Radiation hardend systems on the cheap

Question

Aside form the "buy really costly NASA stuff" solution, what can be done about running a system in high radiation environments like space? (This is mostly a thought experiment.)

Known Issues primarily amount to data errors:

1. main memory
2. cache ram and CPU internal state
3. HDD

What can be done about these?

My thoughts include

• Use EEC Ram
• Shield the CPU and cache with tungsten because it's a good conductor and IIRC a good radiation shield.
• Use RAID 5 (or RAID 6) disks with some sort of hacked setup to correct errors in place rather than just writing off the whole disk as bad.
• Use ROM disks built out of sold state HDDs with the flash chips replaced with ROMs or just a stack of CD/DVD/HDDVD drives.

---

Note: I'm not just thinking of embedded system. I'm also interested in high end system right on up to supercomputers if anyone has any info of them.

Answer 1

For "on the cheap", meaning "relatively economical", I turned my search to Amateur Radio. They fly satellites at a fraction of the cost. Put "AMSAT" in your search. I found this in an article at codec.org.

• "AMSAT satellites are generally firmware-free, lacking even a boot ROM! They use a hardware circuit, without the assistance of software, to initially load their computer RAM from a ground data stream."

• Certain RAM technologies cannot be used. Those with charge pumps, for example.

• Error-correcting systems (extra bits and CRC's)

• "They continuously "wash" the RAM by reading it, checking it against the CRC, and writing the information back to correct for radiation-induced errors. Without this activity, the computer onboard a space vehicle would not survive."

• "Many modern microprocessors would not execute reliably in the radiation environment. AMSAT uses a special radiation-hard silicon-on-sapphire version of the CDP-1802 CPU designed in the late 1970's as its main control computer. They experimented with the StrongARM SA-1100 chip" (but it doesn't state results of that experiment).

Don't forget that an "HDD" is going to have a CPU and RAM on its own controller!

Answer 2

If you're talking about a fairly accessible environment (like an accelerator end-station hall) and a fairly low impact system, you just do lots of verification and complain when it fails, then you try a reboot followed by a field circus.

If it is a little less accessible, you add redundancy and fail-over.

After that it starts to get expensive.

---

The problem with shielding is that for high energy gamma and electron progenitors a little shielding can make things worse! And neutrons are generally hard to shield. You don't even want to talk about muons.

A space environment means cosmic rays which means multi GeV protons. Yuck.