RF design for a space application

[Anton89]

Hello everyone,
I have a question regarding the RF design for a space application. I have to design a transceiver which will be mounted on a satellite. I designed the system and characterized each component (gain, noise figure, etc.). I chose a supereterodyne architecture for RF front-end with digital acquisition (the software design radio).
Now, my problem is this: I need to find the components that are radiation hardness. These components shall be surface mounted (for example, MMIC components).
I spent the last two weeks to look for these components, but did not find anything.
I want to ask for you: do you know any manufacturers that make these kind of components? Alternatively, how can I complete my design? I can't use commercial components because I don't know if they work in the space (because I don't know if they are radiation resistant).
Help me, I'm desperate :-?

 

[chuckey]

have you read :- https://en.wikipedia.org/wiki/Radiation_hardening Might give some clues.
Frank

 

[G4BCH]

In my very limited experience of designing for space, if you are having to worry about radiation hardness and space qualification then you will be backed up by more quality & component sourcing teams than there are design engineers. Ask one of them.
Getting space qualified and or rad hardened parts can be very difficult and very expensive, expect masses of end use paperwork that needs filling in and even more quality paperwork with the parts.
If it is a low earth orbit 'cube sat' then industiral grade components seem to be used.
After all people have been living in low earth orbit and using 'industrial' grade electronics without too many problems for many years now.

 

[betwixt]

I was asked to design some Earth controlled camera steering equipment a few years ago and had to abandon the project due to lack of information and cooperation from parts manufacturers. In some cases they wouldn't even guarantee ICs wouldn't explode in a vacuum because of trapped air bubbles in the packaging.

Brian.

 

[BigBoss]

Space qualified components are manufactured by few high-tech semiconductor companies such as Triquint, TRW, Raytheon etc. They also request very strict agreement that will possibly be signed by government or research institues.
So, it's not so easy to get these components..

 

[c_mitra]

I do not know much, but the current trend is to introduce redundancy in the design and the software so that failure of a component will not affect the performance. This is more efficient because, as you have found out above, most of the modern ICs will not be available in all grades (but most will be available in industrial /military grades) that have been radiation hardened. Even an external shield (even Al) will help.

 

[niccoh]

If you are asking you probably won't be able to get these parts easily or have access to them. If you have to develop this your company then they should have all the info you need to source your parts. Or at least give recommendations where to look.

Space parts go through lots of research, qual, cost big $$$$$ and sometimes are custom made. If this isn't the first time your company has built for space applications you need to ask someone there.

And yes you are correct commercial parts won't work. They might. But they are not guaranteed in the harsh environment.

 

[c_mitra]

ICs wouldn't explode in a vacuum because of trapped air bubbles in the packaging...

It is very unlikely, rather impossible, that an IC will explode because of trapped air bubbles. At most the pressure difference will be 1 atm and the size of the air bubble will be considerably less that the size of the chip. The resulting force will be negligible in my opinion. Even if the chip temperature rises to 273C (where the chip will surely not work), the pressure difference will be 2 atm. If the air bubble is 1mm in diameter, the surface area will be about 3 mm2 and the pressure will be...

 

[vfone]

As was stated, perhaps you will not find much information from the component manufacturers in this regards.
But what I know for sure is that the big manufacturers of communication satellites, they do their intensive qualification of parts or assembled modules.
From now, I think you have to spend more time designing the radiation shields, then finding information about components, which should be done also, but to don't be a breaking point of the process.
I read on the net how some small cubesats were made by university students, and how they use electronic components that has no radiation information about them.

 

[Dan Mills]

It is a risk analysis thing, nobody really cares if a student cubesat fails after a week, so do some qualification of the complete assembly (Bake it in a vacuum chamber, then freeze it in the same, maybe blast it with some energetic alpha, beta and xrays), then just launch the thing.

When your total project costs (including launch) are measured in terms of a few tens of thousands of dollars, and your science program runs in terms of a few days to a week or so, worrying over much about using rad hard parts for the doings is an expensive pain in the arse, and the actuarial risk assessment may well say it is not worth it.

Of course if you are doing commercial work on a billion dollar comms bird, then the only thing to say is WELCOME TO HELL.

Regards, Dan.

 

[c_mitra]

Do not worry too much about radiation hardness of the semiconductor devices. People have spent considerable time in space and consumer grade electronics have survived space trips. You shall find far more difficult to replace all Al-electrolytic capacitors (they are likely to be the first victims in space journey) and the connectors (they have a bad habit of becoming loose at the first opportunity). Also your design must survive the stress of the launch (every part must be very secure)- again the bulky components are the most worrisome. If your device is going to live several years in space, you need to worry seriously about radiation hardness. Our atmosphere is not really a great protection from the high energy radiation you need to worry about in space: about 10-20% of the cosmic rays reach the surface of the earth.

 

[betwixt]

I fully concur that trapped air, which I have never encountered anyway, except possibly in windowed devices, is unlikely to cause issues. My thoughts are the best strategy for non "mission critcal" electronics is to make it entirely surface mounted, no 'wet' electrolytics and no connectors, the whole thing being flooded with a cushioning potting compound afterwards. Care would be needed to ensure there were no voids in the compound as these would be vulnerable to expansion and possible creating cracks. Radiation is going to be a problem wherever the unit is, if a cosmic particle can pass straight through the planet, a few microns of covering isn't going to stop it.

Big problems I see are in RF circuits where screened compartments are necessary and may trap air and the dielectric constant of the potting compound would upset things. For digital circuits, you need to asses whether 'soft' errors caused by instantaneous interference can be worked around or whether complete redundancy is needed.

Brian.