"DoN. Nichols" wrote in message
...
On 2017-04-27, Jim Wilkins wrote:
"DoN. Nichols" wrote in message
...
On 2017-04-26, Jim Wilkins wrote:
"mike" wrote in message
news
On 4/26/2017 3:02 AM, unk wrote:
I want to cut AC power if the temperature drops below a
close-to-freezing
temp, and I want it to work even if there's no power (takes
care
of
the
power fails first then system freezes then power is restored -
bang
goes
pump motor). This needs to be accurate to about 1-2 degrees.
[ ... ]
http://www.ebay.com/bhp/capillary-thermostat
Not the same, but it reminds me of something used for
temperature controlling a circulating water bath.
http://www.philadelphiainstrument.com/thermoregulators.asp
[ ... ]
These made accurate mechanical thermostats obsolete:
https://www.amazon.com/AGPtek-Digita.../dp/B00BVWYH5A
-jsw
Agreed! I have a nice one by Omega for my heat-tread oven which
can be turned on, set for 1800 F, and overshoot by only one degree
F.
(It pauses about 3/4 of the way to temperature and stops heating for
a
bit to calculate the overshoot and compensate for it.)
But he was looking for something mechanical -- or with a minimum
of electrical, so something like what I pointed to, with a couple of
relays, could do it -- with minimal battery backup. And the
learning
curve, with all of the options which can be set in the controllers,
may
be a problem.
Enjoy,
DoN.
I don't know where to buy a cheap, accurate and reliable one. 100C
switches can be found in hot pots and rice cookers, the application
for non-adjustable low temp switches is to protect dehumidifier
evaporator coils etc from icing but
AFAIK they switch somewhat higher
than 0C to allow for tolerances and as noted have a considerable
differential. I tested a Thermocube pipe heater switch that tripped at
an outer temperature of 27F, perhaps because I cooled it faster than
it would inside an unheated house.
One of my early jobs was building custom industrial temperature
cycling ovens. When I compared the scattered readings on a 12 channel
chart recorder I found that measuring temperatures accurately was much
harder than it seems. If possible I insert an electrically isolated
thermocouple into a drilled hole in the component, its heatsink or a
thick washer under a mounting screw.
My home instruments resolve to 0.1F and show that simply approaching
the sensor throws it off by several tenths. Melting snow in a
styrofoam cup is a pretty stable 32F/0C reference but the reading of
rapidly boiling water bounces around by several degrees, on the thin
Inconel probe Omega custom-made for me (twice, to get it right) .
The thermostat in my fridge seems to hold its settings within about
one degree F. The differential is about 5 degrees, ie on at 31F, off
at 26F, where frozen food stays frosted but liquids other than water
don't freeze solid. Room temperature affects the setting somewhat so I
have to reset it for summer or winter which is why I've measured its
cycling with a Fluke 52 thermocouple meter.
I've had 12V 18A-h AGM batteries go bad in about two years, both new
ones in storage and second-hand ones that had been replaced on a
schedule. The voltage is correct but the impedance becomes high.
Charging and cycling restored some of them.
https://www.bobistheoilguy.com/forum...Number=3169459
I personally wouldn't use them in a critical application like freeze
protection unless they were located where they could easily be checked
on a schedule, meaning not in the middle of the garage ceiling. I've
been experimenting with using the HF carbon pile battery tester to
measure how much current they can actually supply when loaded down to
10V. This is quick and easy but does NOT correlate well with remaining
capacity.
http://batteryuniversity.com/learn/a...and_resistance
-jsw