We have a standard Sears programmable thermostat on our
gas-fired furnace.  It's over 10 years old, and we have had
no problems until three days ago (Thursday).
        The thermostat has a horizontal LCD display with black
characters on a gray background.  On Thursday, this display
suddenly *dimmed*.  The entire display was dimmer, and the
lower half of the display was so dim it was almost unreadable.
        Then, a while later, the display was back to normal.
A while after that, it dimmed again.  And so on.
        This behavior has gone on for three days.  During this
period, the thermostat *itself* has continued to operate normally.
Heat is being delivered.  The cats have not frozen.
        So far, then, there is no connection between the display
circuit (which has "issues") and the main thermostat circuits.
        Has anyone had anything like this happen to them?
What is likely to happen in the near future?  Is this a stable
"partial failure", or is the thermostat itself likely to fail
at any time?
        We would really *not* like to try to install a replacement
thermostat during the coldest period in recent years.  However,
we also don't want to be without heat all of a sudden.
        What should we do?

40 responses total.

Does it run on batteries?

Often the display is the first thing to break.

        Did the characters of the LCD itself dim, or was it possibly backlit,
and the bulb dim?

Next time it happens, tap on the display, and see if it flickers. 

Chances are, it's just the display going bad. Probably due to an imperfection
in the LCD sealing.

If it is just the display, it shouldn't have any impact on the operation of
your thermostat, in terms of it turning your furnace on and off when it should
be turning it on and off. However, it could make it difficult to adjust.

As far as replacing it goes, I don't really see why you're against it.
Assuming you know which fuse to pull (or circuit breaker to trip), it should
only take about ten or fifteen minutes to replace. You'd probably have more
than enough time between heating cycles to replace it.

It's probably a do-it-yourself job, and you can find a replacement at just
about any hardware store, including Loew's and Home Depot (or whatever that
orange building on Carpenter is called now).

The thermostat is not normally battery-powered.  It runs off 24V AC current
from a transformer on the furnace itself.  The display itself is dimming;
there is no backlighting.  As to replacement of the thermostat, DTE Energy
(formerly Detroit Edison) doesn't do it.  Sears is a bunch of idiots.
Hutzel will charge $255 to replace the thermostat (new thermostat
included).
        There is a battery backup, and the "low battery" display is not
active.  The instruction manual says "If incoming power should fail,
three...batteries will maintain the stored program for approximately
one year."  So it could be that the battery backup will continue to
power the thermostat for a year.  Or, as the manual literally says,
the *program* will be maintained for a year.
        A friend of mine who claims he knows what he is doing says
that the 24V transformer only powers the thermostat, and nothing else
on the furnace.  In fact, it could be that the dimming is the result
of the battery backup coming on as the transformer intermittently
fails.  He suggeted letting the thing go on until a heating failure
results.
        Is this a good idea or a really bad idea?

The 24 volt xformer probably powers the solenoid valve for the gas, and
other furnace control circuits. Put a VOM on the circuit and see if
the voltage is maintained when the display dims. I would suspect the
display first.

  re #5:  a heating failure in Michigan in the winter can turn out to be
  very costly if pipes freeze..

The thermostat works by shorting its incoming leads when heat is
required; the current it passes opens the gas valve, but while the
voltage across the thermostat is zero its clock and display need
to run on internal power.

One of the first symptoms of a dead battery on an LCD watch is that
the display contrast goes to pot.  In an LCD thermostat which draws
power from the furnace transformer, it would have good power while
the heat was off and flaky battery power when it was on.

Very easy way to test this:  when the display is dim, manually set
the temperature down by five degrees.  This should turn off the
furnace and restore voltage to the thermostat.  If the display
contrast goes back to normal, that was probably the problem.  To
confirm, set the temperature up two degrees above the old setting.
If the display dims again, you've nailed it.

Or just change the battery and see if it's fixed.  It can't hurt.

Won't the battery maintain the display whether the furnace is on or
off? 

Not if the battery is dead.

I've installed two of these things. They're not hard, and I see no reason to
pull a fuse. Goto Lowes, buy one you like. Unbolt the old one from the wall.
Unscrew the screws holding the wires in place. Put the wires on the
appropriate leads of the new 'stat and tighten. Attach to wall, and put a
battery in.

But first try putting a new battery in the old 'stat.

re: 10 - For safety's sake, it's a good idea to pull the fuse, or use whatever
other method you have for turning off the power to your heating and cooling
system. HGTV agrees:
http://www.hgtv.com/hgtv/rm_electric_a_conditioning/article/0,1797,HGTV_369
2_1
385914,00.html

This response has been erased.

try a new battery first - the stat furnctions and the display functoins
are not interconnected. and i disagree that a flakey stat (all by itself)
could cause teh house to burn down.
  
it would be interesting to all of us if you were to determine (as suggested
above somewhere) if the brite/dim relationship coincides with the
furnace blowoing hot air or in idle. 
  
as long as the house temp is staying where you feel it is correct, there
is no great hurry to do much at all - except maybe experimentiwththe
batteries if you are interested.

My parents once had a thermostat fail 'on' while they were gone for the
day.  The house was about 95 degrees inside when they came back, but it
didn't burn down.

This response has been erased.

Shouldn't a short just look to the furnace like the thermostat is turning the
heat on?  I wouldn't expect the current running through a themostat to be more
than needed to sense whether the thermostat is connecting the circuit.

Yeah, the thermostat turns on the furnace by shorting the wires, and doesn't
cause any fires in the process.  The current's limited by the design of the
circuit.  There's no current limiting inside the thermostat -- the
non-electronic ones are just mechanical switches.

If you're worried about 24VAC shorts, you might want to worry more about
your doorbell. ;)  For that matter, there's a lot of 120VAC wires in your
wall that are probably a lot more of a threat.

Another useful safety feature of many furnaces - the gas heater will
cycle off if the blower motor fails and the heat exchanger overheats.

I find it depressing that some people here are suggesting the
replacement of a piece of hardware without doing ANY diagnostic
tests to see what the problem might be.  If the problem is the
battery, replacing the unit is a waste of time, money and effort.

Re #16:  It's enough to open the gas valve, certainly.  However,
that doesn't take much; on furnaces with pilot flames, the pilot's
gas valve is held open by the current from a single thermocouple.

You mean, because inspection is not complete, people should not jump
to speculative conclusions about whether the thermostat is a potential
weapon of mass destruction?

I'd be surprised if it were the current from the thermostat wire that was
opening the gas valve.  That seems likely to have been the case on old gravity
furnaces (where the gas valve would have been the only moving part), but
modern furnaces have fans or water pumps, ignition systems, and all sorts of
other stuff that has to run.  I don't know much about how furnaces work, but
I would guess at this point that closing the thermostat circuit probably
signals a controller of some sort that does the rest.

The dimming of the LCD display is independent of whether or not the
furnace is on.  Sometimes it is dim when the furnace is on; sometimes
it is dim when the furnace is off.  Ditto for being normal.

This thermostat does have a "low battery" display on it in case the
3 AA batteries are low.  This label (actually not a separate display)
has *not* come on.  Ergo, it seems to me that the batteries are fine.

Russ, what does this do to your theory?

I agree with the need for a diagnosis before anything is replaced.

What is the display displaying, other than temperature?  Do you need it to
adjust the thermostat with?  If not, you could just stick a thermometer in
the room to read the temperature from.  If yes, just adjust it during the
times when it is working properly.

It seems like the cheapest test would be new batteries, or ata least measuring
the battery voltage as the batteries sit in the holder.

Re #21:  Then be surprised.  In a standard pilot-ignited furnace,
the thermostat controls the gas valve directly.  The blower is
controlled by a thermal switch attached to the heat exchanger.
(In a furnace with a powered burner, there is a control unit.)

Re #22:  It strongly suggests that the problem is not the batteries,
but is some other kind of intermittent connection.  It might be
possible to take the unit apart and clean the connections.

Our original low-efficiency furnace gas valve was originally controlled
directly by a simple mechanical thermostat. I replaced the thermostat
with a programmable one, which also then controlled the gas valve directly.
When the low-efficiency furnace was replaced with a high-efficiency, with
a control board, the programmable thermostat also controlled that. I
suspect that the gas control valve is an "active" valve, that requires
very little electric power, but works mainly on the gas pressure itself.

Probably only older furnaces still have such a control setup.  The risk is
that if the pilot goes out then the house could fill up with gas.  Instead,
furnaces these days have a little extra control from a heat sensor by the
pilot.

Re #27: That's a different issue.  On every pilot-operated gas device I've
seen, the gas valve will not open unless there's voltage from the
thermocouple in the pilot flame.  If the thermocouple cools, then gas to the
pilot and the burner is cut off.  The little pushbutton you're supposed to
hold down for 30 seconds after lighting the pilot overrides the
thermocouple.

Yes - in addition to what I said in #27, there is a thermocouple that
cuts off the gas if the pilot light is not working. Those go way back.
I'm not sure what is the equivalent on the high-efficiency furnace, which
uses an electric igniter, not a pilot flame. 

Good question.  I would assume if the burner doesn't light in a certain
amount of time, it gives up, but I don't know for sure.

While on the subject: our high efficiency furnace has developed the
peculiarity of repeatedly going through the startup sequence short of
actually starting. That is, the combustion blower starts, runs for a
while, and shuts down. This is repeated up to a dozen or more times,
when finally the flame comes on and the furnace then operates normally.
This has been going on for some years now. I had an annual checkup
a while back and asked about this, but the serviceman had no idea what
it means. 

(That was the symptom that led me to my frozen intake pipe. :)

Re #29:  In the powered-burner furnaces I've seen, the gas valve
will not open unless there is pressure from air moving out the
exhaust.  This air motion is sensed with a vane and a switch.

Some may also have a sensor to shut off the gas flow if the
temperature fails to come up, but I haven't been into the details.

Re #31:  Check your burner airflow sensor if the spark does not
occur, or the gas valve if it sparks but there is no flame.
Cycling the airflow may eventually get a balky vane to move.

You should at least look at the batteries and not assume that the low 
battery indicator is working properly.  Your batteries could also be 
corroding, which would cause a problem.  Try fresh batteries and see 
what that does.

Re #32: the cycling of my system occurs winter or summer. However I have
suspected it may be related to the pressure sensing function. However
since it does come on, I haven't felt very pressured to solve the peculiarity.

Re #35:  Your *burner* air supply will not be on in the summer,
just the interior air blower.

Sometimes, although not often, we have the heat turned on sometime in
the period 21 June to 21 September. 

batteries are cheap, quick adn easy.  after that, other analysis can take
place.

Re #33: My high efficiency furnace use to do something similar.  It had a
vacuum switch on the inlet side of the induction blower.  A .2" dia plastic
tube connected the switch to the blower intake manifold.  Over time,
condensation would fill a dip in the tube with water and the switch would no
longer sense vacuum, causing the main blower to not come on.  I shortened the
tube and made sure there was a downhill run to the manifold.

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss