PDA

View Full Version : Designing low-cost stability chamber



scool
14-06-2010, 02:40 PM
I'm looking to modify a standard, 45 cu ft. commercial foodservice refrigerator to become a stability chamber able to maintain temps +/- 0.5 C between 10C-20C and ~80% humidity.

Based on initial tests on one unit, maintaining a 15C temperature using only the existing compressor created 30 second on / 5 minute off cycles and temperature swings +/- 1C.

I could buy a hot gas bypass stability chamber pre-made by a laboratory supply company for $12K-$15K, but am looking for a more cost effective solution, e.g. pick up a used refrigerator for $2K, add $1K in mods.

The unit needs to be able to restore temperature within a few minutes of a door opening but also need to be able to trickle cool to counteract insulation heat loss.

I've researched and evaluated dozens of different approaches and cooling technologies. Right now the most energy efficient, cost efficient, and simplest is looking like this and I'd welcome any thoughts:

- Immerse the evaporator (located in the ceiling) in an insulated tub of water turning the main refrigeration unit into a water chiller.

- Use a pump and 1/4" piping to run the water to/from a large 120.4 overclocker radiator with four 120mm variable speed fans speed regulated by a PID temperature controller.

- Place a float valve and fill line into the tub of water along with a humidistat controlled atomizer and fan to maintain humidity.

The tub of water will serve as a heat sink / reservoir of cooling power. The refrigerator thermometer will be placed into the water and temperature set to maintain it just above the dew point of the air (~5C lower) to avoid condensation.

As insulation heat loss occurs, or the door is opened, the radiator fans will adjust speed proportionally draw cooling power from the water in the reservoir. The refrigerator compressor will kick on at some reasonable cycle to maintain water reservoir temperature.

I estimate cooling power needed to offset ongoing losses at around 50-100 watts and power needed to recover from a door opening in 1-2 minutes to be about 200-300 watts.

Welcome any thoughts about how effectively this will work and/or suggestions for alternative approaches.