Experimental Physics and
Industrial Control System

That looks like it's the famous VAT valve melting incident on beamline X-16 at NSLS.  Probably around 1988?  Turns out that was due to a piece of debris in the valve air solenoid, not a logic problem.  Though if the beam had dumped as soon as the fault was detected it probably would have been OK.

Mark

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Michael Davidsaver
Sent: Tuesday, April 05, 2016 9:21 AM
To: [email protected]
Subject: Re: Controlling vacuum valves with EPICS

An atypical example of what "vacuum goes bad" can look like.

http://mdavidsaver.users.sourceforge.net/vac-valve.jpg

I don't remember the context of this image other than it's a vacuum gate
valve which has seen too many x-rays.

I'll echo what others have said.  Protection functions should be left to
PLCs, which are designed to be reliable and verifiable.


On 04/05/2016 07:44 AM, Mark Rivers wrote:
> Hi Florian,
>
> We use PLCs to control our valves because they are part of an Equipment Protection System (EPS).  If the vacuum goes bad somewhere the PLC automatically closes the appropriate valves.  EPICS makes valve open/close requests to the PLC, and the PLC acts on those requests only if is safe to do so.
>
> If you don't have any EPS requirements, i.e. it is fine to open/close the valves any time, then using EPICS to directly control them should be fine.  But if you do need an EPS I would use a PLC for that, not EPICS, since it might not be running when you need it.
>
> Mark
>
> ________________________________________
> From: [email protected] [[email protected]] on behalf of Florian Feldbauer [[email protected]]
> Sent: Tuesday, April 05, 2016 2:52 AM
> To: EPICS Tech-Talk
> Subject: Controlling vacuum valves with EPICS
>
> Hey all,
>
> I'm working with EPICS now for several years to develop parts of the
> Detector Control System for
> the PANDA experiment.
> For our luminosity detector we need to control a small vacuum system.
> I had a look at various talks from EPICS meetings to see, how others are
> building their control systems
> and I have a question concerning this:
> All the systems I have seen are using PLCs to control the valves of the
> system.
>
> What is the advantage of a PLC for controlling the valves?
> Why not using an EPICS IOC for direct control running on a realtime system?
>
> I never worked on vacuum systems before, and our vacuum workshop here at
> Mainz
> could not help me with this question.
>
> Many thanks in advance for the help!
>
>    Florian
>
> --
> ----------------------------------------
> | Dr. Florian Feldbauer                |
> |                                      |
> | Helmholtz-Institut Mainz /           |
> | Johannes Gutenberg-Universität Mainz |
> | Johann-Joachim-Becher-Weg 36         |
> | D-55128 Mainz                        |
> |                                      |
> | Office: SB1 / 00-213                 |
> | Phone:  (+49)6131 / 39-29608         |
> ----------------------------------------
>
>