On 4/6/2018 6:10 PM, J.Albert wrote:
What I have:
Burnham gas furnace, hot water, model P205 WI (it's 30 years old but
still in good condition)
There's no stack vent (it had an old HyTemp which I've disconnected
electrically and locked open).
No fans of any kind.
Electric circulating pump.
House is old (but not large) and has big cast-iron radiators.

I was having some problems (furnace wouldn't start correctly), had my
plumber/heating guy come over. Checked the pilot/ignition, cleaned the
sensor.
He installed a new thermostat (old one must have been 40+ years old).
The new thermostat is a Honeywell "round" T87K1007.
There are only two wires from the furnace that connect to it.
No fans, no air conditioning.

The furnace still wasn't right afterwards, but I investigated on my own
and came to the conclusion that the intermittent ignition module (old
Honeywell S86H) had something inside that was failing. I replaced it
with a new Honeywell "universal" module (S8610U3009/U), and that seems
to have solved the problems with the furnace.

But I find that the new thermostat is quirky.
The "room temperature" indication (bottom pointer) seems to have no
relationship to the "desired temp" setting (top pointer that you set by
rotating the dial).

I checked the thermostat's DIP switches (it has these instead of the
"predictor" that the old one has), and found one that wasn't set
according to the manual, and reset it.

But... I've come to the conclusion that the thermostat doesn't do what I
want it to do.
Please keep reading.

I understand that a thermostat is intended to automatically maintain
room temperature within a slight variation from the desired target
setting. Perhaps within 1 degree +/-?

The result is that the furnace cycles on/off too much for what I want.

Emotion aside...
What is it that makes a stable temperature undesirable?
Can you put back the old thermostat?
The thermostat datasheet suggests that the jumpers might be setting
the number of cycles...but that's vague.
You can emulate the predictor by putting a resistor in series with
the thermostat and placing the resistor proximate to the temperature
sensing element. You'd have to do a lot of experimenting to determine
the resistance and proximity.

Try bolting the thermostat to something with large thermal mass
and restricting the coupling between the air and the sensing element.

What I've been doing (it's still cold at night here in southern New
England yet) is setting the thermostat up higher than where I want,
letting the furnace run for an hour, then turning the thermostat down
(and furnace off) until it gets a little cooler than I like.

I can repeat this two or three times a day, and that's all I need.

So....
I know there are "smart thermostats" one can buy, but I still don't know
if they can do what I want. Which is:
- set a high temp limit (furnace shuts off)
- set a low temp limit (furnace turns on)
Say, set my low for 64 degrees and my high for 70 degrees. Or something
like that.

Are there any thermostats out that that work this way?
Or do they ALL "cycle within" a smaller range of temperature change?
They're doing what they were designed to do.

High/Low setpoint thermostats used to be all over ebay for less than $10,
but all the ones I checked were "no longer available".

I'm still struggling with why it's worth a lot of trouble to make
your temperature unstable.

When modifying HVAC, it's important to ask yourself two questions...

1)What might go horribly wrong?
You can easily convince yourself that any catastrophic failure is
highly unlikely. Problem with statistics is that they are useless
in the individual case. A pregnant teenager is not comforted
by arguments that birth control is 99% effective.

2)Will my fire insurance cover my stupidity?