Experimental Physics and
Industrial Control System

Ryan Pierce <[email protected]> · Sun, 06 Sep 2015 17:39:13 -0500

Off topic warning....

A while back, I posted about work I was doing to use EPICS for computer

controlled beer brewing. I recently heard from Ralph Lange that at the

last EPICS conference in Lansing, this project was informally mentioned,

so I thought I'd give an update.

The beer brewing project has been making slow progress. After welding

together the stand and assembling the burners, pilot lights, and

actuated propane valves, I learned an important lesson in

thermodynamics: 100,000 BTUs is a lot of heat, I didn't plan for the

fact that not all of it would be transferred into my brew kettles, and

it had to go somewhere. This became painfully obvious when I saw the

steel of the brew stand discolor and it looked like the hardware on the

kettles was about to melt. So I've been doing ongoing re-engineering

with a friend, and we are working on a system of shields to direct the

waste heat where it won't cause damage. Now I think I'm getting closer

to the point where the physical system can actually work, at least for

one of the three kettles.

The control box is nearly complete. But I need to add an additional

panel for all the jacks that will connect this to the hardware and

instrumentation. Rather than do precision drilling of 60+ holes for the

jacks and mounting screws on a drill press, I'm instead working at a

local hackerspace to repair a Bridgeport CNC mill with a Heidenhain

controller. In theory, this should make short work of my control panel.

I've successfully been able to use CSS BOY on a laptop to communicate

with a SoftIOC running on a Raspberry Pi within the box, which

communicates via asyn with an Arduino Mega running custom software I'm

adapting from Peter Jemian's cmd_response code. I still have a lot of

code to write before I'm ready to brew with this, and I've been focusing

more on the hardware issues lately. But I can make relays open and

close, and display their status, so the general concept is working. I'm

also inspired by the WebOPI work I've seen at BESSY; ideally I'd like to

use it to control the brewery via a web browser on an iPad.

This past month has involved quite a lot of physics-related travels that

seem to circle back to EPICS. My girlfriend runs the Fab Lab at the

University of Chicago's innovation hub, the Chicago Innovation Exchange.

Back in 2013, she and I traveled with another friend from Germany to the

Chernobyl Exclusion Zone. Our friend came to visit us in Chicago, and

due to CIE's connections with national labs, she arranged tours of

Argonne (the central machine shop, glass blowing, optics shop, APS,

nuclear museum, and ATLAS) and Fermilab. Of course we also had to visit

the site of CP-1 (we live a few blocks away), hike around Site A, and

have a picnic on top of Plot M. (Note: despite surveying with multiple

meters, the most radioactive thing we could detect there was by far the

granite in the monument itself.)

Our next stop was Berlin. I owe massive thanks to Ralph Lange and Thomas

Birke. Although Ralph, who I originally contacted about WebOPI, was no

longer a resident of Berlin, he put me in touch with Thomas who gave us

an incredible tour of BESSY and helped us get access to tour BER-II. I'm

especially grateful to him for answering all my questions about the

control systems and showing me in detail how EPICS was used. Given how

peripherally involved I am with EPICS, I am very pleasantly surprised

not only with the help I've received through this list, but also that

this has translated into international hospitality. Again, thanks.

We then went on to Chernobyl. While we did get access to the Reactor #2

control room and Reactor #3 circulation pump halls, I didn't see EPICS

running anywhere. :-)

While I was in Berlin, I met another collector of radiation detectors,

and we did some trading, resulting in me returning back to the US with a

3"x3" NaI(Tl) scintillator, an alpha spectrometer, and a rather large

HPGe (which has barely usable cryostat vacuum; I was told I can use it

in low humidity but I will eventually need to pump it out again) that

only barely fit in my luggage. (This also raised a lot of eyebrows at

the airport.) Of course this meant I had to purchase my first NIM crate.

I found a deal on eBay for a crate that includes an Ortec 659 5kV bias

supply and Canberra 2020 spectroscopy amp. Now while I already have an

MCA (SpecTech UCS-20) that I intended to use for all this, the crate

also included a Canberra 8075 ADC and 556 AIM. This really interests me,

since it seems like a far better MCA system than what I have, and I

think I can integrate these to handle pileup rejection and more

accurately measure live time. Doing more research, I found I could use

Canberra's Genie2k software, which requires a difficult to find dongle.

Or... I could use EPICS. Again, it all comes back to EPICS.

Regards,
Ryan

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System