KSP Community CubeSat

[MrOnak]

Quick heads-up: Interorbital Systems are offering CubeSat kits that include a free launch with them for around 21,000 USD. Details of what's included are in the brochure.

Noteworthy is that their CubeSat is of the same form factor but minor things have changed in respect to the CalPoly spec to there is no guarantee that their specific model would be able to fly on other launch providers, in case something happens with Interorbital System. May or may not be a deal breaker but definitely something to keep in mind.

That said, their price is pretty good compared to the alternatives I've found so far.

[K^2]

Sadly, I think we will have to go with zetaboards, as it's probably the best free option. Other free things like phpbb are nontrivial to set up. I'll go ahead and start up a zetaboards forum for us.

Erm... Why not simply set up a custom board, something like Simple Machines Forum, on a free/cheap hosting site?

[Newt]

Nicholander: It will probably be a bit less easy to follow everything that is going on at the same time, but it should be better organized, and probably simpler therefore to jump into a discussion that you have not been obsessively following. We could try to do some sort of weekly update for the different sections (for example), to try to keep everyone informed. We do have a relatively small group, and therefore getting everyone's input on whatever they know about, is important.

Mr0nak: I would somewhat question the reliability of that company. They do not seem to really have done that much, and we would be flying on a rather unproven rocket. I suppose that is the price of cheapness? Worth looking into at least.

K^2: I am not really sure. This seems to be happening rather quickly to me, which is on one hand nice, and on one hand somewhat too hasty, perhaps? I do not really imagine that LordQ will do something horrible or expensive, however, though, apart from my quick research, I am unsure of the pro's and con's of these different tools we are looking at.

[Mazon Del]

Re-reading your previous post. Bringing the gases to the sensor(s) for analysis is definitely an issue. Thinking about a solution ...

There are a few ways we could go about it, but I don't really like any of them. Tubes, valves, etc. We either end up with a bunch of equipment that costs a lot of mass in order to pipe the gasses around while keeping things safe in case one or more have a leaky seal, or we go a different high mass (and high cost) route and just put a sensor package on each container. Technically we don't HAVE to get the gas data, but it is highly preferred.

Additionally, upon thinking about it, one advantage towards having the larger volume of air is that it ends up being a bit of a heat sink for the plants. Ideally, this should mean that it helps to damp down the oscillation of heat from the sat passing from light to dark.

[MBobrik]

There are a few ways we could go about it, but I don't really like any of them. Tubes, valves, etc. We either end up with a bunch of equipment that costs a lot of mass in order to pipe the gasses around while keeping things safe in case one or more have a leaky seal, or we go a different high mass (and high cost) route and just put a sensor package on each container. Technically we don't HAVE to get the gas data, but it is highly preferred.

I think we have at least four possibilities.

1. A pressurized vessel in which the rotating sample holder, sensors, cameras, servos and heater will reside.
   + simple
   + makes the sample holder simpler because it does not have to deal with the pressure.
   - the bulkiest solution
   - risky, one sealing failure ends the mission.
   
2. The sample holder will be the pressure vessel, cameras and servos will be outside, sensors will be inside.
   + less bulky.
   - risky, one sealing failure ends the mission.
   
3. Individual samples sealed.
   + safe, losing a few samples is not a big deal.
   - no gas sensor data, because while the sensors are small, all need a certain volume of gas to work with.
   
4. The middle way. Sample holder divided into a few sealed groups of interconnected samples, each group with its own sensor.
   + safe, losing one or two groups will still leave enough data.
   - multiple sensors cost more, each sensor ~$100
   - would be most probably too complex in the samples in capsules version
   

[astropapi1]

I'd go for a sealing the entire Cubesat. I mean, it gives us more surface area for radiators and reflective surfaces and it'll be easier to manage the temperature.

I was also thinking that, since the radius of our centrifuge will be so small, the G difference between the base of the plate and the "top" of the moss can be significant. It'll be 0 at the middle, remember?

[MBobrik]

I'd go for a sealing the entire Cubesat. I mean, it gives us more surface area for radiators and reflective surfaces and it'll be easier to manage the temperature.

As we are going to have more problems with losing heat than with overheating, more surface is not a good thing. The smaller the heated compartment will be, the better.

I was also thinking that, since the radius of our centrifuge will be so small, the G difference between the base of the plate and the "top" of the moss can be significant. It'll be 0 at the middle, remember?

That's why the samples will be arranged in a ring. Each will be at the same distance from the axis of rotation.

[Mazon Del]

I'm not 100% sure what the difference is between options 1 and 2 are exactly.

Also, the moss WILL have to deal with a gradient, having the moss in the ring just makes it far easier to calculate the effect of that.

[MBobrik]

I'm not 100% sure what the difference is between options 1 and 2 are exactly.

Also, the moss WILL have to deal with a gradient, having the moss in the ring just makes it far easier to calculate the effect of that.

1. got a larger immobile pressurized container inside of which the unpressurized lightweight sample holder will rotate. servos and cameras will be inside.

2. the rotating sample holder itself will be sealed and pressurized, while the servos and cameras outside will be in vacuum.

And yes, there will always be a small gradient, but if we count 45 mm diameter and 5 mm sample thickness, it will be only a difference of 11 % between outer and inner side, which is negligible methinks.

EDIT:Mazon Del You are of course right. should be radius not diameter... trivial mistake

Edited December 17, 2014 by MBobrik

[MrOnak]

1. got a larger immobile pressurized container inside of which the unpressurized lightweight sample holder will rotate. servos and cameras will be inside.

2. the rotating sample holder itself will be sealed pressurized, while the servos and cameras outside will be in vacuum.

And yes, there will always be a small gradient, but if we count 90 mm diameter and 5 mm sample thickness, it will be only a difference of 5.6 % between outer inner side, which is negligible methinks.

Well you should calculate the difference in terms of radius, not diameter since that is what affects the change in gravity, but regardless, you're right...

http://www.endmemo.com/bio/grpm.php

[Mazon Del]

Ah! Thanks for the clarification. Now, we could do option two, while the whole inner ring is still the separate capsules. We'd end up caring more about the rotation systems in terms of needing to put rotation logic in. This is because the atmospheric sensors would need to be mounted on the ring. It really isn't that difficult to add the logic though. What might be an issue is that the trailing cables would need to have a track of some sort to ensure that they don't flop everywhere, and we may or may not care about how difficult this makes keeping our COG constant.

[MBobrik]

What might be an issue is that the trailing cables would need to have a track of some sort to ensure that they don't flop everywhere

There won't be any trailing cables. the sensor logic on the sample cylinder will be connected by two sliding contacts.

[Nicholander]

We could try to do some sort of weekly update for the different sections (for example), to try to keep everyone informed.

I like that idea, on what day of the week should we have the updates? Saturday or Monday? (Those are the 2 best options in my opinion)

[LordQ]

@Nicholander, sadly, that's exactly the problem. vBulletin is a great forum software, but it costs money.

@K^2, I haven't been considering software like Simple Machines so far because they only offer the forum software and not hosting space. Places like Zetaboards or Lefora offer forum software as well as free hosting space, and so are easier to set up. However, I do see that deljr15 has some server space, so I'll definitely look into Simple Machines as well.

Also, the name KSP Community CubeSat is a bit too long for a forum name. The zetaboards forum I quickly started up last night to explore features rejects the name because it exceeds its 20 character limit for a forum name. Any other suggestions? Some quick and snappy would be best, ideally one word or two. And clearly we should reference Poland ball.

Edited December 17, 2014 by LordQ

[Mazon Del]

There won't be any trailing cables. the sensor logic on the sample cylinder will be connected by two sliding contacts.

Sliding contacts are very VERY icky. Plus you'd need far more than two. Power, ground, signal. And that is per each sensor that is on the system. So if we have temperature, pressure, O2, CO2, and humidity, that ends up being 7 assuming that you managed to find equipment that all worked off the same power requirements and thus could use the same rails. If your sensors are outputting analog, you'll get a bunch of noise that will muck with the A2D conversion later. If your signals are digital, then you are going to have to worry about occasional data corruption due to the same noise. These are the main reasons why in almost every case requiring a spinning turret or similar device, the cables end up being put into some sort of spool that gets wrapped/unwrapped. It's just better.

[MBobrik]

Also, the name KSP Community CubeSat is a bit too long for a forum name. The zetaboards forum I quickly started up last night to explore features rejects the name because it exceeds its 20 character limit for a forum name. Any other suggestions? Some quick and snappy would be best, ideally one word or two. And clearly we should reference Poland ball.

don'tknow... just shorten up the long name to 'KSP Cubesat'

[Mazon Del]

KSP MossSat perhaps? Still banking on the Cossack Kerbal.

[MBobrik]

Sliding contacts are very VERY icky. Plus you'd need far more than two. Power, ground, signal. And that is per each sensor that is on the system. So if we have temperature, pressure, O2, CO2, and humidity, that ends up being 7 assuming that you managed to find equipment that all worked off the same power requirements and thus could use the same rails.

No. there will be a small microcontroller with A/D on the cylinder, which will power up the sensor one at time, gather the signals and deal with error correction, and the signal will be sent over the power line. So just TWO sliding contacts.

I used the this trick a few times in my own contraptions. And the data gathering will be done only when the cylinder is not moving, so no noise.

These are the main reasons why in almost every case requiring a spinning turret or similar device, the cables end up being put into some sort of spool that gets wrapped/unwrapped. It's just better.

Wrapping cables are icky to me, because it would mean we would have to prevent the cylinder from rotating beyond one turn, so it would have to move back and forth, and it would have to be physically prevented from moving further to prevent yanking or tangling the cables if the servo control screws up. Sliding contact allows the cylinder to rotate freely so we can get from and to arbitrary sample by turning just in one direction, and nothing bad happens if the servo logic accidentally tries to turn more than one turn...

[MrOnak]

Sorry if I'm wrong but if you manage to get a well conducting sliding contact through the plastic of the petri dish and remain a pressure tight seal that survives the rigours of launch, you can get a wire through that as well.

[Nicholander]

Just shorten it to KSP CubeSat like MBobrik suggested.

[sndrtj]

I'm very late to this discussion, but where can I donate to this project? Not that I have much to spend, but every penny's something, right?

As a biologist I really love the moss idea! Could someone perhaps summarize what are the targets for the moss experiments? Thanks! :-).

[Nicholander]

If you want to know what's going on, read everything after page 123. So you don't have to read everything before that.

[LordQ]

I'm very late to this discussion, but where can I donate to this project? Not that I have much to spend, but every penny's something, right?

As a biologist I really love the moss idea! Could someone perhaps summarize what are the targets for the moss experiments? Thanks! :-).

We don't have any means for accepting donations yet, but thanks for your enthusiasm.

I believe the targets for the moss experiments are to see how the moss grows in centrifuge 'gravity' of something similar to lunar and martian gravity. I'm not 100% versed i this though, so I can't really give more details. We do have another thread that discusses the moss actually, however, there's not much in that thread just yet.

[sndrtj]

We don't have any means for accepting donations yet, but thanks for your enthusiasm.

I believe the targets for the moss experiments are to see how the moss grows in centrifuge 'gravity' of something similar to lunar and martian gravity. I'm not 100% versed i this though, so I can't really give more details. We do have another thread that discusses the moss actually, however, there's not much in that thread just yet.

Thanks! I'll head over there for the moss discussion :-).

[Newt]

Regarding the cables wrapping around, sliding contacts seem a tad risky to me. More importantly, I do not think that the wire tangle will be too much of an issue if we program intelligently. It can just spin around one way once, and back again the next time. Provided there are no easy places to snag or get caught, all should be fine letting the wires drift around a bit.