Glovebox Connections to Building Services

[kacarrol]

I work on a nuclear research site. As such, there are a number of gloveboxes that handle a number of different materials (plutonium, uranium etc). I am looking at different building services, such as compressed air, running into the gloveboxes. This compressed air system feeds the rest of the site.

In my mind I would feel better if a separate compressed air bottle fed the system, then if there is blow back only the compressed air bottle and the associated piping will be contaminated.

Am I being overly conservative by saying that gloveboxes should not be directly connected to this system? Do other sites connect services in through their gloveboxes? Is an inline HEPA filter and double block and bleed valve considered a reasonable enough guard?

What do you do with process water connections? Just use a double block and bleed valve?

PS I have already considered the issues with compressed air pressurizing the glovebox and water as a moderator.

Thank you for your input,
K

 

[IRstuff]

How do you expect there to be "blowback?" Pressurized systems typically have several atmosphere's worth of pressure. How would contaminated air at room pressure "blowback" against an uphill pressure gradient?

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

Of course I can. I can do anything. I can do absolutely anything. I'm an expert!

 

[kacarrol]

The compressed air system on our site is old. While there has been some recent upgrades and repairs it is frequently down (depressurized). I'm not certain of the current number of outages but I believe in the recent past we were looking at 6-12 loss of pressure incidents per year.

Therefore I would be left with a pipe that is essentially open to atmospheric pressure a couple of times per year. I realize that the glovebox should still be under slight negative pressure through this, but it's my understanding that in industrial ventilation of large facilities a slight negative pressure is not 100% effective at controlling contaminants. There is not much air flow through a glovebox, usually. Wouldn't Brownian motion be able to overcome this slight pressure gradient?

But that is why I'm asking, maybe the amount of escaping particulate is so low it's negligible and I'm being overly conservative. I was trying to see what other facilities do and what operating experience was out there on this topic.

I have read a piece of operational experience in the book "What Went Wrong - Case Histories of Process Plant Disasters...5th ed" section 34.3 for anyone who has the book. Basically there was a carbon dioxide tank, a line connected the low pressure gas from this tank to the reactor (so the tank was supposed to feed towards the reactor). There were three valves in the line, the radioactive gas from the reactor managed to leak through the valves and back into the carbon dioxide supply tank. This is why I'm concerned and wondering what other people do with their gloveboxes and process piping.

 

[racookpe1978]

Yes, your precautionary requirement is prudent.

Total isolation of every system going into (and out of) a glove box is needed. Physical is best, if not physical (flanged and blanked, disconnect, or QD isolated) and operational (valve or plug).

If for nothing else, so the glovebox is not contaminated/sprayed down, vented into, drained into, flooded by the system when the system fails or gets maintained or gets flushed or is rebuilt.