Non-powered gas valve pilot safety circuit

[WattsThat]

On devices like a gas fired hot water heater here in the US, the gas control valves have a pilot safety system controlled by a thermocouple.

The thermocouple output voltage of perhaps 40-50 millivolts (when at gas flame temperature) energizes a solenoid which allows a small amount of gas to flow into the pilot nozzle so long as the pilot remains lit. Should the pilot be extinguished, the thermocouple output drops to a few millivolts, the solenoid opens, and pilot gas flow stops.

My question this: these units have no external power. How can you construct a wound coil solenoid that actuates with virtually no current and very low voltage with the result being virtually no power. Sure, you can have a permanent magnet in the core but I’m puzzled over how you can move against and hold a spring loaded plunger with such a very low energy source like a thermocouple.

These devices have been made for fifty years or more. It’s rock solid, simple and proven technology. But, I can find virtually nothing on line about the design of such a solenoid. Anybody out know anything about the devices?

Please, no conjecture. You either know and can explain it or you’re just making noise and waving your hands. A typical value and operation manual:

https://www.amazon.com/Robertshaw-110-326-Heater-Thermostat-Natural/dp/B005HONZZ6/ref=psdc_2232427011_t2_B001UJS63I#descriptionAndDetails

https://www.robertshaw.com/WorkArea/DownloadAsset.aspx?id=2147485045

[Manul]

How can you construct a wound coil solenoid that actuates with virtually no current and very low voltage with the result being virtually no power.

The "virtually no current" part is wrong. Thermocouple has low output voltage, but also really low impedance. If load resistance is very low, current is certainly not small. If voltage is 15mV and overall circuit resistance 100mOhms, current is 150mA. That is not a small current. If coil has 10's of turns, it is not a very small magnetic field. After that what is left is good mechanical engineering and it is not my speciality, sorry.

[Zero999]

https://en.wikipedia.org/wiki/Thermocouple#Gas_appliance_safety

[WattsThat]

The "virtually no current" part is wrong. Thermocouple has low output voltage, but also really low impedance. If load resistance is very low, current is certainly not small. If voltage is 15mV and overall circuit resistance 100mOhms, current is 150mA. That is not a small current. If coil has 10's of turns, it is not a very small magnetic field. After that what is left is good mechanical engineering and it is not my speciality, sorry.

Thank you, that fits with other info that I’ve read. I have no idea what the specs are for the solenoid coil but from what I’ve seen, they appear to have fairly heavy gauge wire (perhaps #18) and judging from the diameters, 20 turns or more would be quite possible. If a thermocouple can source 100-200 mA, that could provide enough flux to hold against the spring force.

I’m in scrounge mode for an old gas valve to take apart! Even at a few hundred milliamps, we’re still talking about less than 5 mW of power.

[Manul]

I’m in scrounge mode for an old gas valve to take apart! Even at a few hundred milliamps, we’re still talking about less than 5 mW of power.

Electromagnets are actually not about power. I will give you a hint: can you make electromagnet from superconducting winding? Yes, you can. What will be power? Zero. You can excite current inside supercuducting ring and it will continue going by itself (called persistent current) producing zero power (no loss) and creating magnetic field (theretically - forever).

[gcewing]

I’m in scrounge mode for an old gas valve to take apart!

Maybe we could persuade Big Clive to purchase one from eBay for examination? Sounds like the kind of thing that would be up his alley.

[amyk]

There's already quite a few videos on YouTube about this, these just came up in a preliminary search so I'm sure there are probably better ones:



(In an annoyingly robotic voice, but still educational...)


The summary is that the solenoid is not powerful enough to actually pull in the valve, but once it's closed (manually), the magnetic circuit is closed and has a much smaller reluctance so the force required to open it again is much greater. Since the valve should close quickly once the flame goes out and the thermocouple begins to cool, the magnetic force is just barely enough to oppose the spring tension.

[wizard69]

I’m in scrounge mode for an old gas valve to take apart!

Maybe we could persuade Big Clive to purchase one from eBay for examination? Sounds like the kind of thing that would be up his alley.

Not the big man here but decades ago I did work in a die cast foundry.   There was a huge variety of valves operating with these thermocouples, and I think the big factor was lots of turns of very fine wire.  At least that is what I remembered after a fork lift partially disassembled a valve for me. 

[Zero999]

I’m in scrounge mode for an old gas valve to take apart! Even at a few hundred milliamps, we’re still talking about less than 5 mW of power.

Yes 5mW seems sensible. Normally the valve is opened by the user pushing the dial inwards. The coil is only used to hold it open and the gas pressure also helps.

Look at a relay data sheet to give you an idea. The drop-out voltage is often around a tenth of the nominal voltage, which is a power dissipation of a hundredth.
https://www.farnell.com/datasheets/2366559.pdf

[floobydust]

You can find datasheets for the thermocouples (typ. 30mV), thermopile generators (typ.750mV) and the gas valves. The operating current is never mentioned, I faintly remember it could be an amp but losses in resistance are high and any parasitic (metal) junction gives unwanted thermal EMF so the wiring is done one way. Current is something you can't easily measure. The solenoid just has minute hold-in power with mechanical gain (lever) and springs.

Honeywell V800 standing pilot valves 750mV, 2 ohm coil and 3.4ohms system total resistance gives 220mA.
Honeywell V400 standing pilot valves 30mV, 0.02 ohms system total resistance? gives 1,500mA.
Hot Junction: 1400°F (760°C), Cold Junction: 780°F (416°C)

These systems are rarely used today because they are wasteful the pilot is always on, slow- it can take a long time to react to a flame out and by then, some large burners will build up with fuel which is bad. There are combustion safety standards like NFPA-85, 86 or 87, that dictate the pilot-control response times, which are also dependent on the burner size. Electronic control is the norm nowadays.

For more power look, at thermoelectric generator (TEG) systems. NASA uses Radioisotope Thermoelectric Generators for power on the Mars rovers, Voyager, Apollo etc. i.e. 768 thermocouples for 30VDC. Natural gas is also used for remote areas.

[mikeselectricstuff]

I just measuired the short-circuit current of a gas boiler thermocouple at 300mA. Bear in mind that in this application, it only needs to generate current, not power, as no energy is needed to create a magnetic force. other than that needed to overcome electrical resistance ( the same force could be created by a permanent magnet).

The devices that replace these nowadays are quite interesting- high voltage spark igniters which use the spark electrode to measure ionisation current through the flame to cut off the gas if the flame goes out.

[Zero999]

These systems are rarely used today because they are wasteful the pilot is always on, slow- it can take a long time to react to a flame out and by then, some large burners will build up with fuel which is bad. There are combustion safety standards like NFPA-85, 86 or 87, that dictate the pilot-control response times, which are also dependent on the burner size. Electronic control is the norm nowadays.

They're extremely common in gas stoves in the UK. Each burner has its own thermocouple and valve. There is no pilot light. The user pushes the valve in, activating the ignition, which is normally a mains powered coil, but can also be a piezo in cookers which don't require a mains supply. Once the flame is lit, the user has to hold the valve in for a couple of seconds, until the thermocouple generates enough current to hold the valve open, then they can release it.

[floobydust]

Gas stoves I see are using flame rod/igniter per burner, this is where flame detection and ignition sparker is done at a single Kanthal rod, which is a very fast system. Press and hold and wait can be too much for North America lol. But these stoves are less popular here, it's mainly electric and now apparently gas stoves prices have gone up as they are a chef's special and considered "exotic" 

[Zero999]

Gas stoves I see are using flame rod/igniter per burner, this is where flame detection and ignition sparker is done at a single Kanthal rod, which is a very fast system. Press and hold and wait can be too much for North America lol. But these stoves are less popular here, it's mainly electric and now apparently gas stoves prices have gone up as they are a chef's special and considered "exotic" 

My stove doesn't take that long to light. I didn't know Americans are that impatient. I do like my gas stove. It's much quicker to heat, than an electric hob and probably more efficient too. I admit an induction hob would be more efficient, but I like to be able to use my gas stove, even if there's a power cut. It saves having to have a separate camping stove, for that eventuality and I can also use it to heat my house, without electricity.