> Anyone on the list have experience / knowledge of using PIC 
> microcontrollers
> which may be exposed to X-ray beams?  A customer has an application for 
> use
> in radiation therapy rooms, and the electronics package will occasionally 
> be
> in the path of the therapy beam during patient exposure; due to the nature
> of the package, this is unavoidable.  The question is: will this 
> eventually
> cause the electronics - specifically the microcontrollers - to fail?  Some
> empirical testing done Saturday in a Cancer Center showed that the
> microcontrollers will eventually fail, but only after extended exposure -
> approx. 1 1/2 hours in a continuous beam, something that will not happen
> with a patient.  If the package is exposed intermittently, will the
> radiation effects be cumulative?  Tried Microchip & Google with very 
> little
> useful info uncovered.  Thanks,

I remember going to a conference a good many years ago, where one of the 
papers described using a microprocessor based system to build a robot that 
could be put under an x-ray machine to medical students how to recognise 
various items on an x-ray image. He built the dummy with appropriate 
internal parts, and had it working nicely.

Then came the day when they tried it under the x-ray. Every thing went 
swimmingly for a while, then a limb would give a spasm, and then carry on 
normally. Then another would do the same, and so it went as time passed, 
until all the limbs were just behaving in a totally random mix of spasms. He 
got the unit back to his engineering lab, and found that all the EPROMS had 
been erased by the x-rays. The solution was to blow some bipolar PROMS, 
which being non-erasable, didn't suffer the same fate.

As to how to deal with your problem, a starting point is to use some 1mm 
tantalum sheet top and bottom of the PCB. This is what we do with 
instruments on space craft, where chips have doubtful radiation performance, 
but then that is shielding from particle damage rather than x-rays. I guess 
for x-ray shielding, some similar thickness lead sheet would be the answer. 

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist