Cooling serving lines from main chiller

[evan]

Hello all,

We're a 10hl setup with four single FV's and a coldroom with 8 serving tanks. Coldroom is run from our glycol. Chiller has more then enough power, it'll do 15KW of cooling at -4C (51,000 BTU.) There are 600 liters of 30% glycol in the system with half of that being the buffer tank.

We're thinking of also using this chiller to run an ice water bath for our serving lines. In total we'd be cooling 200 meters of python, 8 product lines plus 2 cooling lines, 8mm (3/8") tubing. 19mm insulated. The plan is to put a copper coil into a 50 gallon plastic barrel and use that ice water in the python coolant lines. I suppose we'll need a thermostat and solenoid controlling the ice buildup on the coil too.

Questions:

-How big of a copper coil to make?
-Calculation for watts cooling/meter of copper coil?
-How much ice water flow is needed through the python cooling lines (LPM or GPM)?
-Should we use 4 cooling lines in the python instead of 2?
-Should we also place another ice bath cooler at the opposite end of the python to handle the load?


Thoughts/suggestions? Thanks in advance!

evan

[RobZamites]

Why not tie the glycol into the python? An ice-water bath would seem to be a PITA, at least in my eyes....

[lhall]

I think that using the main glycol would work, but don't run the temperature colder than 28 F or you'll freeze the beer lines.

[evan]

Why not tie the glycol into the python? An ice-water bath would seem to be a PITA, at least in my eyes....

Yup, it will be a PITA, but -4C is 24.8F. Can you say frozen? Even at 28F I'd be a little concerned, so thats why I think we need a water bath as a buffer.

[clydemule]

If you don't have constant flow through the serving lines they will freeze up if the beer flow stops. So your buffer tank idea may be a pretty good one, though as people have suggested a PITA.



Since you are pumping the ice water through the python, you maybe be better off just using a brazed plate heat exchanger on the ice water return line from the python. One set of connections would be for the ice water and the other for glycol. The flow would be this:

TANK->>Pump>>Serving Lines>>Heat Exchanger>>TANK
/ \
Glycol Supply>>S.Valve>>/ \Glycol Return

You would turn on the solenoid valve based on the signal from a TSTAT with
a sensor inside the tank.


Once you determine the flow rates of water (which I don't know), the flow and temperature of your glycol you can then have a brazed plate heat exchanger sized.

This way you are not building you copper coils, which is not hard but sizing them is kind of art more than science.

My ASCII diagram is not coming out properly! but I think you get the idea.