radiation hard devices?

[biff44]

so, I have an opportunity to make some space radar hardware. It is NOT technically a "space qualified" design, but more a design where I can use commercial off the shelf components OF A STYLE the will likely withstand some radiation.

It was actually pretty easy to do the microwave parts...just pick GaAs ones.

But there are voltage regulators, power detectors with op amps and some logic. What sort of chips/family of chips might handle the radiation?

I am pretty much a noob on the whole radiation thing, and could use some pointers, articles, or sources of supply. I found a few actual rad hard parts, but their cost was WAY out of line...like $400 for an op amp, $4000 for a switching regulator. I do not need the full nine yards of qualification....just the basic parts ...

 

[BradtheRad]

* What method does NASA endorse?

* Maybe you can get by with the military spec series of IC's. The TTL family's numbers are 54xx instead of 74xx.

* Advice says make a Faraday shield to protect solid state equipment from radiation. This is a metal enclosure around sensitive equipment. The enclosure has a cable running to earth soil. Even though you can't earth equipment in space, there could be a method like it.

What are usable metals? Don't know if it needs to be ferrous. Maybe lead is okay, since that is commonly used for radiation. Not likely you can send lead up into space, however.

* Vacuum tubes are reputed to be immune to EMP, so maybe they can handle radiation as you refer to. It's not likely to can put tubes in your project.

 

[zorro]

Frequently, non radiation-hardened or radiation-resistant electronic components are protected with shields (usually of tantalum or tungsten) in order to protect them from radiation in space.

 

[dick_freebird]

Frequently, non radiation-hardened or radiation-resistant electronic components are protected with shields (usually of tantalum or tungsten) in order to protect them from radiation in space.

Actually, shielding is a last resort as it's (a) dead weight at
$5-$10K/lb to orbit and (b) only an attenuator, not a barrier
to more energetic ions and gamma. Only precious instruments
like the camera will get a pass on mass used this way.

As a rule higher voltage linears like voltage regulators have
more trouble than lower voltage parts. Analog is more fussy
than digital, transistor hardness even being equal. There is
a small vendor base for hardened and tolerant parts. The
low volume, high touch production and tiny demand pool
makes for very unpleasant pricing. This motivates many to
consider upscreening of industial / automotive grade parts
or buying such dice and getting the assembly, test and quals
done by some third party. That's sometimes more trouble
than it's worth (like if you only want 10).

What you are missing in your hissing about price is that
you looked at parts with all of the testing and certs that
full up hard core high value systems demand (flowed down
from customer prime, yer Uncle). You could buy dice and
do like I said above. You could buy an industrial hermetic
(non rad hard) part and see what space flight does to it.
Vendors of rad hard parts do not generally sell a non rad
rated version or an industrial grade version because they
are hip to all of that and are determined to support their
pricing and profit. But the dice you can get before much
value (whether or not you value it) is put into it.

NASA NEPP does an annual metric butt-ton of testing on
ICs and transistors and diodes in support of all the missions
and labs. Google that and you will find a database beyond
imagining, of parts good and bad and how tested. There
used to be an ERRIC database, similar, but I do not know
if it's still around or merged or renamed or whatever. If
you could get ahold of the JPL approved parts list that
would be a good guide (provided that your interests and
theirs coincide, like no doomsday funny business).

Faraday cages are for RF radiation, will not stop the
X-ray / gamma / proton / heavy ion which "radiation
hardened" semiconductors generally refers to.

Power devices operated near rated stress levels are
often prone to destructive failure when you punch
a plasma track through the blocking junction with an
ion. There is some art to device design and specification
to prevent this failure mode. Other art to make the
power MOSFET threshold shift tolerably little over the
mission duration. Part selection matters and the odds
of you finding a good one cheap and predictable in its
attributes across date code are pretty low (but not
low enough, not to try, until the thing has to get built
and not have such a rollup probability of failure that your
launch insurance provider or your customer refuses to
sign off - then you're over the barrel staring down 6
month lead times and four digit pricing and thank you,
sir, may I have another?

Military & Aerospace Electronics is a good mag for this
stuff, civil and Mil space and weapons systems from
parts up through programs.

 

[123jack]

I'd also caution that mil spec tends to be pulse tested not long term exposure.

 

[vfone]

I'm pretty sure that radiation immunity is very important in space communications, but if you look to the schematics (and parts) that space missions used in '60 and '70, you cannot believe these electronic components survived.
And they were not using only vacuum tubes. A lot of old Germanium transistors were used by communication systems sent to the space.

https://www.hq.nasa.gov/alsj/alsj-TVDocs.html

 

[c_mitra]

And they were not using only vacuum tubes. A lot of old Germanium transistors were used by communication systems sent to the space.

With hindsight, people were extremely worried about space and we thought it very harsh and unhospitable. Now that people have survived extended period in space, we know better.

In a positive way, miniaturization of electronic parts and circuits have effectively reduced the exposure cross-section of devices; a good old vacuum tube is actually more susceptible.

Shielding need not be stone round the neck; a drop of a paint (containing Pb or Ta) on a critical area is often necessary and sufficient.

What is essential is a sane assessment of the vulnerability of different parts and the consequences and building in redundancy in design.

 

[biff44]

how do you guys feel about this "Graded Z" radiation shielding, or adding a plastic sheet of HDPE to the back of a thicker aluminum cover? I read thru some papers about it, but have a little trouble figuring out if it really helps over a simple thick aluminum cover.

Also, for a HDPE plastic sheet, what exactly do they mean when they call out "5 g/cm²" ? Does that simply mean a sheet of plastic thick enough so that a square cm of the stuff equals 5 grams?

 

[dick_freebird]

Graded or compound-Z shields may be of benefit in some
environments where (say) gamma radiation converts to a
higher flux of lower energy (but still ionizing) electrons in
a high-Z compact shield; a low-Z afterlayer will catch the
electrons before they become a nuisance (unlike, say, the
gold plated inside face of a hermetic package's lid, which
will spray electrons right onto the die).

Lightweight plastic films may be for neutron, where momentum
conservation favors hydrogen-bearing material (neutrons may
"bounce off" high Z nuclei and keep almost all of their energy,
while a series of neutronroton (H) interactions will slow the
neutron down 50% each collision). Maybe the same applies to
protons in the space environment.

Both of these approaches' utility would depend on the specific
radiation exposure type (or salad). I'd go back to those papers
and see what they said about that aspect.

 

[pjmelect]

I had a log amp accidentally exposed to very high levels of radiation from a atomic pile. The components that failed were a FET with increased gate leakage and the log diodes due to the forward voltage drop changing. The other components transistors cmos logic etc appeared to be unaffected.

 

[biff44]

this is for a low earth orbit application of a few years duration. I do not know what that means for the type of radiation it will encounter, maybe you can comment.

 

[dick_freebird]

https://llis.nasa.gov/lesson/824

indicates that there is significant variation in radiation dose
even within the "LEO" classification. It's a bit long winded but
declares a range of TID (total ionizing dose) to be expected
depending on some orbital details.

Finding TID-tolerant parts is not difficult, only expensive in
one way or another. Finding a whole system's worth of
single event latchup immune ones, besides, is perhaps as
difficult and ensuring that you do not have a SEFI problem
at the chip or at higher levels of assembly, is a problem
often left for the student. For example, if you have a RAM
based FPGA and one bit flips, are you ever going to get
back to baseline? If your attitude control processor
executes a corrupt program word and starts to tumble the
body, will you ever even have the chance to assert a
ground control reset? Wheels within wheels and one goes
flat....

 

[123jack]

For low earth orbit it may also be a big consideration as to what that orbit is.
At certain places around the globe (thinking southern americas here...) even
existing nasa equipment is suffering issues that have to be dealt with.
There may be other "hot spots" to consider also.
It may be worth a search on the nasa web site about this.

 

[biff44]

I finally got some more info. 350 km height 78 to 90 degrees. So it DOES look like I just miss the bad radiation zones.

Anyone care to comment on what types and levels of radiation I should design my electronics to withstand?

 

[c_mitra]

how do you guys feel about this "Graded Z" radiation shielding, or adding a plastic sheet of HDPE to the back of a thicker aluminum cover? I read thru some papers about it, but have a little trouble figuring out if it really helps over a simple thick aluminum cover.

It is the cosmic rays that produce a shower of high energy particles when they met a heavy nucleus. Neutrons can cause structural damage but not much of malfunction to speak. The shower can be attenuated by alternate layers of heavy and light materials.

Also, for a HDPE plastic sheet, what exactly do they mean when they call out "5 g/cm²" ? Does that simply mean a sheet of plastic thick enough so that a square cm of the stuff equals 5 grams?

I have no idea but if your description is correct, it will be more than 5cm thick.

 

[D.A.(Tony)Stewart]

As I recall space qualified parts required X-Ray bombardment to zap any impurities that can cause partial discharge when energized by gamma rays. like an internal ESD... Then X-Ray pre cap inspection for damage or contaminants prior to encapsulation and ATE. Thus the yield can be quite small and the process is expensive.

Commercial parts are an unknown. All boards must be shielded with cast alum. casing with lead alloy EMI foil tape for added protection along the seams. We used these methods for rockets up to 500 miles above the earth.

The most fit humans can survive but suffer with osteoporosis with significant bone loss for lack of gravity, which is minimized with extra exercise, not to mention the other side effects of adjusting to gravity when they return. Nausea is a common side effect which is also a side-effect of radiation sickness. Immune systems must be in peak performance.