KSP Community CubeSat

[MBobrik]

We're currently trying to figure out how fast the magnetotorquer we will be using to spin up our CubeSat can spin us up, and we think it can spin us up to 44 RPM in 1 minute.

Actually, it's now only 7 rpm/min @10 watt

[Endersmens]

I'm not good at explaining things, but I'll try:

We're currently trying to figure out how fast the magnetotorquer we will be using to spin up our CubeSat can spin us up, and we think it can spin us up to 44 RPM in 1 minute.

Sorry if you already know this, but we have a Google Docs thing where we put all our ideas.

I think that's all I can tell you, like I said, I'm not good at explaining things.

That google docs will make making the website a lot easier.

Gah! My school computer won't let me work on the website. It won't let me switch from my school google account to my personal. I'll try to find a way around it.

Edited December 6, 2014 by Endersmens

[LordQ]

Torquer calculation sheet ... enjoy.

.

Sheet is set to the exact input values as the commercial torquer of LordQ just for comparison, and it still gets better numbers by cca 50 %. Feel free to change them to see the effects.

Nicely done. The one quibble I have left is that coil turns and wire diameter should be inputs instead of outputs, but that's an easy fix. I've added some of my calculations where I did that at the bottom of your spreadsheet for reference. The numbers in there don't necessarily mean anything, so feel free to change them.

In which case, I'll focus on figuring out how fast we can precess the cubesat with a given magnetorquer.

[Newt]

Is establishing rotation at a high acceleration at all important? Certainly, we cannot be left drifting without any control to speak of, but it seems preferable to have just what is needed, and possibly reducing the size and mass of the torquer could free up space/mass for other hardware, possibly for the redundancy in this and other systems.

Excuse me if any of this has already been addressed, the bulk of posts makes it difficult to quickly understand what precisely is decided on or against.

[K^2]

Newt, Mbobrik is trying to convince us to use directional antenna and rotate the sat towards target on Earth. I'm not convinced, at the moment, that it is a great idea. For reasons of reliability, complexity, added weight, and power generation and consumption. I'm not all together certain that it wouldn't affect the experiment, either.

[MBobrik]

Newt, Mbobrik is trying to convince us to use directional antenna and rotate the sat towards target on Earth. I'm not convinced, at the moment, that it is a great idea. For reasons of reliability, complexity, added weight, and power generation and consumption. I'm not all together certain that it wouldn't affect the experiment, either.

Well, I am not trying to convince. Actually, after discovering that even the really overpowered 10 W torquer could give us less than 7 RPM/min I am no longer convinced myself that it is the best approach. It was just an idea how to save us the expensive high sensitivity gear on the ground.

But the calculations have merit on their own ... they show the relationships between torquer weight, input power, and the resulting max available acceleration. We can use them to match and build the torquer to our needs, instead of having to pay thousands of dollars for what a reel of aluminium wire would do.

Edited December 7, 2014 by MBobrik

[K^2]

Oh, absolutely. Paying for a torquer would be a criminal waste of whatever money people end up entrusting us. This isn't an MRI spectrometer, where mag field has to be perfect.

As for ground station, we might be able to just do radio bursts with higher power. But I really am going to take a look at how feasible a ground array is going to be.

[Newt]

Many thanks for the explanations, K^2 and MBrobick.

If all you are wanting to do is point 'down' reliably, could it not be easier to use a passive means? I know on some communications satellites people have set up extendable booms that have weights on the far end, and those leverage the satellite to always point at Earth. We could probably do the same thing with, say, a tape measurer, with something of mass on the end (maybe something of other practical use, maybe just ballast), and use motors to roll the measurer out a decided length. I have done that with LEGO parts, in 1g, and it seems pretty reliably to work. I can start looking into the maths for it.

Regarding the ground radios, I really believe that we should look into the rules of the countries where we are putting them up; there are rules that address the communication with orbital radio stations, and amateur radio communication in general in many countries. I have not heard anything here that I know violates any of those rules, but it would be wise to figure that out sooner than later. Not all of our needs are hard- and soft- ware, some are legal too.

Edited December 7, 2014 by Newt

[K^2]

No idea about other countries, but here in the States, I might need a HAM license.

Tidal orientation-keeping isn't an option, because the sat is going to be spinning as required per experiment.

[Newt]

It seems to me like it would be possible, provided that the boom is along the spin axis. Of course, if you are trying to also keep one side of the satellite pointed at the sun, there will be problems. And a high acceleration to track a precise target would be needed, so as to follow it and get back. That seems like a lot of areas where something could go wrong.

In the US, you would need a ham license. Reading some into it, the FCC regulates the bands that people of three ham classes can use (there are also three classes in Australia), though you should be able to communicate on one of those bands that are available. It would appear that for the first level license in Australia, you need to use a commercially produced transmitter only, and there are some fairly strict rules regarding transmission strength. Also there are limitations on what you can do in the bands, i.e., on the 80m band in USA, only morse code at 200W or less. Probably the station would be between 6 and .23m if we are going by the US's rules. Alas, we seem to be mostly not around the equator, which adds challenges also on whether we will be able to reliably contact the satellite.

[MBobrik]

It seems to me like it would be possible, provided that the boom is along the spin axis. Of course, if you are trying to also keep one side of the satellite pointed at the sun, there will be problems.

The sat will be oriented towards sun with its axis. and also, the rotation speed will be quite high, up to 80-100 RPM. So all passive stabilization schemes like tidal or drag won't work.

Alas, we seem to be mostly not around the equator, which adds challenges also on whether we will be able to reliably contact the satellite.

Our most likely orbit would be ISS-like, with the same inclination, so, the best place to place a tracking station will not be the equator, but something around 51° latitude. Which means Canada, England, Netherlands, Germany or Poland, Ukraine, Russia, Kazakhstan, or on the southern hemisphere Argentina, Chile.

Of course anything in-between would work, though we would get transmission windows less often.

[bounding star]

we need to run some tests to find if this poor lump of moss will even survive a launch intact. I mean, looking at what nasa uses to test their sats for launch gives you a bit of an idea of what would be involved for the moss... http://motherboard.vice.com/blog/to-stress-test-rockets-scientists-built-the-loudest-soundsystem-in-the-world

[henryrasia]

Hey, in light of all of the problems relating to the sat being spinning (solar panels and directional antennae not being able to point to a particular place), how about we ditch that? An alternative would be to have a normal, still satellite with a centrifuge inside. This way we can control both spin rate and direction better. The only problem I see (besides figuring out a centrifuge) is the fact that we'd have to have double navigation for the sat while having a massive centrifuge inside (essentially an uncontrolled gyro). What do you think?

[Newt]

Henryrasia: That makes sense, but I would be concerned about the influence of drag between the two parts, as well as the communication between the two. It seems likely that we would continually need to push the centrifuge faster, as it would impart its force to the main body, which would have its own need to counter the imparted velocity. It might work, and certainly if we could then use a passive means to point down we might ultimately save some power. But that fails to address the solar panels. We would also need a means of communication and power transfer most likely, which could be established by, for example, metal contacts along one axis, but I am unsure of reliability.

Bounding Star, I would be most concerned, not about the moss' well being, but about the satellite. Moss is pretty resilient, from my experience, but being mashed and broken and shaken. Delicate wire connexions, not so much.

MBobrik, there are several schemes to how to arrange your comm network, based on orbital inclination as well as a how you want to communicate with the satellite, many short windows, fewer long windows et cetera. Full equatorial coverage gives two windows/orbit, for example, whereas full 51N coverage, say gives only one, but takes fewer stations, and there are many more. Alas, this is really a discussion that can not go very far, considering that where we have the opportunity to establish a comm station, is really not up to us.

Edited December 8, 2014 by Newt

[K^2]

You're only making it worse. Centrifuge still carries angular momentum, so that's all of the problems with rotating cube sat, plus the new problems of mechanical parts, hooking up sensors, figuring out cameras...

Keep it simple. We don't really have a problem. Omnidirectional antenna will work just fine.

[MBobrik]

An alternative would be to have a normal, still satellite with a centrifuge inside.

having a large part of the sat moving continuously during the entire mission is just asking for a mechanical failure. We should avoid that when possible.

Full equatorial coverage gives two windows/orbit, for example, whereas full 51N coverage, say gives only one, but takes fewer stations, and there are many more as well.

We won't have any continuous coverage. And because we've decided to "outsource" the antenna gain from the sat to the ground antenna we won't be able to have many of them. We will be glad if we have two comm stations and not only one.

Edited December 8, 2014 by MBobrik

[Mazon Del]

Very high chances of it seizing up. I'd try to come up with something that's either more reliable or has redundancy.

With regard to the ring of moss capsules that spin on command to present the fixed camera(s) with a new capsule to look at: I'm not sure why you think it has a high chance of seizing up. Provided that we take care that there isn't any loose bits floating around where the gears are, this should be perfectly fine. We COULD have redundancy in that the cameras also spin to face new moss. But the basic issue is that we either provide every bit of moss with its own camera, or we have a moving system where the moss/camera(s) have to rotate to provide a new view. with the ring system, you have a main ring (think of the Halo, but tiny and not lethal) with the plant capsules facing in. The outside has a gear tooth pattern on it. To keep it in place you have three gears in an equilateral triangle pattern around it. These gears would have little flanges on top and bottom so that the ring cannot slide up or down (more than say a millimeter or so, just for the sake of giving it some wiggle room). Two of these are just idlers that rotate freely, one is connected to a servo. There isn't too much point in having more than one servo power this, because chances are if one of the servos seizes up, then the other one (which would be the same model/power) 'probably' won't be able to backdrive the other servo unless the seizing was purely a controller issue (hmm, which makes me realize we may have to look into rad hardened servos...unless we want to go with motors combined with a potentiometer or other rotation system.). This system ends up having four moving parts and three points of contact. Heck, given this system I could actually create a two/three piece casing that encloses it so it is much harder for something to get in and muck with it.

Thoughts?

[rhoark]

There could be mirrors placed to project all the moss in a single frame.

[Mazon Del]

One of my original ideas (came up with Luis) was to try and have some weird mirror/lensing system so that it looks at all the moss at the same time. In the end we decided that we could probably get it set up, but if it was nudged even slightly in its fixed frame, all the images would be off. Unless we want to have a servo or two to adjust its position...in which case we might as well go with something easier. Keep in mind, we'd have to get some fairly small (and probably specialized) optics in order to fit more than like 4 moss capsules into an image, and even then we'd probably have to up the image resolution so that each quarter/image was useful. Which ends up meaning that most of the vastly increased amount data we end up sending back is wasted because of the blank (or distorted) space. Whereas if we can take multiple low-res images that are almost entirely devoted to the plant in question, we can pick up the macro features that we need to determine things. It would be nice if we could occasionally take a semi-high res image because apparently with low range microscopes its actually fairly easy to make out the internal structure of the moss.

Speaking of Luis!

I sent him an email on Friday telling him about our progress and asking him a few questions. Namely, if it would cause a problem for us to occasionally flash visible light so we don't have to go with IR, what wavelengths might be best for this if necessary, and how often we need to take an image. Chances are he is going to say that less images over a longer period of time are fine mostly because the moss in question tends to work a little like building a lego plant. You put a few pieces on every few hours, and for the most part the old pieces never change. So if we miss 2 mm of growth happening in between shots, we can still see what happened. He said he was excited to hear things were progressing and that he'd get back to me this week at some point.

If we can nail down exactly how we want the structure of the growth chamber to look, I can work at building a mockup to do that initial dark-testing that he said would allow him to apply for NASA funding to begin breeding the moss for the desired gravitometric responses.

[MrOnak]

This might be completely out of place for my very first post in this thread but why don't you all build a 1:1 scale model (cardboard will do) of a 1U CubeSat to get a feeling for how little space you actually have?

I can't help but think that most ideas, while neat, will be really difficult to shrink to a level where they all fit into the remaining space.

The design spec for CubeSats is online - page 23 is what you're looking for as of revision 13 of the spec.

Places like cubesatkit.com have CAD files available for off-the-shelf components like CPU, battery board, communication board etc. So if you want to go overboard you can 3d-print all that and get a very accurate model done - or just take measurements and put another carboard box inside the scale model to indicate occupied space. Even if you won't use the parts from cubesatkit.com yours will probably be of similar size.

[MBobrik]

With regard to the ring of moss capsules that spin on command to present the fixed camera(s) with a new capsule to look at: I'm not sure why you think it has a high chance of seizing up.Thoughts?

First, how many samples we will have ? How big is one sample going to be ? Have they to share common atmosphere, or have they to be isolated from each other ?

I too think that a moving sample ring is the simplest and most reliable solution. Moving the camera would take up additional space and would not be any more reliable.Having separate camera for each sample, if we are not talking about 4 or less samples, would be just too many cameras. Routing the image via as many mirrors as samples to a central camera would be extremely complicated both mechanically and optically... and bulky.

And we can have two redundant motors. Just connect them via ratcheted gears so that when one seizes up, the other can still move the ring in the opposite direction.

[Newt]

Mazon Del, Using servos to spin the thing also seems to pose several problems. For one, we will be pushing off of the rest of the vehicle, using it as reaction mass. We will thus need to hold it back from spinning by some other means. Additionally, the equilateral triangle design and the servo will, unless we get some way to disconnect, have a continual friction, imparting force to the main body, and slowing itself down. That will mean that much of the time we will need to have the servo running, just to prevent the centrifuge from stopping its motion. That will risk problems of heat dispersion, as we don not want baked moss and computers. So we need an exceptional servo, that can run for the duration of the primary mission with few if any pauses, and that puts out extremely little heat. Plus we will need to have an outer body control system that can deal with that force at all times, lest the satellite spin out of control, loosing radio contact and sunlight.

Taking a photograph from a non-rotating camera of such a fast rotating target also seems to be of concern. If we are rotating at 100RPM, and taking a 1/300second exposure, the target will have spun 2 degrees before the shutter closes. That is pretty unacceptable, I think. How bright will the moss be lit up? I saw something about the possibility of a basic moss experiment trial, did that happen, how did it go?

[Nicholander]

Mr0nak, good idea, but we really don't need it, we all know how big a space we're working with here, a 10 cm cube.

henryrasia, having there be a motor that spins it instead of the whole satellite spinning? That's ridiculous, as others pointed out above, it's just asking for a mechanical failure. And how much power would it consume? Way to much is the answer.

[MrOnak]

I didn't mean to imply you don't know what you're doing, I was just wondering whether you guys have a "chart" done with the dimensions of the various parts of your experiments. From the (very good) descriptions of how to position various petri dishes in front of the camera I can't help but think that all the solutions are bigger than the space you have and quite complicated. Not saying I have a better idea, I'm just pointing out that maybe working within a given space budget and then see whether you actually can fit more than a single experiment in is more fruitful than the other way around.

My point is that you don't have a 10cm cube for your experiment. You might have a 10x10x5 area to work with, minus at least 8.5mm squares in each of the corners. That's assuming you can cram the electrical power system, radio communication, main circuitry, batteries and their heaters into 10x10x5, which is tight.

edit to add same goes for weight. Do you have a rough ball park figure how much of the ~1kg weight limit is already taken by the frame, the solar panels, antennae, magnetorquers, batteries, heaters and circuitry? I.e., how heavy the experiments can be?

Edited December 8, 2014 by MrOnak

[henryrasia]

I didn't mean to imply you don't know what you're doing, I was just wondering whether you guys have a "chart" done with the dimensions of the various parts of your experiments. From the (very good) descriptions of how to position various petri dishes in front of the camera I can't help but think that all the solutions are bigger than the space you have and quite complicated. Not saying I have a better idea, I'm just pointing out that maybe working within a given space budget and then see whether you actually can fit more than a single experiment in is more fruitful than the other way around.

Hey friend, no sweat. I think the same too. Though we're debating the general systems components for now, I definitely agree that we'll need to take into account more lowly, wordly issues such as physical size.

Unless we're willing to step up development speed, that is...

And about the centrifuge idea, it's just as good (and bad) as independent antennae/solar panels. But that is a problem that needs to be addressed, as we'll have significantly less signal strength and power if we use passive systems, unless we get really creative, in exchange for mechanical simplicity and safety. So there's a trade-off there between creativity and capabilities.