KSP Community CubeSat

[K^2]

Looks to me like we don't need to sacrifice rad-hard ICs either way. First round of qualification testing can be done with stand-ins, and TVAC bakeout would have to be done with the actual flight build regardless of qualification/protoflight route.

Or am I misreading something?

[Nicholander]

Will such a low cost processor be able to support 2 Cameras? Will we have to have some sort of secondary chip just for handling image processing?

Why would we need 2 camera's anyway? One is enough.

[christok]

Looks to me like we don't need to sacrifice rad-hard ICs either way. First round of qualification testing can be done with stand-ins, and TVAC bakeout would have to be done with the actual flight build regardless of qualification/protoflight route.

Or am I misreading something?

I believe you are. "Qualification testing is performed on an engineering unit hardware that is identical to the flight model CubeSat." For our own internal tests, we can of course go through as many cheap stand-ins as required.

[Kryten]

You'd want a flight spare regardless of whether it's strictly required; if you don't have a sat available for launch, and there isn't enough time to arrange for someone else's to be substituted, most providers will launch a mass simulator and charge you anyway.

[K^2]

I believe you are. "Qualification testing is performed on an engineering unit hardware that is identical to the flight model CubeSat." For our own internal tests, we can of course go through as many cheap stand-ins as required.

Qualification is just stress tests. I can't imagine any reason to have the expensive ICs in there if they are going to be same package installed in the same sockets. There is also a form for waving parts of the test requirements. This might be a sort of thing that can be waved. At any rate, it'd be worth the time to verify all of this at some point. I'll make a note of that.

Why would we need 2 camera's anyway? One is enough.

Forward-facing camera for general pictures. Rear-facing camera taking pictures through the eco-chamber.

You'd want a flight spare regardless of whether it's strictly required;

There would definitely be spares constructed, but they'll probably be built with cheap ICs. There will probably need to be at least one spare solar panel as well, but not a whole spare set, in all likelihood.

Edited July 20, 2014 by K^2

[Nicholander]

About how long do you think it will be until we can start our KickStarter?

[christok]

Months at least.

[indroth]

Cameras:

http://www.diyspaceexploration.com/various-options-imaging-modules-cubesat-payload/

[K^2]

Thought more about magnetotorquing. Rather than making small coils with ferrite cores, it might be a good idea to have frames with large area. (For a current loop, torque is IAxB) We can count on a mag field of about 30μT. Maybe a bit stronger. The loops can carry an amp easy. So a 100 turns on a frame that goes around the full face of a sat, we can get 30μNm of torque. With a 1kg cube, depending on how the mass is distributed, we could be looking at 0.01 - 0.02s^-2 of angular acceleration. That will get us up to necessary rotational speed in about 10 minutes, which is more than we could ask for.

So that definitely takes care of the attitude control. I'd probably want a 50 turn frame on each face with individual amps for a touch of extra reliability. The 8051 handles all data access as reads/writes from the bus, so the coil amps can simply have an address on that bus, allowing up to 256 field intensity settings.

[Nicholander]

Yeah, that's a great idea! We should do that!

[dharak1]

Would we be able to supply an Amp? What kind of voltage would we be looking at?

[Mazon Del]

Are there any COTS magnetotorque motors available for this sort of application, or would we be making them ourselves?

[Rakaydos]

Speaking of the electrical system, what are the best low mass/high energy solar cells I sould be looking at?

[Nicholander]

Rakaydos, I found this Solar Panel Datasheet on Clyde Space: Solar_Panel_Datasheet

[Aethon]

Hmmm. This was just posted in the welcome aboard forum.

http://forum.kerbalspaceprogram.com/threads/87413-KSat-aerospace-students-say-hello

Maybe they can help.

[K^2]

Are there any COTS magnetotorque motors available for this sort of application, or would we be making them ourselves?

The above calculations are for 6 coils of 50 turns each with no core. I see no point for paying for anything but the spool of wire for that. There might be a point in paying for something special to drive these coils, however. I'm still looking at best way to build the bus for this thing. I don't know if it's better to have a single address decoder and a single point of failure, or have each channel decode its own address, increasing complexity and reliability.

The coils, however, I just plan to wind to the inside of the aluminum frame of the sat. If there is a need to 3D-print an endoframe to support the coils and sensors, that's always an option. Exoframe will be custom-cut aluminum to meet specifications, and will have multiple points of attachment to the plastic endoframe. That should provide sufficient resistance to any vibration during the ride.

[Mazon Del]

Well it is a matter of precision I'd guess. Considering the environment we are working in doesn't dampen out slight misalignments of things like thrust, I'd at least look into how much it would cost for a professionally wound system. If it wasn't too much extra budget, I could just pay for it, especially as it would likely be cheaper than having high res magnetic imaging done to determine if we are good or not.

As far as aluminum goes, there is a guy near me who owns a CNC waterjet cutter that is industry grade. He bought it for his own projects, and he does the work super cheaply for others on small projects. Seriously, someone my group knows went to him and he apologized that for the work he was doing (something like cutting an intricate design in a like 2ft x 2ft space) he had to charge the guy his minimum fee of $40. The other guy laughed and said it was ok, as he had been quoted something like $4,000 from a standard place.

Are their any space-worthy plastics we can currently 3D print with? IE: Non-outgassing and/or capable of ignoring vacuum/crazy temp changes.

[K^2]

Alignment would be a factor if we needed precise maneuvering. But if the sat is going to spin at 60 RPM, any mis-alignment will go into tumble, which is easily detected and corrected for.

I'll run some simulations with slightly randomized fields, but I honestly don't expect it to be a problem.

Once in orbit, temperature shouldn't be a factor. Plastic could get brittle when cold, which could be a problem on the ride up. Any idea where we could check on the temperatures of the payload during the ride? I did not think about outgassing. Hot + vacuum could be an issue. I'll find out what the relevant factors are, and if any of the available 3D printables are fine for it.

[Nicholander]

Also, is it me or are a lot less people posting on this thread? It's probably that everyone is playing 0.24. .

[AngelLestat]

So anyone has an idea of the possible total cost?

I guess you should compare all ideas with a list of components, mission objectives and cost of each part (showing sources if its has).

With an average aproximation of space lifetime of each part.

That is a required step before choose an idea and enter in a deep mission design step

When the mission design is finish, only then we can create the kickstarter campaign and advertise the project to get extra funds.

[dharak1]

Last time I checked we were aiming for a cost around 10 - 25 thousand.

[K^2]

Plus launch and whatever we end up doing for ground station/comms, yes. Still working on the details. I keep getting side-tracked into specifics of the board layout. But that's not really a bad thing.

[dharak1]

How big of a problem would board layout be when we have to deal with heat and all? Since PCB is a better heatsink than the vacuum around the chips would they all end up overheating each other from the heat they take in from the sun and the heat they generate?

[K^2]

8051 draws 25mA at 5V. That's nothing compared to solar input. The only part that might run a bit hot is the comms unit, but that should only be used in bursts. That leaves the Sun as primary source of the heat to deal with.

Worst case, we can cool electronics with an active heat pump. (Solid state.) These things are cheap enough, and you can use them for noise-free cooling of a CPU on a modern PC. It's a bit of a power hog, but since we only need it when solar power is abundant, I don't see a major issue with it.

[Mazon Del]

One thing I have been a little concerned about is how the radio will react to the rotation. Rotating as fast as we are going to be might cause issues if we use a semi-directional (depending on where it is mounted) antenna, and while an omni directional is my personal choice being that it reduces the need to aim the sat with any particular care, they waste more power (the signal radiating off to space) and thus will require more to match the signal strength of a directional.

Have we had any particular thoughts on radios yet?

Edited July 23, 2014 by Mazon Del