Honeywell T87 mercury thermostat

dcaheat

A quick question about the Honeywell T87 mercury thermostat: Can the thermostat be damaged if the heating system (gas valve) recommends an anticipator setting of 0.8 and the anticipator is set at 0.3?

Jamie Hall

No, Set the anticipator to what works best in your situation. The recommendation is only a starting point. But... remember to keep the thermostat level.

109A_5

Hello @dcaheat,
I suppose it is possible, but I would expect the anticipator heat would just be over active and make the run time shorter. Is the Gas Valve current actually going through the thermostat or is there other control equipment in-between the thermostat and the Gas Valve ?

Mad Dog_2

Doubt it unless it got a line voltage surge.   Mad Dog 🐕 

mattmia2

Maybe if you go to the extremes you could heat the resistor enough to damage it or burn it out, if the control draws say 1.5 a and you set the anticipator to .2a. It would also shut off almost immediately on a call for heat.

HomerJSmith

The thermostat should be measured against a separate accurate thermometer and adjusted to read the room temp accurately. There is a thermostat wrench that makes the adjustment on the thermostat dial. Old bi-metal mercury thermostats will get out of adjustment. Just an added note.

WMno57

I purposely mounted a Honeywell mercury (not T87) off level because I wanted a low temperature thermostat. Kept the house at 58 on it's lowest setting when it was level. Now it the maintains the vacation home at 48 when set to it's lowest setting. No batteries or electronics to worry about either.

Jamie Hall

WMno57 said:

I purposely mounted a Honeywell mercury (not T87) off level because I wanted a low temperature thermostat. Kept the house at 58 on it's lowest setting when it was level. Now it the maintains the vacation home at 48 when set to it's lowest setting. No batteries or electronics to worry about either.

That works. I've done that on occasion. It won't work, though, if some of the occupants depend on the numbers on the dial to tell them whether they are comfortable or not...

dcaheat

I would like to measure the current draw of the primary control on my boiler so as to be able to set the T87 anticipator correctly. The T87 instructions say to connect an AC ammeter between the R and W terminals on the wall plate or subbase. So, this is what I think I need to do:
1. Turn off main power to boiler.
2. Remove thermostat from the subbase.
3. Turn on main power to the boiler.
4. Set multimeter to measure AC amps.
5. Touch multimeter Red probe to R subbase terminal and Black probe to W terminal. The boiler should turn on.
6. Let the boiler run for a few minutes and then read the amps on the multimeter.
Is this procedure correct?
Thanks for your help!

109A_5

Hello @dcaheat,
That should work. Make sure you use the correct amperage range. With some meters you may have to use the 10 to 20 Amp range.

Big Ed_4

T87 setting , First measure the mili ohm of the thermostat circuit . Setting the anticipator to that reading will give a ballpark of 4-6 cycles an hour . Low mass radiation and steam you what to set it longer ....

A short will fry them ..

dcaheat

Thanks for the info.
I am trying to determine why the T87 anticipator is not working. Here is some background.
My heating system is gas-fired cast iron hot water. I replaced the 40 year old HydroTherm boiler with a new WM CGA3. The old boiler worked perfectly with the T87 for the 36 years I have been in the house. (Although I did not know it at the time the new boiler was installed, the anticipator was set to 0.3 even though HydroTherm recommended 0.6) When the new boiler was installed, nothing was done to the T87. Now, the boiler runs for about an hour and the thermostat is not satisfied until the room temperature is about 2 degrees above the set temperature. WM recommends the anticipator be set at 0.8. Changing the anticipator does nothing.
I was able to obtain a NOS T87 (very pretty!). I set the anticipator to 0.8 and the heating tech mounted the stat on the wall. Same result! I opened the stat and discovered the tech had moved the anticipator to 0.6. I tried several other anticipator settings from 0.6 to 0.2, but the results are always the same.
I am wondering if I have burned out anticipators.
Any ideas as to what is happening will be appreciated.

Jamie Hall

It's rare for them to burn out, but it has happened. If you have a multimeter, set it to ohms and read the resistance across the anticipator with the thermostat set to off. If it's open circuit, it burned out. If you have two which are burned out, find out why...

Big Ed_4

T87 is so old school , no need to use them anymore . We can program in the cycle in now . I like the T4

109A_5

Hello @dcaheat,
If the anticipator resistor is burned out you won't get any heat unless it is set to 1.0 which effectively eliminates the resistor from the series circuit. I suspect the system current may be so low that the anticipator resistor simply does not produce enough heat to be effective. You need to measure the current to see if it is in the normal range of the anticipator settings, 1.0 Amp down to 0.12 Amps. If below 0.12 Amps the anticipator resistor simply does not produce enough heat to be effective.

yellowdog

Is the inside of the thermostat clean of all dust and cobwebs so it does not interfere with or affect the movement of the mercury bulb.

dcaheat

Thanks for the suggestions.
I measured the resistance across the anticipator. It measured 1 - 5 ohms, depending on where the anticipator was set. So, the anticipator is not burned out.
Tomorrow I will measure the current.

jesmed1

You should also be able to derive the current by measuring the voltage across the resistor with the thermostat connected as usual and the contacts closed so the boiler is running and current is flowing through the resistor.

Since you know the anticipator resistance (let's say 5 ohms at one end of the scale), by Ohm's Law V=IR, so current I=V/R. So if you measure 5 volts across 5 ohms, that's 1.0 amps flowing through the resistor. If you measure 0.5 volts across 5 ohms, that's 0.1 amps.

Needless to say, you do not want to short across the resistor accidentally while doing this.

dcaheat

I plan to measure the current by touching the probes to the R and W terminals on the base plate, as I described in the procedure above. Question: does it matter which terminal I touch first?
Thanks

109A_5

Hello @dcaheat,

dcaheat said:

Question: does it matter which terminal I touch first?

No, it does not matter.

@jesmed1 as much as your Ohm's law is correct, many multi-meters have poor accuracy at low Ohms, so I think actually measuring the current (if done correctly) is better.

dcaheat

I measured the current. 0.025A, well below the threshold needed to activate the anticipator. Also well below what one would expect given that the gas valve is rated at 0.7A and Weil-McLain recommends a setting of 0.8A.
What would cause the current to be so low? Is there a fix?
Thanks

mattmia2

Is there a solid sate control? If the control is solid state or only uses a very small relay it may draw almost no current. You could put a resistor between w and c to allow additional current to heat the anticipator, otherwise you'd need to go to a microprocessor control that can time the cycle length.

pecmsg

dcaheat said:

I measured the current. 0.025A, well below the threshold needed to activate the anticipator. Also well below what one would expect given that the gas valve is rated at 0.7A and Weil-McLain recommends a setting of 0.8A. What would cause the current to be so low? Is there a fix? Thanks

Bad reading?

Jamie Hall

109A_5 said:

Hello @dcaheat,
I suppose it is possible, but I would expect the anticipator heat would just be over active and make the run time shorter. Is the Gas Valve current actually going through the thermostat or is there other control equipment in-between the thermostat and the Gas Valve ?

mattmia2 said:

Is there a solid sate control? If the control is solid state or only uses a very small relay it may draw almost no current. You could put a resistor between w and c to allow additional current to heat the anticipator, otherwise you'd need to go to a microprocessor control that can time the cycle length.

25 milliamps is very low for almost anything except a solid state control of some kind -- but is actually quite reasonable for solid state. The cure, as @mattmia2 suggests, would be a nice little resistor between W and C. Try around 50 ohms power resistor -- say 25 watts or so -- to start with the anticipator set to 0.5 amps. Keep the resistor away from the thermostat -- like far away! -- it will get hot.

mattmia2

Are you sure you had the meter on AC amps and not DC amps?

It is very possible that the page in the manual with the control module and the table about the gas valves was the result of a technical writer who didn't exactly understand the system merging the instructions for the control module in to the existing instructions. It would be reasonable to ask their technical support about this.

109A_5

Hello @dcaheat,
Well the current kind of matches the symptoms. Vent damper current maybe ? Maybe verify the controller, maybe it is different from the manual below.

Hot water ? Maybe the thermostat is seeing a zone valve controller and not the boilers control module and/or gas valve.

From the CGa Series 3 manual...



https://www.weil-mclain.com/sites/default/files/field-file/CGa Series 3 Boiler Manual 0520.pdf

dcaheat

I'm still working on my low amp problem. Thanks for the suggestions.
Question: Why is the gas valve current draw used to set the anticipator? What's special about the gas valve?
Thanks!

Jamie Hall

dcaheat said:

I'm still working on my low amp problem. Thanks for the suggestions.
Question: Why is the gas valve current draw used to set the anticipator? What's special about the gas valve?
Thanks!

Nothing. But in many older systems, the thermostat was directly connected to the gas valve and nothing else. Really simple -- power supply, thermostat, gas valve, back to power supply. Those were the days...

Nowadays the thermostat, more often than not, doesn't control the gas valve (or ignitor for an oil burner) directly. Instead, there may be a relay. There may be a solid state switch. There may be a zone valve. There may even be a computer.

What the anticipator needs to know is what the current draw of the controlled device is when the thermostat switch is closed (thermostat calling for heat). And this can vary all over the map, from as much as almost an amp for a gas valve or zone valve to as little as 10 milliamps for a solid state switch.

109A_5

Hello @dcaheat,
Not all gas valves draw the same current. Different size valves and designs, etc. Not all heating systems heat the room that has the thermostat in it the same way. The heat anticipator is a variable resistor so you can tweak for these issues. The resistance of the variable resistor was chosen to provide a certain amount of additional heat to gently bias the thermostat to end the call for heat slightly early. Since typically (historically) the thermostat directly controlled the gas valve so the gas valve current is the major factor to calibrate the variable resistor current to. Providing at least a starting point.

Presently the thermostat may not directly (electrically) control the gas valve. Electronic control boards, relays, Zone Valve controllers, etc. may be in the middle. So the thermostat (in come cases) does not have the actual gas valve current through it during a call for heat.