[tschmidlin]

Looked hard at the freezers...... but between getting one and the thermostat setup it was too much $$. They're flying off the shelf on Craigslist here.....

So, I got a Sears fridge, up/down, freezer down. This way you don't have the hump in the bottom of the fridge. Took out all the shelves and it will fit 6 cornies easily, pin or ball with plenty of head space for plumbing. Taps are going in the door, there's no risk of running into refrigerant lines in the door.

Craigslist, $50 out the door, no additional electronics required.

Jeez. No-ones giving stuff away around here like that. How many cf?

Yeah, I'd probably switch my serving fridge from the keezer to one of those if I could get one for next to nothing!  My main issue with fridges has been the freezer on top, which makes the taps too low and the tendency to push the taps open if you open the freezer when there is a handle on the faucet.

[Mikey]

The popular digital controllers, that many of us here use, have settings to adjust the deadband. Say you have it set for a +/- 2 degree deadband. That means if your setpoint is at 50F the controller will not turn on the compressor until the temperature is at 52F and then leaves it on until the temperature drops to 48F.

Be warned that terminology varies with different manufacturers. Some will call it total deadband so +/-2 is 4 degrees total. Some will not center it about the setpoint, so a 50F setpoint may not turn on until 54F and then back off at 50F. I'm using 4 degrees total deadband as an example, because that's usually a pretty safe setting to prevent short cycling. Hope this helps.

[tumarkin]

thanks, all, that does help a lot. the problem is, I've got an older analog johnson that doesn't have that capability.

[narcout]

I've been using a 1 degree differential for about 2 years now without issue. I work form home a lot, and the freezer doesn't cycle very often at all.

[slobrewer]

The other thing to consider is that depending on the design you have to lift the keg pretty high to get it over the collar on a chest freezer.  Based on the way I mounted mine that was a hassle and once in a while I'd accidentally knock open a tap while lifting the keg into place.  Not a lot of fun.

Also, you can end up with a lot of condensation in the bottom of a chest freezer.  They're simply not designed to deal with condensation, which happens at fridge temps, so you get a pool of liquid that soon goes funky.  Some people use desiccants or towels with some luck but for me it was a pretty big hassle.  I had to periodically pull out all the kegs and clean things up, which is difficult due to the height of the collar.

All of this convinced me to switch to a front door fridge like this:
http://www.katom.com/598-TBB2448GS.html?CID=GoogleBase2

Keep an eye on CraigsList for restaurants that are going out of business.  I picked up mine for $500.  I don't have to override the temperature settings at all (saving me from using a temp controller and worrying about short cycling a compressor.)

[Mikey]

That's a nice fridge.

One degree either side of setpoint is probably fine as well. Just remember to add some mass to the temperature sensor or let it sit in a glass of water. I wouldn't recommend immersing it below where the wires enter the probe.

[tschmidlin]

I agree, 1 degree is probably fine, especially when it's just sitting there being cold.  If you're getting in and out of it a lot and losing the cold air out the fridge door, then 2 degrees is probably better.

[Tim McManus]

I went with a fridge because they're easier to find and they're easier to work with.  The initial setup can be a chore, but long-term there is very little worry about.

•  The fridge will keep beer within serving temperature without any modifications.  It's a fridge.
•  Thermal temperature change from opening the door is actually less than if you opened a regular fridge.  The kegs--when full--will cool the surrounding air and the temperature of the beer won't be affected.  It's a fridge, it's designed to be opened and closed.
•  Mounting external faucets is a challenge because the door needs a good amount of modification.  Depending on the amount of faucets you want to use, the cost is on-par with buying a faucet post.
•  The freezer door on the top won't get in the way if you use the standard black handles that come with the faucets, and measure before drilling (the height of the faucets is obviously important).  Using custom or longer taps handles is a bad idea, not because of the freezer door but because of the angle required to close the faucet.  The top of the handle will hit the fridge before the faucet closes completely.
•  Fridges are easier (IMHO) to find.  I picked up a dent-n-scratch fridge for $200.  I use the freezer to store hops and other treats.

I think this is a good thread.  Lots of good discussion.  I just wanted to throw my $0.02 in from my fridge conversion experiences.

[birchvalleybrew]

Thanks to all - your remarks have been very useful!!!

bvb :

[tom]

Just remember to add some mass to the temperature sensor or let it sit in a glass of water.

I have heard that frequently, but it makes no sense to me. Could you explain it further?

You want to control the temperature of the beer. Putting the sensor into something else will only delay the precision of the "control" of it.

[Mikey]

The additional mass is just slow it down. It doesn't effect the setpoint.

[Tim McManus]

I'd argue that you'd want to keep the sensor unchanged and let it sense the ambient temperature of the air in the cooling unit (fridge or freezer).

I understand the goal is to keep the liquid in the cooling unit as close to a certain temperature as possible, but these units are designed to cool the air, not a liquid in a container.  If you put a probe in the liquid, it would be harder to control the temperature of the liquid.  The cooling unit would not turn on until the liquid exceeded the range set on the thermostat.  This would almost certainly mean the ambient air would be much warmer than the liquid.  It takes more energy to heat and cool liquid than it does a gas.

These devices work on the concept of keeping the ambient air within a certain range.  As long as the gas surrounding the liquid is maintained within a certain temperature, the temperature of the liquid should remain fairly consistent.  Unless you're directly cooling the liquid with another liquid (like glycol) there's no reason to regulate the cooling based on the temperature of the liquid or any other liquid.  It would be inefficient.

[Mikey]

It depends on how much water you have in the glass. Yes, if you dunked the probe into a keg of warm beer, the freezer would get as cold as possible before the probe actually got to setpoint. The idea is to stabilize it a little. Controlling air temperature is very difficult. Imagine trying to control the temperature of a hairdryer with a probe in front of it. I know this is stagnant air, but some of the same principles apply.

Also, it doesn't have to be water. Strap some copper to it. Anything to slow it down a little. Depending on the type of probe, you may not need it anyway.