Warning: this page describes the methods I used to use to keep
the temperature constant when brewing beer. The solution is not only not optimal: it works
quite badly under some circumstances. The current
version works much better.
The remainder of this page describes the old method I used. If you're thinking of copying
it, you should know:
The temperature control is not very good, especially during vigorous fermentations, where
the wort temperature can rise to several degrees above the ambient.
The thermostat is sensitive to dew, such as can form when moist outside air is cooled.
The symptoms are that the display goes blank and the device essentially ceases functioning.
So far I have been able to recover from this by removing the batterries and drying it out,
but it's possible it might die some day. I've since put the thermostat in a sandwich bag to
keep outside air out.
There is mains voltage inside the thermostat. This is a particular concern in view of the
humidity issue.
This is the power takeoff for the thermostat. I've rewired it internally, which probably
violates some safety standard. Each connector is wired differently: the one on the right
connects the neutral line to the thermostat, which then switches it to one of the other two
depending on the temperature. The cooling socket connects to the fridge, which is turned on to
always cool, and the other socket connects to the 60W globe at the bottom of the fridge.
The setup isn't ideal: the thermostat always turns on one of the circuits, so you have to
decide whether to heat or to cool. It would be easier if it would have a range where both
things were off, but the only way I can see to do that with this equipment is to buy two
thermostats. An alternative is Paul Sorenson's
thermostat, which does have hysteresis. On the other hand, it requires installation directly in
the fridge, and it's not clear to me how to use it to heat.
In addition, the temperature control could be better: I've seen the wort temperature rise by
as much as 6° C above the ambient temperature. Paul's thermostat won't help there directly,
though he's thinking of inserting the probe directly into the wort, which would help. The last
word on temperature control hasn't been spoken yet.
One obvious better solution would be to measure the wort temperature more directly. One
possibility involves a computer
and Quozl's Temperature
Sensor, which is available as a kit
from Ozitronics in Australia
and Carl's
Electronics in the USA. See the web page for
other sources.
