Experimental Physics and
Industrial Control System

Ryan Pierce <[email protected]> · Tue, 22 Dec 2015 23:12:48 -0600

Has anyone been able to run CSS BOY successfully on a Raspberry Pi?

My original plans for an EPICS-based computer controlled beer brewing

system involved a control box with no user input except for an e-stop

button. The operator interface would either be a laptop running CSS BOY,

or an iPad with a WebOPI backend. But with the significantly faster quad

core Raspberry Pi 2, and the availability of inexpensive panel mount

touchscreen displays that work with Pi, I'm wondering if I could take

advantage of some spare panel space on my control box and run CSS BOY on

the same Pi that runs my Soft IOC? I believe you can run Java and

Eclipse on a Pi, so I'm wondering if CSS BOY would work, and if it would

perform reasonably well? Has anyone had any experience with this?

In other news....

I was quite fortunate to meet with Ralph Lange and Michael Davidsaver

when they passed through Chicago on their way to an event at Argonne.

They toured the makerspace where I was constructing this system, and we

had a late night EPICS hacking session where they helped me fix some

issues I'd been having with CSS BOY, a calcout record that wasn't

updating like I thought it should, and they showed me some neat tricks

with alarms. Thanks so much!

A friend and I did some redesign and fabrication to solve the problem of

directing waste heat from the burner to keep it from damaging the frame

or the actuated valves and sensors on the kegs. This past weekend I did

my first test with the control box connected to the actual brewing

hardware. I was using CSS BOY, a Soft IOC on the Pi, StreamDevice and

asyn, and software I wrote on the Arduino Mega. For this test, I

connected one propane burner, which heats one of the kegs, and a single

DS18B20 temperature probe to the Arduino. While the test worked, I've

now got a control algorithm question about preventing overshoot that is

unrelated to EPICS.

For now, the system just has on/off control of the propane burner. (I do

have a needle valve on each burner's propane supply, and I am

considering putting stepper motors on them.) I didn't think PID would be

terribly useful with a binary control output, so I coded an extremely

simple thermostat algorithm: if temp < setpoint - deadband then turn on

the burner; if temp > setpoint then turn off the burner. The results are

here: http://elna.mackenziegems.com/20151220/20151220-2.png

I originally had a 70 F setpoint and 0.5 F deadband. Then I changed the

setpoint to 120 F. The burner cut out at 120 F but the temperature

overshot by 3 degrees F, and it took over 80 minutes to fall back down

below 119.5 F which triggered the burner. That's more than I'd like. But

from that point, everything happened as I expected. It took less than 30

seconds to heat the kettle by 0.5 F, it overshot by about 0.4 F, and

took about 25 minutes to fall back to 119.5 F. Those numbers are

perfectly acceptable.

I'm guessing the initial overshoot happened due to a number of reasons.

(Residual heat in the stainless steel base of the keg needing time to

transfer to the water? With 10+ gallons of water in the keg, convection

to equalize temperature may not be efficient? Lag time in the thermal

probe?) I'd think an ideal control algorithm should anticipate that with

a large commanded temperature change, cutting off the burner early is

necessary to prevent an overshoot, but that would be unnecessary with

just a small 0.5 degree F change. I'm guessing this is a common control

systems problem, so rather than reinvent the wheel, is there a standard

algorithm that handles this? Or can PID be used in a situation like this?

Hopefully over the holidays I'll be able to make some blog posts with

pictures of all this. Still, I'm glad I'm getting closer to being able

to brew my first batch of beer with this system. (Would that be an IOC

IPA? ;-)

Thanks,
Ryan

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System