[HH] X-ray machines & mobile electronics

[Greg London]

Ah, yes, programmable logic can be problematic.

A satellite I was working on, we had an vendor
that sold fuse-based FPGA's swear to us their device
was rad hard. We spent a year doing the entire design using their
FPGA's and a couple specialized ASIC's. When the design
was done, the vendor said 'oops', they couldn't certify
their chips to meet our radiation requirements.

Basically, their fuses might blow when exposed to radiation.

If I recall, they told us this about a month before
the project deadline was up.

Good times. Good times.

Greg


> Greg London wrote:
>> But if the device is off when the radiation hits it,
>> then you can't get any of that sort of behavior or failures.
>> If there is no power to the system, then a flip or a short or
>> whatever can't cause a current surge and burn out one of the
>> metal layers tying the gates together.
>
> I think the exception to this would be the Flash EEPROM.
>
> http://forums.parallax.com/showthread.php/98219-X-ray-tolerant-chips
>
> http://www.prutchi.com/pdf/implantable/Radiation%20Hardness%20Testing%20Implantable%20ICs.pdf
>
> But the anecdotal and scientific data seem to indicate that this is very
> low probability.
>
> The first thing I'd try doing, if you haven't done so already, is
> removing the batteries for a period of time, to fully reset the devices.
> (Either wait a while or maybe try powering them up with the battery
> removed to insure any filtering caps have discharged.)
>
> The next thing I'd try is to reflash the ROM, if that's possible with a
> bricked device.
>
>  -Tom
>
> _______________________________________________
> Hardwarehacking mailing list
> Hardwarehacking at blu.org
> http://lists.blu.org/mailman/listinfo/hardwarehacking
>


--