incidence but emitting from a single side, a second-surface mirror reaches 300K. Averaging with Mercury's night side helps a lot, especially at the batteries. Active steering of cooling surfaces is less reliable. The probe doesn't need a general fridge cooling cycle.

----------

A thin polyimide film for the sail (arbitrary a=0.3 e=0.9 both sides) facing the Sun reaches survivable 454K=+181°C and radiates infrared to the probe's body but from a limited solid angle. Choose well the polymer deposited after the metal.

Cooling the telescopic booms is difficult.

----------

The solar cells could populate sparsely a Sun-facing area. If they're transparent to the near-IR (a=0.7), cover 1/4 of the area, the rest offering a=0.03 and e=0.95, and both panel's faces emit (e=0.95), sunlight alone heats to 436K=+163°C, 600K Mercury in the back 550K=+277°C shortly. A single emitting face must be less populated. Few mm thick aluminium foil spreads the heat of small 20mm*20mm cells without heatpipes.

Wide bandgap cell semiconductor is the obvious choice. The soft bond with the support conducts heat away easily but must resist the peak temperature.

----------

The surface of Mercury is still hot when the Sun sets. 200h night cool very roughly 0.2m soil whose surface gets chilly but is blasted clear by the landing engines, so insulating the prospector's bottom seems better. Deep soil probes would come warm.

----------

De-orbiting and landing on Mercury consume about 4.3km/s, taking off and navigating to the orbiter 4.0km/s. Two stages of hydrogen and oxygen shall do it.

The descender starting with 700kg burns 0.94+0.14kg/s at 30bar to push 5kN in 1*D0.6m. Including 2.2g/s in the 15kW 8kg fuel cell, the I_sp=4622m/s=471s lands 276kg, of which 50kg are abandoned, so the unpolluted crawler+ascender weigh 226kg.

Minus the crawler with instruments but plus the samples, the ascender takes off with 100kg to meet the orbiter with 42kg only, leaving maybe 10kg for the packaged samples. Check my lightweight boxes
http://www.scienceforums.net/topic/85103-mission-to-bring-back-moon-samples/?do=findComment&comment=823276
Multiple prospectors look hard. The Earth reentry capsule must stay with the orbiter. A heavier and slower mission injected to Venus flyby by the launcher would be much better. Without the descender+crawler+ascender, the return leg isn't as long.

Hydrogen needs active cooling at least at the ferry-orbiter. I won't detail the easier oxygen, and as mass at the ferry and the descender is easier, I concentrate on the crawler+ascender.

7.5kg hydrogen for the ascent plus 5.0kg for mean 50W over 70 days fit in D=0.75m. Polymer belts holding the tank leak negligible heat, but from the 300K surrounding ascender, 50 plies of e=0.02 multilayer insulation leak 300mW. A cryocooler, third-Carnot efficient from 20K to 250K, uses mean 12W, rather with stronger intermittent runs and a mechanical separation at idle; its heat dump operates in Mercury's shade, when awaiting the orbiter as well.

Or let the hydrogen evaporate at the crawler+ascender. 300mW evaporate 4.2kg over 70 days, so the fuel cell taking hydrogen mainly gaseous would keep the liquid cold. That's marginal with the present figures but feasible at a bigger prospector.

I wouldn't burn methane in the engines and crack it for electricity. Lose 20% mass at the descender, 20% at the ascender, double the oxygen+fuel consumption for electricity, add complexity and mass.

Marc Schaefer, aka Enthalpy

[Enthalpy]

More about the Venus flyby for the Mercury samples mission. It brings much more than I hoped. But as these computations are slippery, I may have made mistakes.

Kate Davis gives an angle in relation with the asymptotic speed
http://ccar.colorado.edu/imd/2015/documents/BPlaneHandout.pdf
which I rearrange slightly with an escape speed at the periapsis

50m/s lost in drag need 10^-10kg/m³ residual atmosphere, 0.5µm erosion about as much, and 400W/m² a bit more. Having to atmosphere table for Venus, I made horrible scaling factors with Earth and decide the periapsis shall be at 683km over the 6052km planet's radius. The escape speed is 9825m/s there. And here's an angle-versus-speed graph for this periapsis at Venus

Ariane 5eca can put 4550kg at 3475m/s above Earth's gravity, Ariane 6 hopefully too. Aimed 38° Sunwards, or outwards in an other launch window, the probe arrives at Venus with 4713m/s and 60° sunwards and forwards. Well, if Venus were in the ecliptic plane, sorry folks.

The Venus flyby can give the probe 4066m/s polar speed, zero radial speed (this isn't optimized) and 2365m/s below Venus' orbital speed, so the probe's orbit is in Mercury's plane (...if Venus were in the ecliptic) and with 0.648AU semimajor axis.
VenusFlybyNum.zip

From there, the probe sails to Mercury's orbit, which costs roughly 10760m/s. At (mean!) 0.648AU, the sail accelerates 4550kg by 55µm/s², so this step costs 4 years. Or a bit more, since Mercury's apsis line is not exactly in Venus' orbital plane, and so on.

The rest of the mission stays the same, but with more mass. 1100kg sail leave 700kg for the bus and 2750kg as the descender starts: four times more mass than previously. The ascender has now a comfortable size, and since the electricity doesn't scale up, it brings more than 40kg samples back. Two prospectors look possible.

The return path, with 1840kg, takes only 1.5 years to accelerate towards Venus.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Researchers let fungi chew some components of Norway spruce and sycamore to lighten them
researchgate.net
their goal is to build bowed string instruments, but if fungi can lighten balsa wood, sandwich materials improve.

I proposed a graphite-balsa-graphite sandwich for the walls of the solar sail's booms, in this thread
Mar 15, 2015
but such sandwiches have many uses.

The paper reports on figures 3a (axial) and 3b (radial) 15% mass gained after 20 weeks chowing on Norway spruce (-19% E modulus matter less here). If balsa too gets 15% lighter, a 10-segment sail is 30kg lighter. That's one science instrument more.

Merry Christmas!
Marc Schaefer, aka Enthalpy