Do two wire thermostats short the thermostat wire when they turn on?

In_New_England · November 30

For reasons best not disclosed to save you all from being summoned as accomplices at the inevitable trial, I need to know if a two wire thermostat simply shorts the two wires together with a relay when there is a call for heat.

i.e. I have a battery operated thermostat that connects to our oil boiler. The boiler's relay control board supplies 29V AC across the thermostat wires (I know the specs say 24V, but I measured what I measured). When there is a call for heat, I hear a click from the thermostat and I'm guessing that's a relay that simply shorts the two wires together, is that correct?

I suppose I could experiment with the thermostat to figure that out, but I thought I'd ask you lot too.

Thanks and, for those of you in the North East, hope you can stay warm this week!

109A_5 · November 30

Hello In_New_England,

Technically the word 'Short' or 'Shorts' implies an unwanted (or never wanted) connection. The thermostat during a call for heat provides a connection between the R terminal and the W terminal. The connection could be made by the contacts of a relay, a bi-metal switch, a solid state switch, etc. depending on the thermostat's technology.

This switch closure (switch turning on) activates the control circuit of the heating equipment.

In court I would not use the word 'Short' or 'Shorts' where a switch, relay or other switching device makes a connection that is part of the normal operation.

In_New_England · November 30

@109A_5 Noted carefully, thank you. You are on my shortlist of expert witnesses to call.

EdTheHeaterMan · November 30

NO it is not a short Circuit… It is a completed circuit on a call for heat and an open circuit when the thermostat is satisfied.

If you turn on a light switch and the kitchen light illuminates, would you call that a Short Circuit?. I don't believe any electrician would say that the switch closing the circuit is a Short Circuit. I would think of short circuit as placing a paper clip in a wall outlet and having some sparks then the circuit breaker would trip or fuse would blow. That is a Short Circuit

However some may call the unintentional connection of a thermostat wire that completes a circuit a Short Circuit. So by those terms you could think of the thermostat completing the circuit is just like a short. You would be incorrect but if it makes sense to you, in your minds eye, then by all means use that idea to understand how the thermostat works. The thermostat connects the Red wire and the White wire so the heater will turn on.

I believe @109A_5 has it almost correct, but in your case he has it right.

And I would be Shocked, @In_New_England, it you didn't believe my answer is Grounded in fact

HVACNUT · November 30

The relay in the thermostat closes a set of Normally Open contacts.

A Short, as mentioned, is just that. A Short cut off of the designated path. Like a track runner who cuts across the field.

Steamhead · November 30

@EdTheHeaterMan said:

"And I would be Shocked, @In_New_England, it you didn't believe my answer is Grounded in fact"

OK (groan) we've had our daily Dad-joke…………..

PC7060 · November 30

https://forum.heatinghelp.com/discussion/comment/1823334#Comment_1823334

agree with @109A_5 “Open” or “Closed” is the common term with relays. Note the term “Normally” applied to either state denotes the power off condition of the relay.

As in the thermostat closes the relay applying input transformer voltage to the boiler control.

EBEBRATT-Ed · December 1

Thermostat does not have to have a relay could be a set of contacts or a mercury switch in the old days.

In_New_England · December 3

@EdTheHeaterMan I consider myself empowered by this information.

For my next crime: I need an off the shelf module that can detect the presence/absence of the 29V AC signal on the thermostat wires. Basically: I want to splice the thermostat wire pair and detect the presence of 29V when the circuit is open, and 0V when it is closed by the thermostat.

I thought this would be easy to find, since obviously every boiler/furnace in the America has this as part of the control circuit board, but alas I can't find something suitable.

I could build an op-amp circuit myself with opto-isolators, but I'm too lazy!

Jamie Hall · December 4

Eh? Why 29 volts AC? Why not 24, which is the nominal loaded voltage of most heating control equipment?

And which of the two control wires do you want to measure voltage on, and in reference to what ground? Or do you want to measure the voltage between the two control wires?

I presume you mean to splice the sensor in question into one of the two control wires — not splice the control wires together? How would you decide which control wire to splice into?

And is this for long term — that is, effectively permanent — installation, or occasional measurements? If this is short term… any decent modern multimeter will do. Almost all of them have input impedance on the voltage scales of 1 megohm or better.

Can you describe a bit better just what it is you are trying to do?

EdTheHeaterMan · December 4

If the ohms on this bulb are high enough so not to produce a holding circuit, this bulb is rated for 24 volts. In theory the bulb would illuminate when the thermostat is not calling for heat and the bulb would go out when the thermostat is calling for heat.

The true test would be to spend about $6.00 and test the theory. If there is enough resistance in the filament so the relay drops out, then you have your indicator lamp.

https://www.bulbamerica.com/collections/24v/products/osram-sylvania-7511-21w-24v-bayonet-base-s8-automotive-light-bulb

EdTheHeaterMan · December 4

By the way. we refer to the control voltage as 24 VAC. Not 29 volts.

The reason you have 29 volts on you meter is because there may be no load on the transformer. Once something is actually using the 24 VAC the 29 will drop somewhat.

The other thing is that the primary voltage to the transformer may be as high as 130 VAC and as low as 110 VAC. So your house current must be at the high end of the allowable input voltage. Higher input = higher output

109A_5 · December 4

You just want an indicator ? Two LEDs (or one LED and a diode) and a current limiting resistor. When the thermostat is off (open) the LEDs will be on. Probably won't draw enough current to activate the heating equipment. More than just an indicator needed, use a SSR (Solid State Relay) AC to AC or whatever you want there are many types.

Another way a bit different, LEDs are on when the heating equipment is on.

ethicalpaul · December 4

There are tons of 24v LEDs out there you can use as indicators

Here is one

109A_5 · December 4

Yeah there is, but which spec do you believe 12, 24, 20mA or 0.06W ??? Maybe it is the Power Factor correction messing with the numbers.

" Features:Rated Voltage:AC/DC 12-24V,Rated Current:20mA, Rated Power Consumption: 0.06W;Light Color:Red/Orange/Blue/Green/White. "

Jamie Hall · December 4

Have we figured out what the OP wants to do?

In_New_England · December 4

Thanks all for the info! Yes! I shoulda thought of loading. Ah the good old days when things were loads and your transformer terminal voltage dropped when things happened.

Here's my expert drawing of my proposed project

[Thermostat] ———+———————————————————————————————[ BOILER ]
[          ] ———|————+——————————————————————————[        ]
                |    |
                |    |
[Box X     ]————+    |
[          ]————|————+ 
[          ]    |    |
[Sensor Y  ]————+    |
[          ]—————————+

The thermostat to Boiler connection is the usual one

Box X is my magic box that sends a heat call when the outdoor temperature falls below a certain level, say 0C / 32F. (Why? see here please. In short: run the baseboard periodically to avoid pipes freezing.)

It's attached in parallel to the existing thermostat as I felt it would be the most fail-safe. I had a scheme where it would be a pass through for the existing thermostat and be part of a more complicated home made integrated controls that switched between baseboard and mini-split, but I'm going to develop that system as a series of separate layers, with this one being a basic fail-safe that always runs.

Sensor Y is an optional feature I am imagining here.

Box X as it stands now is open loop - it sends periodic calls to the oil boiler but doesn't know if the thermostat has already recently called for heat. Sensor Y senses the voltage across the thermostat wire and tells Box X what the most recent call for heat was (regardless of who did it) allowing Box X to skip a run if the thermostat is working.

PS. In reality, we probably don't need this, since I'm following the boss' advice (All the married men know who the boss is) and running the oil heat at sub zero rather than the mini-splits. That was the consensus solution on this forum too. But I have all the parts and the code, so I'm going to do it anyway.

ethicalpaul · December 4

Your drawing helps me to understand what your saying I think, but did you tell us in words what it is that you want Sensor Y to do?

It seems like you just want to have several thermostats wired in parallel, so that if any of them gets too cold, the heat will come on. That's what your drawing looks like anyway.

If that's what you want, then yes, that will work.

Edit: I read your description again, and I admit I remain confused. Why does Box X have to care about who asked for heat?

Jamie Hall · December 4

Interesting. Clearly there will need to be some electronics related to your sensor Y to keep track of time, if nothing else, and produce a suitable output signal to Box X, perhaps to inhibit calls from that box for a certain amount of time.

The key consideration for the sensor portion of sensor Y is that whatever it is must — repeat, must — be at least moderate impedance. I'd want to see at least 1K ohm if possible. Why? To avoid any possibility of interference with either the thermostat itself or with the actual boiler control. I'm not sure what device — or devices — I would use for that. I suspect that your initial thought of a triac optoisolator would probably suit, although a reed relay might also work perfectly well. You might be able to get higher impedance, if needed (I'd experiment with your specific boiler control) by specifying a lower trigger/coil voltage and a series resistor. I'll let you worry about that.

Remember that your output from the device will be a logical low — conducting — when the thermostat is off! You'll need to invert that if the rest of your circuitry requires a logical low when the thermostat is on. I'll let you worry about that, too!

Digikey has a nice selection of both devices…

In_New_England · December 4

@ethicalpaul Sensor Y is a micro-optimization.

If the mini-splits are doing all the heating all the time, there is no need for sensor Y at all: only Box X will be periodically activating the baseboard heat.

If the baseboard heat is being used, the pipes will be warm enough and Box X's call for circulation is redundant. It's not a big deal - the brief burst of circulation will not alter the indoor temperature by much.

Therefore Sensor Y is a nice-to-have option, where Box X keeps track of external calls for heat by monitoring the thermostat wire. If it detects, via sensor Y, that the thermostat is regularly calling for heat, it will stop its own operation.

@Jamie Hall - yes, absolutely - I'm looking for a high impedance solution that will draw very little current. I saw a chip (HCPL3700) that might work: it rectifies and optoisolates and gives a binary output, which is very convenient to interfacing with Box X which is a digital micro-controller.

ethicalpaul · December 4

Thanks for that. I am not understanding why not just use the temperature of the room to determine when to turn on a system. One thermostat could be set lower than the other so it doesn't come on unless things get chilly. But I have no doubt I'm missing some aspect of what you're looking to do, so I will just offer that observation without expectation of response! Thanks again!

109A_5 · December 5

Hello In_New_England,

I would do it something like this.

Say you want the boiler the run for 10 minutes every 8 hours when the temperature outdoors is less than 32 degrees Fahrenheit. Mostly independent of anything else. (the times are just an example)

If the boiler's normal thermostat has activated then the microcontroller box delays its call by 8 hours (or what ever you want).

The microcontroller would ignore a call from itself (relay closure).

So above 32 degrees Fahrenheit or if the boiler's normal thermostat calls are more frequent than 8 hours the microcontroller will never generate a call for heat.

Fairly simple program for an Arduino or the like.

What is the maximum AC input voltage of the HCPL3700, I could not find it in the data sheet.

The temperature input can be anything you want a temperature switch, Analog or Digital. If you use a sensor that reports the actual temperature to the microcontroller (analog LM34, digital DS18B20, etc.) the frequency of the boiler running can increase as it gets colder outside.

I would probably buffer the relay coil with a transistor and don't forget the diode across the relay coil.

109A_5 · December 5

HCPL3700

Going by the symbols in the HCPL3700 circuit, D1 and D2 appear to be Schottky Diodes (low forward voltage drop). D3 and D4 seem to be Zener Diodes (Limits the input Voltage). You may not be able to apply 24 VAC directly to pins 1 and 4, you may need (at least) a current limiting resistor. If you use this device I would use the DIP package and install a socket.

Edit: Also look at - Application Note 1004

https://docs.broadcom.com/doc/AV02-3699EN

You will need to filter it too unless your microcontroller's code can deal with the microcontroller's input pin switching on and off at a 120 Hz rate during boiler inactivity.

In_New_England · December 5

@109A_5 that’s a wonderful diagram. Yes that’s the design I’m building, except that I have a DS18B20 and the code compares the value against a setting ( the switch is a software construct).

Thank you for the notes on the optoisolator. While glancing through some circuits I thought I found one with capacitor smoothing on the AC side. Perhaps I imagined that.

For this project I think I’ll skip the thermostat state sensor: one more minute of baseboard heat never hurt nobody.

I’m using a simple reciprocal formula to increase the frequency of the circulation as the OD temp gets lower.

You all have been fabulous. I’ll post some photos when I’m done.

In_New_England · December 6

https://forum.heatinghelp.com/discussion/comment/1824599#Comment_1824599

You are absolutely correct - the baseboard hydronic heat thermostat is what drives the regular calls for heat from the baseboard.

The oddity here is that I'm worried about the case where we do not use the hydronic heat, relying on the mini-splits down through freezing weather. I worry that the unused hydronic piping will freeze.

To guard against that, I'm developing this control circuit that will run the hydronic system briefly (say 1 min/hr) to just keep the pipes from freezing, gated off the external temperature which is a proxy for the water temp in a pipe at risk of freezing.

109A_5 · December 6

Yes you can just add a capacitor at pins 2 & 3, see page 6 and 7 of the App note.

109A_5 · December 6

Ok , I read your other thread. Since you only intend to have the microcontroller box run the system for a minute, I would use a normally closed contact of the microcontroller box relay to disable the boiler from firing. I see no point in firing a boiler that supplies DHW for only a minute multiple times a day, why add the wear and tear to the boiler equipment ?

ethicalpaul · December 6

https://forum.heatinghelp.com/discussion/comment/1825047#Comment_1825047

ok I see. So why not a second thermostat in parallel, controlling the hydronic system, located on the pipes themselves set for 50 degrees, say? Or whatever value you like

But I dare say that if it’s cold enough out to freeze your pipes, you will be running the hydronic system anyway.

In_New_England · December 6

https://forum.heatinghelp.com/discussion/comment/1825058#Comment_1825058

The design I described is indeed a thermostat like system that is connected in parallel
I agree that pipe temperature is what we are after, but I believe gating it off the outdoor temperature will be similarly effective: the pipes, where they are accessible, might be warmer than the coldest parts. The coldest parts will not be colder than the outdoors.
You are correct (and I did mention that earlier in the thread) that practically speaking, this winter, we are running the oil heat when it is below freezing. This is mostly me tinkering and getting a simple insurance policy out of it :)

In_New_England · December 6

https://forum.heatinghelp.com/discussion/comment/1825055#Comment_1825055

I think the additional use of the boiler will be minimal.

I don't know the control circuitry of the boiler, but from observing it work, I think it's quite simple: The boiler has an internal thermostat (set at 175 I think). Whenever the temperature drops below this, the boiler fires to bring the temperature back up.

When DHW is used, when the baseboard is run, or simply over time, the boiler loses heat and at some point the internal thermostat drops below set point and fires the oil to bring the temperature back up.

I suspect that the short bursts of circulation I will call for will only slightly increase the boilers normal background activity, probably by making it fire for a little longer than usual.

109A_5 · December 7

A dedicated call 1 minute once a day is minimal to me, but still too much, 1 minute every hour is a lot. To me oil heat is a high maintenance system, I can't see adding a lot of extra cycles per day as being good.

I think 175 is the high limit.

Due to the added extra heat load an extended DHW call to me is much better. Not a dedicated call a minute long just to keep the pipes warm.

" I suspect that the short bursts of circulation " to me that is different than a call generated by a contact closure of the thermostat circuit. If 1 minute so circulation is the goal I would just control the circulator(s) (assuming no zone valves). If you have zone valves I do a boiler call via the thermostat wires and disable the burner system for that minute.

If you can wire up a ESP32 and do the code I'm sure you could figure out a boiler control if you wanted to.

In_New_England · December 9

Running the zone will lead to the boiler losing heat. If the temperature drops below the low limit the burner will fire.

If I suppress the burner during my run it’ll just fire after I’m done. It’ll have to fire at some point to put the energy back in that was taken out.

Jamie Hall · December 9

https://forum.heatinghelp.com/discussion/comment/1825738#Comment_1825738

Well, true. The boiler and the water in it will lose heat. However if the boiler is left controlled by an aquastat with a normal high setting and a nice differential (this is not the same as the high limit safety shutoff), so what? You'll get your nice short burst of circulation from time to time — which won't lose that much heat, but some — and at some point the boiler will hit the low limit and fire for a reasonable cycle until it hits the high setting. If you want to make that firing cycle longer, may I suggest a buffer tank to increase the mass of water?

I have to admit that I have a sneaking suspicion here that we are looking at a solution in search of a problem.

Perhaps we need to step back out of the weeds long enough to define exactly what the problem is we are trying to solve here?

In_New_England · December 9

@Jamie Hall

Perhaps we need to step back out of the weeds long enough to define exactly what the problem is we are trying to solve here?

Quite!

From my perspective:

Initial (and over all problem) that I'm trying to solve: If I stop using the base board heat all winter I may run the risk of a baseboard pipe freezing. Solution: A device that periodically runs the baseboard whenever outside temperature is below freezing.

Note: This winter we've been using the baseboard heat instead of the mini-splits at freezing temperatures, so the issue is moot. This may change in future winters.

Concern raised by @109A_5 : Short calls for heat like this will increase wear and tear on boiler. I'm not convinced this is perceptible and I don't think there is anything that can be done about this given the design of the setup I have.

Jamie Hall · December 9

Thank you! And you do indeed have a completely valid concern — perhaps even more valid than you might realise: if you have much heating plumbing wandering around inside outside walls, the risk of having it freeze if your only heat in the place is the mini-splits (or, I might add, any forced air system) is insanely high. Why? Because, for better or worse, they heat the air, and only that heats the structure.

So… we need to do something about that. Quite right.

And I will make on more assumption: you wish to minimize the use of the baseboard heat in favour of the heat pumps. Not sure about that one myself— but there are valid reasons why one might want to do that. Just that I'm not convinced, having a very good mini-split system myself (with a superb steam system to do the heavy lifting). But that's not part of this discussion.

Since you seem to be handy with electronics, I think what I might recommend is to locate somewhere in your heating plumbing which you think is the most vulnerable location with regard to freezing (needs to be on a return line), and strap a thermistor with a suitable range on there (or it could be a thermocouple — depends. Do you want to play with a voltage (thermocouple) or a resistance (thermistor). Then hook up a simple circuit — I doubt it needs to be fancy — to call for heat in the pipes whenever they dropped below … oh, let's say 40 to give a bit of margin — and turns off the call when they rise to some convenient value — perhaps 50 or so. And treat the output of that circuitry as a normal call for heat to the boiler and circulators(s) (you could even have one for each zone, if you wanted to get fancy).

The potential problem here which I think may be what you were trying to get around is the possibility of having a conventional heat call from the space thermostat coming close after — or ending shortly before — a call from the freeze protection circuitry, but I honestly don't think that that will be a problem as, if the system is being used for space heating it is very unlikely that the pipe temperature will be low enough to trigger the freeze protection, at least if the space heating is on any kind of reasonable cycle timing.

With the electronic circuitry in there — and your ability to fiddle it — it would not be at all difficult to adjust the cutout temperature for the freeze protection to give a reasonable cycle time for the system — say five minutes or so — if it turned out that the system heated up too fast.

But all this is my way of approaching it… and every engineer approaches a problem differently!

In_New_England · December 10

@Jamie Hall

I think what I might recommend is to locate somewhere in your heating plumbing which you think is the most vulnerable location with regard to freezing (needs to be on a return line), and strap a thermistor with a suitable range on there (or it could be a thermocouple — depends. Do you want to play with a voltage (thermocouple) or a resistance (thermistor). Then hook up a simple circuit — I doubt it needs to be fancy — to call for heat in the pipes whenever they dropped below … oh, let's say 40 to give a bit of margin — and turns off the call when they rise to some convenient value — perhaps 50 or so. And treat the output of that circuitry as a normal call for heat to the boiler and circulators(s) (you could even have one for each zone, if you wanted to get fancy).

I love your idea of attaching the temperature sensor to a baseboard heat pipe! It has the genius of simplicity and directly addresses the issue we are trying to fix. It therefore avoids the issue of running the boiler when it is already in use.

My concern would be to find a suitable place to put the temperature sensor. The basement is unfinished and I can trace a pipe on the return line to where it goes up and perhaps that would be a decent place to put the sensor. I could watch it on cold nights with the boiler off and see how cold it gets relative to the outside temperature.

Since everyone here has been so kind and interested, here's an update on the hardware side:

My original plan was to place the temperature sensor on the building exterior and activate the system when the exterior temperature dropped below freezing.

@Jamie Hall the box, in its soul, is what we spoke about: It senses the temperature from the probe (silver cylinder + wire, coiled up on the right) and activates the relay module (blue component) which will be connected to the boiler (empty socket on top right)

In keeping with these times it has some more features: The micro-controller (chip on left) connects to the home router and offers a simple webpage which tells us the temperature and the last time the relay ran.

I have some long term plans of using the micro-controller as the hub of an integrated temperature control: I'll make a bunch of controllers that my family will use to set the temperature in the various rooms and the controllers will talk to this one which, depending on the outdoor temperature, will decide whether to run the baseboard or run the mini-splits. The mini-split controller is an untouched project as of yet.

Anyhow! After the most recent discussion, I'm off to see if I can find a suitable heating pipe to attach the the temperature sensor to.

109A_5 · December 10

My experience with frozen pipes is they don't always freeze where you think they will. High winds will find a path to your pipes that you never knew existed and in the oddest places. So if you put 1 sensor or 10 in your favorite places (easy access) the pipes will freeze elsewhere. I like your strategy about taking action around 32 degrees outdoor temp. Since your boiler is also used for the DHW I would not activate the boiler's burner for the sole purpose of freeze prevention, it will do that on its own if needed.

If your intent is to run the freeze protection for one minute and increase the frequency of the application interval the colder it gets outside the usefulness of monitoring the zone's thermostat or the boiler's T-T may be minimal. You may want redundant DS18B20 sensors and/or a software means to determine the data is plausible.

yellowdog · December 10

Why not just change your circ relay to a taco EXP with the pump exerciser card? Put it on the freeze protection setting and you're done.

EdTheHeaterMan · December 10

https://forum.heatinghelp.com/discussion/comment/1825990#Comment_1825990

That would not be as much fun!

Do two wire thermostats short the thermostat wire when they turn on?

Comments

Categories