KSP Community CubeSat

[henryrasia]

K^2 glad you're back! Have you seen the doc? I thought it'd be good to have a place with all the information. And thank you, astropapi, I agree that we're not making the progress we should be doing. I know no one is to blame, but we should be getting through research and decision-making faster than we are. That's why I thought of assigning responsibilities to people so that their efforts would be more focused. I just don't feel like we actually believe we're going to carry out an actual space mission, and so the tone of the designing process seems very hypothetical and as just a exercise for thought. I do not know if you will agree with me on this, but there's the problem: I don't know you guys, and I'd like some kind of confirmation that we're actually doing this. How will we cause hype for a kickstarter if we're not hyped ourselves?

I can say I am! Let's do this team!

[Nicholander]

Endersmens, how's that website going? And yes, I'm hyped up too about this henryrasia!

[DarkStar64]

K^2 glad you're back! Have you seen the doc? I thought it'd be good to have a place with all the information. And thank you, astropapi, I agree that we're not making the progress we should be doing. I know no one is to blame, but we should be getting through research and decision-making faster than we are. That's why I thought of assigning responsibilities to people so that their efforts would be more focused. I just don't feel like we actually believe we're going to carry out an actual space mission, and so the tone of the designing process seems very hypothetical and as just a exercise for thought. I do not know if you will agree with me on this, but there's the problem: I don't know you guys, and I'd like some kind of confirmation that we're actually doing this. How will we cause hype for a kickstarter if we're not hyped ourselves?

I can say I am! Let's do this team!

All aboard the hype train!!

[K^2]

K^2, welcome back! Anyway, when are you going to update the first post, and how's the sim going?

Working on it. And I'll need to regroup on the sim. I still have one of the orbital elements being miscalculated.

[NERVAfan]

Glad to see you back K^2!

[Engineer of Stuff]

Is your Kickstarter page up, K^2?

[henryrasia]

NO NO NO NOOOO!!!

No kickstarter before having a solid project to show. That's how kickstarters fail: no one believes you're going to deliver and if you don't, a second kickstarter is out of the question because you've lost credibility. The best kickstarter is the one that could still be made without it, but not as quickly / as good as if with the money. At the very LEAST we need to have a prototype built before starting. So we should get to work on that. I agree we're running late on ourselves. We're not hired, so we need to set up our own deadlines.

[Nicholander]

Deadlines? Okay, here's an idea: (Feel free to criticize it and change it, I'm not that good with setting deadlines)

Early-Mid November: Have a basic concept on the CubeSat.

Mid December: Have the sim be done.

Start of 2015: Start actually engineering and designing the CubeSat.

March 2015: Have a design/blueprint that would actually work.

April-May 2015: Start talks/discussion with a space agency to send the thing to space.

Mid June: Start the Kickstarter.

Mid July 2015: Kickstarter finished, and assuming we got the money, start building parts of it to test. (Like put on a balloon)

October 2015: Tests done, start building the CubeSat.

November 2015: Start setting up communications systems around the world, we can ask people on Reddit and the KSP Forums to do that, I think someone mentioned something like that above.

End of 2015: Have the CubeSat be done and tested, then we wait until the launch provider will put it on a rocket.

Whenever the LV provider says so: Have it done and able to launch.

[LvL]

looks good, but, uh, don't we already have a basic concept on the cubesat?

[henryrasia]

Not if this is this will be an actual, real life space mission.

We need specifics: Materials, components, designs, dimensions, spacing, prices, physical requirements (resonance frequency, pressure and stress limits, heat limits, radiation limits, etc), and many other things NERVAfan should have read in the books. And yes, we're very behind in that respect. And Nicholander, that's a great deadline list. I would only say that we can compress the designing/prototyping/testing in less time. Also, the kickstarter and negotiations don't take time to do, per se, but yes a long time waiting. We should probably be doing things while those are going on. I say we have the concept of the cubesat completely done before the end of this year (no prototype though) and while that happens K^2 would do the sim. BTW are you using STK? Cause you should probably use something made instead of coding something from scrach, if nothing else to avoid bugs.

[Nicholander]

By concept you mean like what I put "Design/Concept that would actually work"? Also, K^2, can you put "NEXT DEADLINE: End of 2014, have a design/blueprint that would actually work and have the sim be done" in the first post?

[NERVAfan]

As for design - basic limits...

Dimensions: 1U is limited to 10cm x 10cm x 10cm, maximum mass 1.33 kg, center of gravity must be within 2 cm of its geometric center in all 3 dimensions.

Price: Not sure yet, depends on kickstarter so we can't really have a finalized design until after that, though yes, we certainly need a working design going in to the kickstarter.

Power: I don't know what the limits are or how much we will need. K^2, do you know?

Time/labor: This is tricky also. We are all in different parts of the world, so how will the physical assembly be done?

A lot of things, like the basic structure & solar panels, we can buy "off-the-shelf" specifically for Cubesats, and I think we should so that we can focus our efforts on the mission-specific stuff.

http://www.clyde-space.com/cubesat_shop/structures/1u_structures

http://www.cubesatshop.com/index.php?option=com_virtuemart&page=shop.browse&category_id=1&Itemid=66

http://www.cubesatshop.com/index.php?option=com_virtuemart&page=shop.browse&category_id=17&Itemid=79

[K^2]

Well, if you don't want to risk opening panels, you can at most expose 3 faces at 1/sqrt(3) efficiency, which gives you effective area of 170cm², or 0.017m². At 1.3kW/m² and 20% efficiency, we're looking at 40W maximum. Realistically, it will probably be less, and only available for half of the time. Average power consumption needs to be kept under 10W with this setup.

Fortunately, the only seriously power-hungry system is transmitter, and we can burst that from battery.

But I still think we should be at least considering fold out panels. We'll need to spin the sat for attitude control either way, so we might as well make panels deployable. That will give us effective area of up to 500cm², more than tripling available power. I'd be much more comfortable with this overhead and added redundancy. Though, I admit, lack of mechanical parts does appeal to me as well.

[LvL]

I don't really think we should use fold out panels, because if we're really unlucky, we could have a mechanical failure. And the panels will be more fragile.

[K^2]

Cube rides up to orbit folded up, and once in orbit, whether it's fragile or not won't matter. And if there is a mechanical failure, worst case scenario, the cube defaults to the same configuration you'd use if you didn't have deployable panels. There is added complexity, but no actual increase in failure modes. On the contrary, you end up with more options to fix any problems that do develop.

[LvL]

I just realised, we're more of a space agency (in development) rather than a group of people just making a cubesat. We have our own goals, and are making the cubesat ourselves. So YAAAAAAAY! what should we be called? perhaps the kommunity cubesat organisation (KCO)?

(just being a bit silly at the end there )

[henryrasia]

Yeah! That's the spirit!

I really think we shouls have a name (though that's a low priority). I think it should be unrelated to KSP and more to the fact we are a bunch of noobs (effectively) making a space program. The slogan could be: "We can into space" (referencing also some internet culture).

NERVAfan, we should definetly prefer to use off-the-shelf components to minimize liabilities (they could even be space-graded already!). But we should at least have a price range to know how much to ask in the kickstarter. Also have a list of capabilities to show off in it as well. Having a "range" will also allow us to make concrete stretch goals such as "double battery size for faster transmission" or whatevesr it may be.

K^2, Deplyable solar panels? Hmm... Do we really need that much power? We should begin with answering that. Also, the whole thing is going to be spinning wildly for 3 weeks straight, would't that cause structural stress on deployable panels and be minimized with surface panels? On the other hand we could have them along the axis that does not spin, but then it would have to be spinning at the same crazy velocity backwards to "stay still". I know they can move to track the sun, but at that speed for 3 weeks? I'm not dismissing it, but we should look into that.

[Nicholander]

I think we can determine if we will have deployable solar panels or not is to know how much power our CubeSat will consume.

[NERVAfan]

Also, the whole thing is going to be spinning wildly for 3 weeks straight, would't that cause structural stress on deployable panels and be minimized with surface panels? On the other hand we could have them along the axis that does not spin, but then it would have to be spinning at the same crazy velocity backwards to "stay still". I know they can move to track the sun, but at that speed for 3 weeks? I'm not dismissing it, but we should look into that.

That does sound potentially problematic. And it would probably be cheaper without the deployable ones. So if we CAN do it without the deployable ones, maybe we should set that as the baseline, and consider the deployable ones as an "extra"/stretch goal if we get more money than we expect on the kickstarter.

What will consume power?

-Transmitter

-Magnetorquers (attitude control)

-Camera and IR LED(?)

What else? Will thermal control be just insulation?

[LvL]

If our thermal control will just be insulation, we'll need some pretty good insulation.

[deljr15]

http://www.clyde-space.com/cubesat_shop/structures/1u_structures

http://www.cubesatshop.com/index.php?option=com_virtuemart&page=shop.browse&category_id=1&Itemid=66

http://www.cubesatshop.com/index.php?option=com_virtuemart&page=shop.browse&category_id=17&Itemid=79

I have to say as a manufacturing engineer those prices are outrageous for what your getting. The structure could be custom made far cheaper and possibly better. (tailor made)

[dharak1]

This is the kind of thought we need. Especially for us. If we have a custom frame we wont need the test material to fit in a proper "stack" or anything for that matter. It lets us build around our camera if it's too long also.

[deljr15]

I just read the google doc on this project.

The idea of running 2 experiments could be accomplished by placing the experiments an appropriate distance from the center of mass. Allowing for a constant rpm of the satellite with separate simulated gravities.

[LvL]

Can we get the google doc link in the description K^2?

[K^2]

I think we can determine if we will have deployable solar panels or not is to know how much power our CubeSat will consume.

Yeah, that's probably the best way to go about it.

If we are still ok with rad-hard version of 8051 CPU, that draws about 25mA at 5V. That's 125mW. RAM, receiver, and other circuitry will increase the minimum, but we should be able to keep basic "not dead" state in under 1W. If our average power supply drops bellow that, we're screwed. But that can be achieved with just one 5x10 panel at a weird angle to the Sun. So overall, it's pretty safe.

This will be enough for the sat to remain in the stand-by mode, waiting for either a signal or timed operation. So what we need overhead for is to reduce time between operations. Transmitter might require 200W+, for example, which will be the biggest drain. But taking pictures and even getting sensor data will also require extra power. And the baud rate will be pretty bad. I'm thinking 50k-100k, which means that even a single JPEG might take a minute to send, which is about a full pass. If we wish to catch it again on the next pass, which will be about 100min later, it will need at least 2W of average overhead to recharge. If we wish to have it try and communicate with several stations along the way, that's even more power.

So 20W I'm confident about with fixed panels is enough, but it's a bit less breathing room than I'd like to have. If we don't end up having any other power-hungry systems, I would probably lean to call it sufficient in the name of simplicity. If something else comes up, we'll have to think it through.

Can we get the google doc link in the description K^2?

I don't think I have a working version of that link? I'll probably find it as I catch up, but if somebody can drop it here, it might save some time.

I have to say as a manufacturing engineer those prices are outrageous for what your getting. The structure could be custom made far cheaper and possibly better. (tailor made)

Agreed. There is no reason to waste money on cube-specific frames. We can get custom ones machined to tolerances for much less.

So far, the only must-have expensive thing we need is solar panels, with rad-hard CPU and RAM being on the short list if we can get the funds. Almost everything else can be either off-the-shelf consumer grade or custom built.

Edited November 2, 2014 by K^2