Printable Version of Topic

Posted by: vjkane Aug 8 2020, 05:14 PM

[ADMIN NOTE: New topic started using moved posts from Pallas telescopic observation topic. Please read and keep rule 1.9 in mind. Thanks!]


I was hoping that NASA would select the Athena mission to do a flyby of Pallas using a cubesat spacecraft under the SIMPLEx program. It selected the Janus mission to flyby two near Earth asteroid binary systems instead. The Athena team plans to submit again.

Posted by: antipode Aug 9 2020, 12:14 AM

That's good news vjkane! Cubesats for initial exploration of interesting NEOs and objects out to the main belt seem like a great way to go.
I think there's a need for an ongoing program that sits well under the Discovery class cost-wise that can take advantage of cubesat architectures with limited but focused payloads,
only solar power, and new cheaper smallsat launchers or rideshare opportunities.

P

Posted by: vjkane Aug 9 2020, 01:15 AM

Don't get too excited. While it was ranked Category 1 (fully selectable on science and implementation), NASA didn't select it. There also has to be a launch opportunity. The previous proposal made use of the Psyche Mars flyby and the fact that Pallas is currently close to the ecliptic.

All we know about the new proposal, based on the PI's statements, is that the next version will be much better.

Posted by: JRehling Aug 9 2020, 07:51 PM

Given the low mass of cubesats, I wonder about a mission that sends many of them on a Jupiter (or, possibly, Earth) flyby and spreads them out enough, before the flyby, for the gravity assist to scatter them into solar orbits that cross back through the asteroid belt and fly by many different targets from an initially-similar trajectory. There have to be many opportunities for such orbits to intersect the orbits of two targets (out of thousands to choose from) so that the number of asteroids observed could be on the order of 1.5 per cubesat. Thereby, one low-mass mission could give us flyby observations of many asteroids.

The investment in delta-v for the Jupiter flyby would be high if you only wanted to observe a single target, but for the payoff of observing many targets, the science / cost would become more appealing.

The devil is in the details as to what a Jupiter gravity assist can achieve, but a couple of inspiring examples are Pioneer 11, which gained a high inclination en route to its Saturn flyby, Ulysses, which gained a 79° inclination for observing the Sun. Getting to Pallas is easy in comparison to either of those.

Posted by: vjkane Aug 10 2020, 01:51 PM

That's almost certainly ballastically possible. Two issues: 1) cubesats still have low reliability and therefore lifetimes and 2) no one has figured out how to as dramatically cut the mission operations costs.

Posted by: HSchirmer Aug 18 2020, 07:44 PM

It occurred to me that it might make sense to reverse direction first - Venus is closer and IIRC aligns more often for flybys(?) So an initial Venus gravity/Oberth boost might be a faster way to get out to Jupiter. Also, I VAGUELY remember that rockets using RP1 for fuel (kerosene/diesel) are denser, and pack more delta-v for Oberth maneuvers boosts, compared to cryogenics (H2 or CH4) or hypergolics. IIRC problem is, if you fly RP1 past ~1.0 AU it turns into Jello unless you heat and stir it. So, perhaps aim IN towards Venus, use cheap Lox-RP1 and get the Oberth/gravity boost there?

Also seems to me that a cubesat heading towards the asteroid belt could benefit from some sort of solar sail, for course correction on the way out, for station keeping once you get there, and as a solar collector to reflect light back onto the cubesat solar cells to keep the battery charged and the cubesat alive.

Question- how much of an Oberth/gravity boost CAN you get from a sling shot maneuver around the moon?
I have to ask, because years ago, somebody speculated that the GRAIL data and a super accurate lunar gravity map would allow a bonkers "Tors Twillinger"-"REVERSE HOVER-SLAM" massive Oberth & gravity boost from a lunar flyby.
It's essentially the "Thor's Twins" turn from The-Hunt-For-Red-October, (but with a rocket.)
You skim over the surface of the moon around 300 feet/100 meters above the surface, now that you are deep in the Lunar gravity well you light up a high delta-v Lox-RP1 engine, blast across the lunar seas, and then thread the gap through the mountains at the west end of the Ocean of Storms, then make the final course correction that gets you to the outer solar system in a hurry.
Could it work?

Posted by: antipode Aug 18 2020, 10:34 PM

You don't want to mix up your metric and imperial units for THAT one!

P

Posted by: stevesliva Aug 19 2020, 02:33 AM

(re: Star Wars lunar gravity boost)

Right, you instead mix up whether you're using units of distance or units of time.

Posted by: HSchirmer Aug 19 2020, 02:40 AM

Ah, the "good old days." Friday, April 20, 1962. Neil Amstrong bounced an X-15 off the atmosphere, then executed the fastest U-turn in human history, and came in low and fast into Rosamond dry lake.

-"Tower, AirBoss wants to know, just how far WAS Neil from the Joshua trees?"
- Capcom Petersen "Um, tell AirBoss, that, eh, Neil was AT LEAST 150 feet away from the Joshua trees."
- "This is AirBoss to Tower. Were those trees on Neil's right or his left..."

Posted by: JRehling Aug 19 2020, 09:55 PM

1) Venus does not align more often for flybys; this is easy to lookup and check. The synodic period of Venus is 19 months and for Jupiter, 13 months… not that the difference would matter much in planning a single mission.

2) Adding a solar reflector for two purposes to each cubesat immediately defeats the principle of simplicity. I was suggesting that there might be an affordable, elegant solution and you're proposing an expensive solution; of course, those always exist. That's just an entirely different topic.

3) The magnitude of possible gravity assists depends upon many factors, but they vary with the amount of time that the spacecraft spends within the body's gravity well… time integrated times the acceleration due to gravity at that distance… and Jupiter has a mass 390 times that of Venus. It has easily the greater potential for bending the paths of objects on similar trajectories to different destinations.

Posted by: HSchirmer Aug 20 2020, 03:45 AM

Fair points. My points
1) cubesats rely on batteries & solar panels, not plutonium RTGs.
2) A big shiny flat solar sail would allow a cubesat to make course corrections without needing heavy cold gas thrusters.
3) The same big shiny sail, when parabolically curved, would focus dim sunlight onto a cubesat solar panel, allowing a cubesat to survive on solar power much further out that it could otherwise.

4) Of course, the sun has a mass 333,000 times that of Venus, is only 1 AU away and has NO synodic period, so it's thousands of times better for bending orbital paths.

Posted by: HSchirmer Aug 20 2020, 06:04 PM

Looks like the cubesat-solar sail is now in the extended mission...

https://spacenews.com/solar-sail-spacecraft-begins-extended-mission/

The next size up seems to be CubeSail, designed by CU Aerospace (Champaign-Urbana Aerospace) and the University of Illinois, is a low-cost flight experiment based on the UltraSail concept. This experiment uses two CubeSat satellites to deploy a 20 sq. meter solar sail, which can play the same role as propellant on a conventional spacecraft.
https://www.cubesail.us/

Posted by: JRehling Aug 21 2020, 01:43 AM

There are multiple reasons why the Sun is not an easy object for gravity assists.

Most compelling for the reason at hand, the Sun is not close in terms of delta-v; it is actually the single object in the universe that requires the most delta-v to reach. It is much less delta-v to reach Jupiter.

Posted by: HSchirmer Aug 21 2020, 02:59 AM

Excellent points, as usual.

Here's a quick run from the excellent "Atomic Rockets" website...
http://www.projectrho.com/public_html/rocket/mission.php#oberthe0


Fastest rocket yet was New Horizons using an Atlas/Centaur at 16km/s.
(For comparison, the Falcon Heavy tossed a Tesla at Mars at around 12km/s.)
So, an solar Oberth effect maneuver by an Atlas/Centaur should net ~10x that, about 160km/s.

Posted by: Holder of the Two Leashes Aug 21 2020, 03:07 PM

Yes, that's true in terms of going there directly. In a different sense though, it's the same delta-v as Jupiter, provided you go to Jupiter first. You can use Jupiter's gravity to slingshot a probe arbitrarily close to the sun. So if you are willing to add several years to the journey plus some hefty radiation hardening for Jupiter's environment, you can pull off getting to the sun with a lot less energy.

In early designs for the Parker probe, a Jupiter assist was being seriously considered. The result would have been a probe that could reach a point three times closer to the sun than it is in fact going to (pretty much the minimum survivable distance with current technology). All at the expense of spending very little time at the sun.

Here is an article at the Parker website that discusses that option:http://parkersolarprobe.jhuapl.edu/News-Center/Show-Article.php?articleID=104

Posted by: HSchirmer Aug 21 2020, 04:37 PM

Has anyone worked out a "RETROGRADE Grand Tour?"

Launch from Earth, and align for gravity assist/Oberth burns using-
1) Moon
2) Venus
3) Mercury
4) Sun

Net result- you STILL get where you want, but FASTER?

Posted by: JRehling Aug 21 2020, 07:28 PM

As a rule of thumb I keep in mind with orbital mechanics, escape velocity is sqrt(2) times the orbital velocity.

We're already orbiting the Sun. To send a rocket to the further point in the universe requires delta-v that is sqrt(2)-1 = 0.41 times the Earth's orbital velocity around the Sun.

To send a rocket to the Sun requires 1.0 times the Earth's orbital velocity of delta-v. So, roughly 2.5 the delta-v of a trip to infinity.

Of course, reaching Jupiter requires less energy than reaching infinity.

Posted by: HSchirmer Aug 21 2020, 09:34 PM

I think that's apples to oranges here.

The Atlas-Centaur with the New Horizons probe on it's nose was orbiting the Sun at ~1AU with an orbital velocity of 30km/s.
For comparison, the Hayabusa2 probe is orbiting the Sun at ~ 1AU at an orbital velocity of 30km/s.

The escape velocity needed to leave Earth "to infinity and beyond" is MUCH different than the escape velocity needed to get from Hayabusa 2 "to infinity and beyond". HOWEVER, thanks to the laws of conservation of momentum, you CAN do a meaningful Oberth burn when leaving earth. When leaving Hayabusa2, not so much.

Posted by: nprev Aug 22 2020, 12:40 AM

ADMIN HAT ON: Alright. This thread was created from posts that veered wildly off-topic in a previous thread. It has now veered wildly off-topic from what seemed to be the topic it veered into. In fact, it has now essentially turned into a debate between two members about orbital mechanics.

If this persists the thread will be closed and archived.

Posted by: vjkane Aug 22 2020, 01:19 AM

This is a paper relevant to this topic


Nanospacecraft fleet for multi-asteroid touring with electric solar wind sails

Abstract:
We propose a distributed close-range survey of hundreds of asteroids representing many asteroid families, spectral types and sizes. This can be implemented by a fleet of nanospacecraft (e.g., 4-5-unit CubeSats) equipped with miniature imaging and spectral instruments (from near ultraviolet to near infrared). To enable the necessary large delta-v, each spacecraft carries a single electric sail tether which taps the momentum from the solar wind. Data are stored in a flash memory during the mission and downlinked at an Earth flyby. This keeps deep-space network telemetry costs down, despite the large number of individual spacecraft. To navigate without the use of the deep-space network, optical navigation is required to track stars, planets and asteroids. The proposed mission architecture is scalable both scientifically and financially. A fleet of 50 spacecraft will be able to obtain images and spectral data from 200 to 300 near-Earth and main belt asteroids. It allows study of those asteroid families and spectroscopic types for which currently no close-range observations are available. This paper presents science objectives, overall science traceability matrix, example targets and technical challenges associated with the mission. We open to discuss preliminary requirements, mission and spacecraft designs.

https://ieeexplore-ieee-org.offcampus.lib.washington.edu/abstract/document/8396670

Posted by: HSchirmer Aug 22 2020, 01:32 AM

Understood; and fair point, a polite discussion which has a run of 4 post by 2 members isn't unusual.
We're just starting to see cubesats for exploration, asking whether the delta-V is better spent by waiting to launch when you have an outward synodic alignment, or launching inward sooner and trying to make up the difference by an Oberth burn is, well, a cool and geeky question. The way that orbital mechanics was figured out was by people debating cool and geeky questions.

The first question I'm trying to get a handle on is that, while the Oberth effect is largest when "falling towards the object at rest from infinity" what percentage of this would be available when falling "from rest" at a L1 or L2 Lagrange point?

The second question whether the 28 day rotation of the Earth-Moon L1 & L2 points could be useful for sending the cubesats on their initial course via an Earth Oberth effect burn. See "The Lagrangian points of the real Earth-Moon system" https://www.researchgate.net/publication/28120754_The_Lagrangian_points_of_the_real_Earth-Moon_system

Won't know until you try, or at least try to think it through.

https://xkcd.com/

Posted by: mcaplinger Aug 22 2020, 08:43 PM

As interesting as the Oberth effect is, it seems to have seen limited use in interplanetary missions so far, largely because spacecraft either don't have significant post-injection delta V capability or they need it all for orbit insertion at the target. (It does get used to the extent possible during that orbit insertion.)

For example, look at where the burns were in the Juno 2+ deltaV-EGA trajectory. https://trs.jpl.nasa.gov/bitstream/handle/2014/45416/08-2728_A1b.pdf

Posted by: HSchirmer Aug 22 2020, 10:19 PM

You are absolutely right. My fault.
Until I read the story of Hiten/Delbruno https://press.princeton.edu/books/hardcover/9780691094809/capture-dynamics-and-chaotic-motions-in-celestial-mechanics and watched Virginia Tech's Department of Aerospace and Ocean Engineering - Ross Dynamics Lab https://youtu.be/fV0kUmtQWZU?t=586 I didn't realized that this is OBVIOUS to me because it is logically identical to chemistry principles of "activation energy" and "phase space." And these have NO analog in orbital mechanics.

So here goes-
1) Launch a rack with cubesats into LEO.
2) Launch a fully fueled booster rocket into LEO.
3) Do a half-century ago Gemini program dock of cubesat payload & fully fueled booster.
4) Use a "Hiten-DelBruno maneuver" to move cubesats & booster from an Earth-Moon Lagrange point to a Sun-Earth Lagrange point.
6) Light the booster rocket and send a payload of cubesats to Mars & the asteroid belt using 10% of the fuel you'd usually need.

Posted by: mcaplinger Aug 22 2020, 11:39 PM

Did it turn out that step 5 was unneeded?

Would this work in theory? Probably. Is it practical from an engineering perspective? Not really, at least not right now. For example, there are no "space tugs" with high delta V and long on-orbit duration. The Centaur upper stage, for example, has a lifetime measured in hours or maybe a few days.

There are a lot of competing constraints in mission design, it's not all about orbital dynamics.

Posted by: HSchirmer Aug 23 2020, 01:19 AM

Do you think it could be adapted to last 4 months? That would allow a Hiten-style path and put an upper stage into lunar orbit that arrives 90% full.

I looked but could not find Centaur on orbit specifications; can you post the link to those refernces?
(That is why I went with the Atlas Heavy RP1-LOX "double stack," it appears the RP1 lasts longer than cryogenic H2. However I'm guessing that MethaLox is stable for at least as long, given that SpaceX starship reliance on MethaLox at Mars.

Do the pros/mods have a list of the rocket fuels at used t Mars? I would guess that somebody here knows the history of Mars mission rocket fuels? I ask because that IS part of the basic blueprint for cubesat fuel to Mars and the asteroid belt. (If not, sounds like a great conference poster!)

Posted by: mcaplinger Aug 23 2020, 01:38 AM

All US Mars missions have used either hydrazine monopropellant or MMH/N2O4 bipropellant. I'm unaware of any deep-space application for any cryogenic fuel of any kind after initial injection by the launch vehicle.

Centaur duration: "The present day Centaur vehicle looses [sic] upwards of 17-20 % lbm of Hydrogen per day" https://www.ulalaunch.com/docs/default-source/extended-duration/centaur-extensibility-for-long-duration-2006-7270.pdf (There
are also avionics thermal control constraints but I'm not sure how driving those are.)

Keep in mind that any mission of reasonable cost pretty much has to use off-the-shelf systems; they typically can't afford to develop their own from-scratch flight infrastructure.

Posted by: HSchirmer Aug 23 2020, 05:45 PM

(EDITED) Good points about cryogenics and good references as usual! That paper confirms a few months IS realistic. "It is feasible to achieve a long-duration cryogenic stage that can sustain itself in a ready state for weeks, months and potentially up to a year in an autonomous coast mode." Summary Page 6. So given how the Muses-A/Hiten project got 456 lbs to lunar orbit, in 5 months (instead of 3 days) but only used 10% fuel/delta-V we can expect to do the same with a similar weight payload of 150 standard cubesats. Once they're in lunar orbit, it's possible to get them to Mars, and beyond.
I had NOT seen the "Cubesat to Mars" proposal January 29,2020.
Jeff Dillon - Cubesat to Mars - 22nd Annual International Mars Society Convention
https://www.youtube.com/watch?v=9K8_VkIjU9c
That "Cubesat to Mars" shows just how little fuel is needed to get a cube sat to Mars, IF you can get it to Lunar orbit.

Another way to avoid loosing cryogenic fuel is to not use cryogenics The British that launched their 'Prospero' satellite on a rocket called the "Black Arrow" https://en.wikipedia.org/wiki/Black_Arrow which burned RP1 & H2O2(85%) and could put 135kg into LEO. That's a VERY interesting fuel mix: non cryogenic, stable, and 85% H2O2 is a possible monopropellant for reaction control & ullage. Also a VERY interesting payload weight, it's essentially 100 standard cubesats. If you were to remove the Black Arrow's "Waxwing" 3rd stage, you can pack in 300 more cubesats for a total of 400 cubesats out to Mars or the asteroid belt. Let's call the 2 stage version of the Black Arrow with 400 cubesats the "Black Bolt."

Replace the standard second stage of a Delta-IV heavy with the Black Bolt. Have the Delta-IV heavy put Black Bolt on a Hiten low fuel trajectory to lunar orbit. You've now got 2 stages @ 2.6 km/s each parked in lunar orbit with 400 cubesats.
-One option- an Oberth burn around Earth heading to the asteroid belt.
-Second option is to continue the "low fuel trajectory" from Lunar orbit to Mars. That would take longer, but you could deploy ~200 cubesats during the journey and use a Mars gravity assist to direct them into the inner asteroid belt. That leaves the Black Bolt time for a Mars gravity assists to put it into the outer asteroid belt with 200 cubesats in reserve and two 2.6 km/s stages available to disburse them. Use the first stage to spread 100 cubesats into the outer asteroid belt. Use the last stage to spread 100 cubesats among the Jovian Trojans.

Posted by: HSchirmer Aug 24 2020, 12:17 AM

Wow, synchronicity. Rocket youtuber Scott Manley just noted plans to update the Black Arrow.

Posted by: djellison Aug 24 2020, 01:18 AM

I'm afraid some extreme bubble bursting is required here.

How are you going to pay for, operate, communicate with and build these?

Note - the only cubesats to successfully operate beyond the EM system were the MarCO pair each of which were 6U cubesats of 13.5kg each. They were not 1kg 1U 'standard' cubesats. They lasted ~9 months. They required DSN time (which is tens of thousands of dollars per hour). They had deployable high gain antennas, a finite attitude control system, and a budget of over $18M for the two.

They had basically no instrumentation beyond a UHF radio for InSIGHT relay and two 50g worse-than-SD-quality cameras (of which only 1 of the 4 flown actually worked properly) They could communicate at only 8kbps during their prime mission.

The Delta IV heavy you propose using is a $400M launch vehicle. You can't just replace its second stage - that's required to even reach LEO. A Delta IV heavy WITH it's second stage can't even get a Black Arrow to GTO, let alone a trans lunar injection.


You're definitely going a bit beyond blue sky thinking here.


The way to do Cubesats beyond the EM system is to piggy back a couple at a time on a larger dedicated missions as MarCO did with InSIGHT - or to pay for a dedicated ride with someone like Rocket Lab whose upgraded Electron rocket can now deliver Cubesats beyond the EM system.

What cubesats are doing is incredible.

They are not some magical panacea that can be scattered across the solar system by KSP style magical engineering.

Posted by: HSchirmer Aug 24 2020, 10:09 AM

Fair enough, these are ideas, not CAD drawings.

-Funding the science is supported by mapping the billions of dollars in metals that reside in the asteroid belt; same way that the science of geology piggybacks off oil/natural gas drilling and locating ore deposits. Funding might also reflect the need to have a "repeater" for the Deep Space Network to relay signals, where a mesh of satellites with radios provides redundancy for deep space data, and a VLA the size of the asteroid belt.
-Communication would not be to any one satellite, but to a "mesh network" probably leave a trio of larger, smarter "repeaters" in orbit around Mars to provide 3x throughput and tridundancy for talking to the "mesh". Basic cubesats would be similar to the concept of SpaceX's Starlink, although each Starlink is much larger, 260 kg or roughly 200 cubesats. I had not based the idea on Starlink, but those basic specifications are a good place to start: 4 phased array RF antennas, standardized solar panels, ion propulsion for station keeping, a star tracking camera, and collision avoidance/database of objects in orbit.
-Initial operation would be a 3d version of the "picket fence" that detected that Arrokoth was a contact binary using 21 occultation observations over time. Sometimes the LACK of data and for how long, IS the data your looking for. This is comparable to router firmware "load balancing" that detect data dropouts- except here the data dropouts would be loss of visual signal as an asteroid passes between a cubesat and the tracking stars, or loss of radio signal as an asteroid passes between two cubesats. This builds up a 3d map of occultations, which gives you asteroid locations.
After you build up a basic map, you'd move to an extended mission of trying to actively "ping" a specific asteroid and have other cubesats listen in to triangulate position like GPS. Then ping from different cubesats to build a 3d map of interesting asteroids, and finally ping at different frequencies to build up a texture map of really interesting asteroids.

-Building would be by the lowest qualified bidder. There would probably be a mix of cubesat sizes, 1x, 2x, 5x, 10x. Starting with 200 to 400 units, you would trade performance for redundancy.


Good point, IIRC, the early specs indicated they could do LEO by throttling down the center core after lift off to conserver fuel, then throttling up after the outer 2 separated. This might have needed a vacuum nozzle for the center core and been unstable at sea level.
Better solution would be, since the 1st & 2nd stages of the "Black Arrow" have the same 2.6km/s but the 1st is heavier, use a pair of the lighter 2nd stages for the "Black Bolt" to get a lighter craft with the same specific impulse, that would meet lunar payload limits.


Yes, that's the traditional way to do cubesats. But not the only way. I'm thinking in NK, number of nodes, number of connections. Since NASA/ESA/JAXA will need more deep space bandwidth, a distributed communications network further out isn't unprecedented, and using data dropouts to map the asteroid belt, is, admittedly thinking counter to expectations.

Posted by: nprev Aug 24 2020, 10:54 PM

This discussion has degenerated into a bottomless morass of 'if onlys'. Thread closed.