Radiation and PLC

IanM8040 said:

sure!

Look up the story of a lady who magically received a very specific number extra extra votes in an election, 4096. equal to 2^12.... eventually diagnosed as a SEE*

https://www.youtube.com/watch?v=AaZ_RSt0KP8 Good video

Click to expand...

So that’s how Biden got elected

I was wondering about that

Oh man, did y'all see that VT100?

Engineerj22 said:

Uh, where are you that this is happening?

This is too specific to be a wild thought.

But yes if the radiation is high enough I can see PLCs dying quickly

Click to expand...

He's in New Mexico. I grew up there, however don't tell anyone. He's probably working at Sandia National Laboratories. In Weapons Testing.

Magnetic fields, not radiation

Fermi Labs called one day wondering why a 'smart' pressure transmitter lost its memory every time they ran a test. I drove over and the guy took me up to the 3rd or 4th floor balcony overlooking the 'test' area. Opposite the balcony was an electromagnet several times the size of a mansion, its height reached to the roof.

He pointed to the lower left corner and said that the transmitter was down at the floor level at the oorner of the magnet.

The area was plastered with warning signs about the prohibition of any metals, particularly ferrous metal, on your person during tests because the magnetic field would attract the object.

The guy was quite disappointed that I wouldn't involve the manufacturer in 'testing' the transmitter for operation in dense magnetic fields, when his application could not be reasonably duplicated. His was the 'test' site and the result was that the memory was wiped every time the magnet fired up. He asked how to shield it to work in that environment and I passed, it wasn't my job to tell him how to shield it - they're the ones working with massive magnets on a day-to-day basis.

My client was a water district with a remote pump station out on the Olympic Peninsula, controlled by an SLC-5/03 with a radio modem. Every few months the controller would fault and they would arrive to find a blinking red light and the memory corrupted. A power cycle always reloaded from EEPROM successfully, but a watchdog relay wasn't considered an acceptable solution.

We replaced the controller twice, and spent a lot of time unproductively pointing fingers back and forth at the modem maker. The failures weren't correlated with lightning storms, or with higher rates of packet loss on the radio network: the controller just gave up the ghost at random.

The water district's technician was out there one sunny afternoon, the sort of day when you can see every peak and valley of the Olympic range from Seattle. Perfect weather for sailing or flying, and not a bad time to take a Company pickup truck out to the woods with a sack lunch to take a look at a troublesome system.

He had the door of the control cabinet open when an EA-6B Prowler from NAS Whidbey Island came over the ridge, banking hard and low enough that he could read the pilot's name stenciled on the ****pit. By the time he recovered from the surprise, he saw the SLC was faulted again.

The Navy, of course, isn't going to tell you when they fly electronic warfare aircraft with their jamming radars painting the terrain. But there's no hiding a Prowler once she's overhead.

danw said:

Fermi Labs called one day wondering why a 'smart' pressure transmitter lost its memory every time they ran a test. I drove over and the guy took me up to the 3rd or 4th floor balcony overlooking the 'test' area. Opposite the balcony was an electromagnet several times the size of a mansion, its height reached to the roof.

He pointed to the lower left corner and said that the transmitter was down at the floor level at the oorner of the magnet.

The area was plastered with warning signs about the prohibition of any metals, particularly ferrous metal, on your person during tests because the magnetic field would attract the object.

The guy was quite disappointed that I wouldn't involve the manufacturer in 'testing' the transmitter for operation in dense magnetic fields, when his application could not be reasonably duplicated. His was the 'test' site and the result was that the memory was wiped every time the magnet fired up. He asked how to shield it to work in that environment and I passed, it wasn't my job to tell him how to shield it - they're the ones working with massive magnets on a day-to-day basis.

Click to expand...

Nickel will do a pretty good job of shielding..

I had an old prof that worked on satellites before retiring to teach and one task was radiation effects. They found it only impacted the electronics they were using when they were powered up and that after a certain amount of time on the shelf they would restore themselves to operating condition. The solution was installing multiples and rotating which ones were on. No idea if that was what was actually implemented, but it was interesting.

40mpg said:

This is a wild thought: Have anyone experienced PLCs(ControlLogix) experiencing unrecoverable fault (solid red OK light) due to radiation bombardment, lets say neutron. And lets say the number of PLCs experiencing the unrecoverable fault is high in the range of 10+ in a short time?

Click to expand...

More likely to be a generic EMC problem.
The usual suspects:
Noise.
Poor ground - of the entire plant, not just individual items.
Poor or missing equipotential bonding.
Poor or missing screening of noise sources.
Poor or missing screening of networking cables.

40mpg said:

Hello,


This is a wild thought: Have anyone experienced PLCs(ControlLogix) experiencing unrecoverable fault (solid red OK light) due to radiation bombardment, lets say neutron. And lets say the number of PLCs experiencing the unrecoverable fault is high in the range of 10+ in a short time?


Is this possible?

Click to expand...

Unlikely, almost impossible that radiation would cause electronics to fail in such a way.

You might get some problems, IE a bit turning on and off, but it wouldn't kill the whole CPU.

If you're talking about putting a PLC inside a spent fuel pond, then yeah I could see it being a problem.

I once worked on a project entailing Electron Beam Guns being used to cut through the titanium hull of a decommissioned Russian submarine (sold for scrap). Every time the gun fired, some piece of nearby electronics would fail somewhere. We had an EE dedicated to finding out how to properly shield everything, it was a long project for him but he eventually came up with a reliable system. I don't know what he did though because it was all hush-hush proprietary.

Ex-nuke here.
I doubt one neutron would cause damage.
Our old RMC100 was used to move cargo containers. The cargo containers were to be scanned with a neutron beam to look for contraband. Just for jollies they put our old RMC100 in the electron beam and it survived a quick scan. The RMC100 used much older chips, not the new 10 nano-meter chips they have now so the RMC100 chips were pretty robust.


I like jraef's story. Delta once made power supplies and controllers for titanium electron beam furnaces. The most powerful one was 1M eV.
The Delta president before me liked to Intel CPU. One time we got a call from a customer saying they had an arc down. It killed the CPU board and erased the e-proms. The engineer flew up from Nevada to get a quick replacement. My current business partner and I scrapped the idea of using CPUs and software for the controllers as they weren't robust enough. Our new design didn't have a CPU or software. Everything was burned into prom. Yes, this was long ago. A PLC could select the pattern, amplitude and offset for the electron beam. The controller board even corrected for angles using e-proms. This design was robust. They never failed. No CPU, no software.


What was funny is that the 1M ev electron beam guns came from east Germany. The cold war was still on at the time.


I know that electrical surges or fields can wipe out electronics. I seriously doubt a neutron or two will kill a PLC.

NetNathan said:

LOL..
That is like looking for the "queer" electrons that blow fuses...

Click to expand...

Those are lucky fuses