sensing and controlling switch on and off

[olsn500]

Long winded here hope you can help, I'm new to this and will try to not sound like an idiot.

I removed the boiler from my home and needed to install multiple furnaces. The furnaces can NOT start and or run at the same time due to a drop in fuel pressure. Installing a larger diameter fuel line is cost prohibitive.

I want to build a device that will control the simple two wire mercury thermostats, they are basically on/off switches.

The device needs to see that if one thermostat is calling for heat it does not allow the other one to come on. Then once that area of the house has reached temp the thermostat will stop calling for heat. At this time if the temp has dropped at the other location that thermostat will then be able to come on. The first thermostat will then be disabled until the second thermostat has completed its cycle.

I have found several articles for controlling switches but nothing that really talks about sensing other switches and only allowing one or the other to operate.

I have a raspberry pi 3 currently and decent soldering skills/tools. My programming knowledge is zero but I need to do something to get heat in my home for the winter. I'll figure it out just need some direction. Thanks

[pcmanbob]

Hi.
How many furnaces / stats are you talking about ?
Will it be only one stat per a furnace ?
I am asking because I think you could actually do this with just some relays and it would be more reliable and not affected by a sudden loss of power which might upset your pi.

And also what voltage are the stats working on ?

[olsn500]

Two furnaces, one stat per furnace, im waiting on kids to get home so i can have some help measuring voltage and the voltage drop when a furnace kicks on. That said, I cant imagine its any higher than a 5V system. I will check though.

[pcmanbob]

Are the stats feed from the furnace controls ?
Would expect them to be more than 5v.
In the UK they are normally 240 v, but in some countries they are 24v. They may also be AC and not DC.

[olsn500]

https://www.alpinehomeair.com/related/S ... ams%20.pdf

24v lol Still waiting on kids, but have found this link that matches my particular furnace.

The stats are both old fashioned two wire W and R on the picture.

My thought was that all I need to do is interrupt the power to one stat while the other is on, then when it shuts off it allows power to the other stat.

[pcmanbob]

Not on my pc at this time of night. Will have a look tomorrow, check that the circuit I have in mind won't cause the relays to chatter and draw you up a diagram.

[olsn500]

That would be fantastic, thank you.

[davidcoton]

On both thermostats:

• Check that the red wire is 24V, not mains (preferably before letting the kids help). DO NOT CONTINUE if mains! The red wire should (see below*) be the unswitched supply to the thermostat.
• Check whether it is DC or AC.
• Check that the white wire switches from 24V (thermostat demanding) to open (thermostat satisfied). Remember that turning the thermostat down will satisfy it, turning it up will cause a demand.

Get a relay with a 24V coil, AC or DC as required. Contacts need to be able to handle 24V AC or DC.
Connect the relay coil from one white wire to common. (Blue in the diagrams, but is that accessible if you only have two wires? Possibly only at the furnace controller.)
Route the other furnace's white wire through the relay contacts, so that it is open when the relay is on.

The furnace with the relay coil connection will always heat when required, the other will only heat after the first is up to temperature.

Make sure all wiring is safe, fully insulated and tidy, even if only 24V.
If you don't understand this, ask before guessing. Especially you must be sure about checking the voltage safely.

*Be aware that wiring is not always done how you expect, or how it should be. DO NOT CONTINUE if the measurements don't make sense.

I suggest you wait for pcmanbob or others to check this before proceeding! I can make mistakes.
Also you may want to report your own OP and ask the mods to move it to "Off topic" since the current solution is not Pi based.

[pcmanbob]

Hi.

First I am going to repeat some of what has already been said.
Safety first so if you are no confident working on mains powered equipment seek professional assistance, when working on your furnaces even the 24v side switch them off at the main switch, make sure all working is tidy and relays are installed in suitable insulated enclosures and above all make sure all earthing is connected and by that I mean protective mains earth.

From the provided diagram I am assuming that the control circuit works on 24V DC but please check as this will affect the relay requirements.
you will probably have to do this at the furnace its self as you required the common terminal to test to be sure to check both furnaces unless you installed them from new and know they are both exactly the same.

The relays you require will need a 24v dc coil ( see statement above about testing) and one or more change over contact.
My recommendation would be to look for octal relay they are plug in and you can get a base which has screw terminals that will accept mains type AC wiring, which is what you will be using to connect between your thermostats and the furnace. These tend to be industry rated so very reliable but if they should fail you can just un-plug and replace without having to do any re-wiring.




Now for the circuit



So you can see that when we turn on a furnace using its stat we also turn on the relay coil that controls the output from the other stat so preventing it from commanding its furnace to run.
You do need to be careful when wiring this as the relay coils will be powered by the opposite furnace to which they are controlling, so you must make sure you use the correct furnace common for each relay coil.
I have simulated this on a bread board so I know it works and does not cause any relay chatter.

Hope this helps and if you have any questions please ask.

[olsn500]

Thank you for your time here, you make it this way and I'll buy you all the beer you can drink.

[davidcoton]

Just a comment about the choice of one relay or two.
With one relay, the thermostat driving the relay coil will always get priority. While it is heating, the other cannot. If the other was heating when the one thermostat switches, it will stop.
Using two relays will allow whichever system demands first to come right up the temperature before the other get a turn.
Your choice of logic, there is no technical advantage either way (unless your furnaces are best with an uninterrupted run once they start). Of course one relay is less expensive than two!

[boyoh]

Hi.

First I am going to repeat some of what has already been said.
Safety first so if you are no confident working on mains powered equipment seek professional assistance, when working on your furnaces even the 24v side switch them off at the main switch, make sure all working is tidy and relays are installed in suitable insulated enclosures and above all make sure all earthing is connected and by that I mean protective mains earth.

From the provided diagram I am assuming that the control circuit works on 24V DC but please check as this will affect the relay requirements.
you will probably have to do this at the furnace its self as you required the common terminal to test to be sure to check both furnaces unless you installed them from new and know they are both exactly the same.

The relays you require will need a 24v dc coil ( see statement above about testing) and one or more change over contact.
My recommendation would be to look for octal relay they are plug in and you can get a base which has screw terminals that will accept mains type AC wiring, which is what you will be using to connect between your thermostats and the furnace. These tend to be industry rated so very reliable but if they should fail you can just un-plug and replace without having to do any re-wiring.




Now for the circuit



So you can see that when we turn on a furnace using its stat we also turn on the relay coil that controls the output from the other stat so preventing it from commanding its furnace to run.
You do need to be careful when wiring this as the relay coils will be powered by the opposite furnace to which they are controlling, so you must make sure you use the correct furnace common for each relay coil.
I have simulated this on a bread board so I know it works and does not cause any relay chatter.

Hope this helps and if you have any questions please ask.

I might be wrong , you are showing both thermoses in the open position (Temp Satisfied)

Starting from cold, Both stats would be made Closed ( Temp Down ) This might cause relay.s
To oppose each other on start up, Please correct me if I am wrong,

[davidcoton]

There is a theoretical possibility that, if both thermostats demanded at exactly the same time, they could "chatter" (actually they would buzz -- many mechanical buzzers were effectively relays that would turn themselves off when energised). In practice, one channel will always respond faster, and will then inhibit the other, so it can't inhibit the faster one. There is an element of (practical) uncertainty, which is why I originally suggested a single relay. However, as outlined above, the single relay design has other potential problems.

[pcmanbob]

So I built my simulation again, this time replacing the stats with direct wire connections to simulate both stats calling for heat.

when I switch on the power to the circuit both circuit energise but in my example relay 1 is powered a fraction before relay 2 so the LED representing furnace 1 comes on for a fraction of a second but furnace 2 comes on and stays on. swapping the relays over caused the circuit to operate the other way round with furnace 1 staying on.

now I doubt that in the actual working system furnace 1 would have the time to do anything, its only that I have LED's representing the furnaces that allows me to see the momentary operation, the LED does not even reach full brightness.

In reality no 2 circuit or relays are exactly the same in resistance and capacitance which is what prevents the relays chattering.

[davidcoton]

Couple more thoughts:

If the system is DC, don't forget to put inverse parallel diodes across the relay coils to prevent damage from back EMF.

If chattering is a problem, and the system is DC, a capacitor of about 1μF and suitable voltage across one relay coil should fix it (giving priority to the other thermostat).

[boyoh]

So I built my simulation again, this time replacing the stats with direct wire connections to simulate both stats calling for heat.

when I switch on the power to the circuit both circuit energise but in my example relay 1 is powered a fraction before relay 2 so the LED representing furnace 1 comes on for a fraction of a second but furnace 2 comes on and stays on. swapping the relays over caused the circuit to operate the other way round with furnace 1 staying on.

now I doubt that in the actual working system furnace 1 would have the time to do anything, its only that I have LED's representing the furnaces that allows me to see the momentary operation, the LED does not even reach full brightness.

In reality no 2 circuit or relays are exactly the same in resistance and capacitance which is what prevents the relays chattering.

You are introducing a element of component tolerance in the circuit switching, this not a good practice
In your test you are manually setting the thermosats, introducing not knowing witch furnace starts first.
There will be undetermined operation of the relays, Also there is the main part of the system being
not being taken in to account , that is the fuel ignition system could be eratice, Also the thermostats
operation will be random to each other, the possibility of operating at the same time

Regards BoyOh

[pcmanbob]

So I built my simulation again, this time replacing the stats with direct wire connections to simulate both stats calling for heat.

when I switch on the power to the circuit both circuit energise but in my example relay 1 is powered a fraction before relay 2 so the LED representing furnace 1 comes on for a fraction of a second but furnace 2 comes on and stays on. swapping the relays over caused the circuit to operate the other way round with furnace 1 staying on.

now I doubt that in the actual working system furnace 1 would have the time to do anything, its only that I have LED's representing the furnaces that allows me to see the momentary operation, the LED does not even reach full brightness.

In reality no 2 circuit or relays are exactly the same in resistance and capacitance which is what prevents the relays chattering.

You are introducing a element of component tolerance in the circuit switching, this not a good practice
In your test you are manually setting the thermosats, introducing not knowing witch furnace starts first.
There will be undetermined operation of the relays, Also there is the main part of the system being
not being taken in to account , that is the fuel ignition system could be eratice, Also the thermostats
operation will be random to each other, the possibility of operating at the same time

Regards BoyOh

In the above test I was simulating a cold morning were both stats could be calling for heating at the same time to prove you would not get relay chatter.
under normal dally operation which ever stat calls for heat first it will prevent the other one from calling for heat until the first stat is satisfied.
we are not changing the operation of the rest of the system, each furnace will still respond to a call for heat as it has always done, so it will control its ignition / fail safe circuit like it always has. in essence what this circuit does is no different from you turning one stat up to call for heat and turning one stat down to prevent it from calling for heat.

The original brief for the project was to just have only one stat being able to call for heat at any one time by preventing the other stat from calling for heating which is exactly what this circuit will do there was no request for any stat to have priority.

Edit.

To help people understand the operation made a short video of the breadboarded circuit.
https://www.youtube.com/watch?v=Wmm1HFgc3og

[davidcoton]

The original brief for the project was to just have only one stat being able to call for heat at any one time by preventing the other stat from calling for heating which is exactly what this circuit will do there was no request for any stat to have priority.

Nor was there any request for a solution not giving one thermostat priority
That is a missing element of the spec, not a justification for insisting on either solution without clarification. Situation normal when engineers start reading specs with an eye to implementation.
The two-relay solution as given can in theory fail by chattering. It may never happen but there is the possibility of random failures due to the exactly incovenient timing of the two thermostats switching.
The single-relay solution cannot chatter, but has the side effect (desirable or not) of giving one thermostat priority. Its behaviour is more predictable and it is simpler. Unless equal priority (first come first served) is a requirement, I cannot think of a reason to prefer to use two relays.

[pcmanbob]

If the original poster is not happy then all he needs to do is re-write the specifications, until then there are 2 solutions 1 or 2 relays.
Both options will work both have there own advantages - disadvantages.

Of course you could always go back to his original idea and put a pi in there to do the control.

reminds me of the old designer joke.

[drgeoff]

I think I first saw that joke on the lab pin-board in 1972. The drawings were exactly the same but the words were not so closely aligned to software roles. More like https://www.businessballs.com/theme/rem ... g_70s.jpeg

[olsn500]

Ok, first I'd like to thank the mods for moving this to off topic discussion for me. I'd also like to thank everyone that has contributed to my solution. Parts will arrive this later week, and Ill either shoot a short video or snap a couple pics. It's getting cold here in the midwest now, the family and I will appreciate the heat.