Solar panels and radar in Earth's orbit

[guss]

Hello, I am planning to design, and maybe build, a CubeSat, or a satellite that is 10cm x 10cm x 10cm, and weighs less than about 1.5 kg. I am looking to have 6 solar panels on the cube, one on each face, and also have a full 360º in all directions radar, along with a GPS and some radio components for communication.

The primary purpose of the satellite will be to map space junk, or objects in Earth's orbit ranging from about a third of a cm to 30 cm or more. I would like to detect this "junk" from as far away, and as accurately as possible. There isn't much "junk" out there, and I probably will get very few readings even if it can pick up objects within a 10 mile radius, maybe a few readings per day. To map the junk, I would have the surrounding area scanned every 5 to 10 seconds, and whenever it picks up an object. Once the data is transferred back to me, it's easy to compute the velocity of that object (based on the two position measurements, and the velocity of my satellite), and therefore I should be able to map that piece of "junk" accurately for a while.

My plan is to have 6 radar sensors/receivers, one in each corner, but I don't know if this is practical, because I don't know if the receivers are big enough. If it went into orbit, the rotation of the cube would be difficult to control, so using the rotation to help with the radar will not work.

How would I get radar to be extremely accurate so it could pick up small objects at distance? How accurate could I get this radar on this small of a device(keep in mind it's a vacuum)? I could put something in place of some of the solar panels on the faces on the cube for better radar detection.

One more question, where can I find small solar panels that are good to use outside of the atmosphere? Will any solar panel work with nearly the same efficiency as any specifically built for space? The main concern is that the panel cannot be airtight, and I am not sure if most panels used on the ground are airtight. Do you think a solar array like this that produces about 4 Watts on average will be able to run the radar and GPS consistently, and run the radio communicator once every day or so? I'll do the exact calculations out myself when I know what's going on, but a guess would be helpful.

Final question, what is the maximum altitude GPS works accurately at? For some reason I could not find this information.

Thank you!

 

[JaredJames]

There isn't much "junk" out there

Just a note on the above, there is a lot of junk out there and it currently causes a lot of problems.

I'm not sure if you'll get a radar system on that sort of scale and with such low power available.

 

[guss]

Just a note on the above, there is a lot of junk out there and it currently causes a lot of problems.

I'm not sure if you'll get a radar system on that sort of scale and with such low power available.

There is a lot of junk, but at the scale of Earth's orbit it is minimal. There are currently ~600000 objects 1cm or greater. Not much when you consider the vast nothingness that's up there, is all I'm saying. I realize it causes problems, which is part of the reason I think this could be a useful idea. Sorry for the confusion.

What if I charged up a battery for a few hours, used the battery's energy for radar for half an hour, then just repeated? That could be more efficient. The goal is obviously just to get the most coverage possible, so a 5 mile range for 15 minutes, then off for 3 hours would be better than a .5 mile range on all the time. It would also be useful to take into account the elevations that should be covered, since the orbit would probably be mostly circular.

 

[JaredJames]

The Earth's orbit is irrelevant - the junk surrounds the planet all the time.

Do you have a radar system that small?

 

[guss]

The Earth's orbit is irrelevant - the junk surrounds the planet all the time.

Do you have a radar system that small?

The area surrounding the planet is so large, though, is what I'm saying. I get what you're saying, I think that it's just my poor wording, sorry.

No, I'd like to find one though.

 

[JaredJames]

No, I'd like to find one though.

That may be tricky.

You can certainly get sensors for that scale, but their range is rather limited (we're talking metres).

I'm not aware of anything on the level you're talking for that size and power.

 

[Drakkith]

Just wanted to throw this in.

Currently a LOT of space debris is tracked OPTICALLY by several stations around the world. I happened to be deployed to Diego Garcia, a small island in the British Indian Ocean Territories, which had one of these stations on it.

It was really really cool. We visited and saw the telescopes they used and various other systems and such. They showed us a screen that showed the REAL TIME position of everything around Earth that they were tracking. While the dots and such weren't to scale, it did look VERY crowded.

 

[guss]

It seems very odd to me that there is no effective radar this small. I was thinking I could ditch the sensors on the corners, and just convert one or a few of the sides into radar receivers.

It seems it would be be much more effective just to use an HD or thermal camera or something then use some software to map the objects.

 

[Drakkith]

It seems very odd to me that there is no effective radar this small. I was thinking I could ditch the sensors on the corners, and just convert one or a few of the sides into radar receivers.

It seems it would be be much more effective just to use an HD or thermal camera or something then use some software to map the objects.

Well, I'm not sure, but the wavelength of the radio waves you are using could drastically effect the size of the radar system. Radio waves, being very long wavelengths, typically require much larger antennas than shorter wavelengths.

I could see this resulting in a "minimum" size of the system depending on the range you needed. But I'm not sure.

 

[guss]

I could use a shorter wavelength and it may work well too, though. Wouldn't the fact that this will be operating if a vacuum have a large benefit if conventional radar was used?

 

[Drakkith]

I could use a shorter wavelength and it may work well too, though. Wouldn't the fact that this will be operating if a vacuum have a large benefit if conventional radar was used?

I don't think so. I THINK your major losses in range for radar on Earth aren't due to it being absorbed, it is the range. Since you lose strength of signal with the square of the distance, to get a longer range you must use a much greater amount of power. You would get SOME increase in range and line of sight, especially because you don't have to deal with the curvature of the Earth's surface, but I don't think it is THAT much.

 

[JaredJames]

My whole point so far has revolved around the power requirements for radar.

You're not talking a small amount. Everywhere I see rates given, even for small systems, you're looking at kW.

 

[Drakkith]

My whole point so far has revolved around the power requirements for radar.

You're not talking a small amount. Everywhere I see rates given, even for small systems, you're looking at kW.

That as well. Like I said in my post you will lose signal strength quickly. So to double your range you need to quadruple your power I believe.

Edit: Please realize we aren't experts, so you will need to do some research on this to be sure.

 

[guss]

I will. They key will be finding the golden ratio of the range of the radar to time it is on, so I can get the most coverage. Even if it would be in an always-on state and had the range of 100 meters but could pick up very small objects, it could still pick up some good data if it was placed in high space-junk traffic area.

What about GPS and solar panels up there?

 

[Drakkith]

What about GPS and solar panels up there?

What about them?

 

[JaredJames]

You're asking a lot from a 10cm³ box.

 

[guss]

Haha, I know. All theoretical though.

What altitude does GPS work up to? Do conventional solar panels work just as well in space or should I get ones specifically designed for space?

 

[PajoTheDwarf]

Lidar would be less trouble I think.

A high quality focus at 250mW to 500mW should reach 10 miles through space and with sensitive detectors you would spot reflections off of brighter or larger objects at that range.

If you used phased array detection your resolution would be more than you need and a phased array substitutes much of the mechanical design and circuitry complexity for computational complexity which is probably going to mean it will be easier and cheaper to shoestring and possibly repeat and improve.

The multiple detection elements used in a phased array also provides redundant reliability. Loss of a few detectors will lower your resolution but the system will still work fine otherwise.

You will want to try and keep the entire system down to around 5 Watts I think to allow dark side operation on reserve power and fast recharge on dayside.

You really will want to control your orientation so build in some gyroscopes which you might want to also use as part of your reserve power system - spin them up when you have good dayside power from your solar panels - and drain power from them on darkside transits.

The main problem from gyroscopes that you can build or afford is the vibration. A slightly shaky laser will swing across a wide area at its extreme range and target focus will be degraded so that you will paint objects weakly.

The easiest way to keep that from being an issue is to live with it. Isolate the Gyroscopes. Try not to physically connect them to the sensor part of the sattellite.

For the laser to not become confused by vibrations, stutter it worse than any possible vibration effects. Don't use continuous beam - deliver a pulsed beam. Some lasers are rated so that a 250 mW rating will allow maybe up to 2.5W pulses at something less than 10% on cycle with each pulse width limited to only a few microseconds.

Phased Array will work very well with that type of much stronger pulsed signal and if the sensor array has a rigid enough frame with nearly neutral vibration interaction - you will resolve perfectly in spite of the vibration in other parts of the sattelite.

 

[guss]

Lidar would be less trouble I think.

A high quality focus at 250mW to 500mW should reach 10 miles through space and with sensitive detectors you would spot reflections off of brighter or larger objects at that range.

If you used phased array detection your resolution would be more than you need and a phased array substitutes much of the mechanical design and circuitry complexity for computational complexity which is probably going to mean it will be easier and cheaper to shoestring and possibly repeat and improve.

The multiple detection elements used in a phased array also provides redundant reliability. Loss of a few detectors will lower your resolution but the system will still work fine otherwise.

You will want to try and keep the entire system down to around 5 Watts I think to allow dark side operation on reserve power and fast recharge on dayside.

You really will want to control your orientation so build in some gyroscopes which you might want to also use as part of your reserve power system - spin them up when you have good dayside power from your solar panels - and drain power from them on darkside transits.

The main problem from gyroscopes that you can build or afford is the vibration. A slightly shaky laser will swing across a wide area at its extreme range and target focus will be degraded so that you will paint objects weakly.

The easiest way to keep that from being an issue is to live with it. Isolate the Gyroscopes. Try not to physically connect them to the sensor part of the sattellite.

For the laser to not become confused by vibrations, stutter it worse than any possible vibration effects. Don't use continuous beam - deliver a pulsed beam. Some lasers are rated so that a 250 mW rating will allow maybe up to 2.5W pulses at something less than 10% on cycle with each pulse width limited to only a few microseconds.

Phased Array will work very well with that type of much stronger pulsed signal and if the sensor array has a rigid enough frame with nearly neutral vibration interaction - you will resolve perfectly in spite of the vibration in other parts of the sattelite.

Thanks! I have never heard of lidar. So, basically shine a ~350mw laser, that has, say, a 450nm wavelength around, and detect when light within, say, 400nm-500nm is is reflected back? What would be the ideal wavelength for the laser and ideal wavelength range for the received (considering light emitted from stars, etc)? It sounds like this could be done with small components as well, and the sky could be scanned very quickly, depending on if the lasers/received were somehow pivoted by a motor or they just spun normally with the satellite. How many of these components could I fit within the power budget? Are you thinking power down the detectors when the sun is gone, then power them up when the sun is out of sight for optimum efficiency?

(damn, lots of questions, sorry)

I think that a gyroscope uses too much power and may be too large to fit, I very well could be wrong though. I think it would be effective enough just to put 3 to 6 panels on it. I was thinking of a sun-synchronous orbit, but that would be poor planning because it would cost more to launch something into orbit that specifically, unless there was some luck. Not to mention, the sun's EMR could get in the way of the detectors.

 

[PajoTheDwarf]

You would do better to put a laser on each side of the satellite and rotate the whole assembly. Again, Gyroscopes would be a major part of this type of control improvement.

Now that also begs is it better to make one side the main sensor side and get your 360 coverage from the satellite rotation? A small satellite is pretty nimble so you could rotate it about 120rpm which is a fast enough sweep rate.

Now I will go to your first question and tie this all together. I would use several lasers of different wavelengths. One on each side probably.They would be able to detect more items more clearly and give you a more identifiable signature from their comparative frequency absorption/reflectance. To pick which specific frequencies you would want is a more complicated question than I can answer without research or experiments, but if you picked a few colors it would likely cover most requirements.

Also by having several lasers as in one on each side of a rotating shell, you will get a better 360 coverage. Rapidly pulsing each of 4 directional lasers in turn will give you almost 4 times the effective sweep rate of the rotational speed of the shell. So from 120rpm you can get 480 angular sample slices per minute.

 

[Mech_Engineer]

Some quotes from a conversation in another forum (http://www.gpsinformation.org/forum/viewtopic.php?f=20&t=7506):

Being at 20,000 metres with a civil GPS is rather being a bit out of your own world and obviously up to no good. Similar with velocity rectrictions, like what are you trying to do?

No it has nothing to do with any of your theories, the manufacturers have to comply with the operational design specifications imposed by the "owners" of GPS. GPS will operate out to 3000km, the space shuttle in fact uses GPS.

My understanding is that GPS manufacturers are required to comply with the ITAR limits (International Traffic in Arms Regulations) [emphasis added]. Consumer units are restricted by design to operate only at elevations and velocities that will meet all practical non-military applications, but would prevent you from using a consumer GPSr to guide your own home-built guided missile, for example. Military GPSrs will certainly work at much higher elevations and velocities than consumer GPSrs.

http://www.fas.org/spp/starwars/offdocs/itar/p121.htm

So some GPS units may have restrictions and others might not. My senior design project launched on a high-altitude weather balloon; the GPS-based locating system on it was good to over 100,000ft.

 

[guss]

Thanks, useful GPS information. What about solar panels in space, like can conventional solar panels be used or is it best to used ones designed for use on satellites? For some reason I am having a really difficult time finding this information.

You would do better to put a laser on each side of the satellite and rotate the whole assembly. Again, Gyroscopes would be a major part of this type of control improvement.

Now that also begs is it better to make one side the main sensor side and get your 360 coverage from the satellite rotation? A small satellite is pretty nimble so you could rotate it about 120rpm which is a fast enough sweep rate.

Now I will go to your first question and tie this all together. I would use several lasers of different wavelengths. One on each side probably.They would be able to detect more items more clearly and give you a more identifiable signature from their comparative frequency absorption/reflectance. To pick which specific frequencies you would want is a more complicated question than I can answer without research or experiments, but if you picked a few colors it would likely cover most requirements.

Also by having several lasers as in one on each side of a rotating shell, you will get a better 360 coverage. Rapidly pulsing each of 4 directional lasers in turn will give you almost 4 times the effective sweep rate of the rotational speed of the shell. So from 120rpm you can get 480 angular sample slices per minute.

I have thought about this, and I think I have a good solution. There would be 2 lidar lasers and sensors, on opposite corners. The detector and lens for these would lie right next to each laser. I think for simplicity sake, each laser would have the same wavelength and sensor range.

There would be 2 operating modes. The default mode would be the normal scanning mode. In this mode, I would use a control moment gyroscope to position and rotate the cube so that the lasers spun about parallel to Earth's surface (but this could always be adjusted). This would give me two detectors basically scanning the seemingly 2 dimensional space for 5-10 miles around the satellite (because the lasers are only detecting objects in 2 dimensions).

This brings us into mode 2. Once it picked up an object (assuming it could figure out the distance that object is away), it would re-adjust the control moment gyroscope to scan the area where the object would be at a certain point of time more slowly, so it had a higher chance to get a second reading. Once 2 position readings are acquired, it could calculate the velocity. It would then return to normal operating mode. If it can't get a second reading after, say, 4 minutes, it returns to normal operating mode anyway.

(Unless the lidar sensors could somehow track velocity more easily, then the second operating mode wouldn't have to be there)

The time that it is in operation is dependent on the energy it has left in its battery. It judges if it should be in operating mode or powered down by detecting if it has enough energy to stay in the second mode for 4 minutes or whatever.

I think this is at least a decent plan so far. It can obviously always be improved upon. Thanks for the great help so far everyone, please keep sharing information.

 

[PajoTheDwarf]

Two sides would be plenty.

Doppler shift maximum averaging will give you a good enough indication of speed differentials.

Basically average the approach and departure maximums and that will cancel out the speed of your sattelite. Your doppler shift should diverge farthest when the object is entering and leaving sensor range. If it

Basically you need sensors to be sensitive enough to frequency to spot the change in frequency from redshift and blueshift in the reflection.

The same thing is done with lidars used by police catching speeders. I mention that as an idea of how you might be able to pick up the technology off the shelf. Or it is an idea of systems that you might copy or adapt.

Also I thought of one problem with Lidar. Light pollution. Looking at the sun for example will mess up your readings.

 

[guss]

Two sides would be plenty.

Doppler shift maximum averaging will give you a good enough indication of speed differentials.

Basically average the approach and departure maximums and that will cancel out the speed of your sattelite. Your doppler shift should diverge farthest when the object is entering and leaving sensor range. If it

Basically you need sensors to be sensitive enough to frequency to spot the change in frequency from redshift and blueshift in the reflection.

The same thing is done with lidars used by police catching speeders. I mention that as an idea of how you might be able to pick up the technology off the shelf. Or it is an idea of systems that you might copy or adapt.

Also I thought of one problem with Lidar. Light pollution. Looking at the sun for example will mess up your readings.

I think that as long as I don't aim the detectors at the sun, I should be good as far as light pollution. I could also cross-check the information each time it gets a positive reading, to see if that part was getting positive readings before the light from the laser hit it.

Can lidar be used to detect direction very accurately, though? What if something was going perpendicular to the direction of the laser? It needs to be very precise.