C-RAM Nuclear Apocalypse Demonstrator

Community Projects
1

I would like to build a demonstrator which will basically show that C-Ram (http://www.baesystems.com/download/BAES … RAM-NV-2Mb) will survive a nuclear apocalypse. My partner ask me to make a demonstrator which will play some music from the chip. I would like to hear from yall whether my approach below is the most optimal.

Here is my proposal: Use the GPIOs on a Raspberry Pie B+ to interface with the chip. Since the Pi already has an SD card and an audio jack, moving files on and off the C-RAM chip shouldnt be too hard.

For this project here are my priorities:

  1. Quick turn around

  2. Low effort on my part

  3. Must not have a kludgy appearance

  4. Should cost <$4000

I have a strong background in C, AVR microcontrollers (using WINAVR), and also with Arduinos. This would be my first Raspberry Pi project and my first real exposure to Linux. I do have a friend who is a linux genius so that should help.

Does my approach seem reasonable? Any alternative hardware you might recommend that would be quicker/easier?

2

Other than its radiation hardness and large power demands, it really not much different than any other NVRAM. I assume to test your nuclear apocalypse theory, you and your partner have access to a radiation source. In that case, your Raspberry Pi will be inadequate because it has no radiation hardening at all. It will fail (bit errors all the way to latching up) at levels far lower than what the C-RAM will tolerate.

Unless you have a highly focusable radiation source to isolate the C-RAM, you will need to harden everything else in your project to an even higher level to guarantee the RAM won’t be the first thing to fail.

Radiation hardening is not trivial or inexpensive.

-Bill

3

phalanx:
Radiation hardening is not trivial or inexpensive.-Bill

mforce:
4. Should cost <$4000

Lead is currently trading at about $.84/pound :smiley:

4

If you think that’s all there is to it, you should volunteer to be his project lead!

-Bill

5

Does survive mean operate through the event ? Or just work afterwards ?

6

Thanks for the responses! Actually I think it would be fabulous to for the RPi to get totally fried. That would really bring the point home. Then i could remove the chip out of the fried RPi and into a fresh RPi to show that the CRAM preserved the data.

Lets not get too caught up in the radiation aspect, for me that’s the easy part. Let me be more explicit on the feedback i am looking for:

  1. Are there any pitfalls for using geany to write a c program which will move a file onto the c-ram chip? I’m used to working with MCU which have deterministic execution timing.

  2. Are there any pitfalls for using geany to write a c program which will move a file off the c-ram chip and play it on the audio jack (probably using omxplayer)?

  3. I have noticed that there are a lot of learning resources for python based development. Should I switch to python for this project?

  4. I also just noticed that the C-RAM chip would ideally require 28 GPIOs (RPi only has 27). If i use a sound file that is <30 sec long then use only have of the memory and leave one pin disconnected. This would be just fine. It would be nice if i had a way to start and stop the playback but there are no pins left for this. Any suggestions? Another option would be to use the Beagle Bone Black with a USB sound card.

Mee_n_Mac:

Survive means work afterwords.