Tag: milivolt

First, a thermocouple is not a flame rectifier like a modern flame sensor. A thermocouple actually generates a milivolt potential difference when it is heated by a flame.. Just to get that out of the way for any of you newer techs who are used to modern flame sensors.

With higher efficiency gas fired equipment being the norm for replacement systems, thermocouples and standing pilots are becoming a thing of the past. Newer appliances do not typically utilize a standing pilot, opting instead for hot surface or spark to pilot ignition. These types of ignition systems have benefits over standing pilot, from increased reliability and longevity to higher efficiency numbers. But there are many appliances in the field that still use a standing pilot, and a good service technician should be able to diagnose a thermocouple issue.

Many of you will say-

“Why even check the thermocouple? It’s a 5 dollar part, just throw a new one in!”

“Why are you so lazy? Do you even HVAC in real life or just on the internet?”

Yes, I know thermocouples are cheap and I am all for replacing them when they need to be replaced, or while replacing a gas valve or pilot assembly. But over the years I have seen a lot of guys ( me included) go on calls for pilot issues, find a pilot blown out, relight the pilot, and then because it’s the easiest, quickest fix, replace the thermocouple, only to have the same customer call in a day or two later with the pilot being out AGAIN. And when the tech goes back and relights the pilot, then what? Is that brand new thermocouple bad after a few days? Probably not. There is probably some other issue, but checking the thermocouple millivolt production is the first step for a proper diagnosis.

So how does a thermocouple work? Well, I’m no scientist ( I’m barely a writer), but I’ll tell you what I know. When different metals are joined, and there is a temperature difference between them, a magnetic field occurs between the joints where the different metals meet. The heat of the pilot flame is the source of the temperature difference in a normal pilot system. Through this process, a small amount of current is produced, generally around 30 millivolts. This voltage is sensed by the gas valve and is used to keep the pilot valve internal to the main gas open. If the pilot goes out, the heat that is generating the potential (voltage) is lost, thus current stops flowing to the gas valve, and the pilot valve is closed, closing off fuel to the pilot assembly. The thermocouple is a safety device. If the pilot flame goes and the pilot valve doesn’t close, the burner compartment and potentially the room the equipment is in can fill up with gas. That the consequences of that would require a different article.

When should you check a thermocouple? I am in the habit of checking thermocouples when I encounter them, whether it’s on a maintenance inspection or a service call. If you are in the habit of checking them, it usually doesn’t take more than a few minutes. If the millivolt measurement is less than 26-27, I typically recommend replacement.

To check a thermocouple, you need a multimeter that is able to measure millivolts. It is typically shown as mV or is just the third decimal over on the DC voltage reading. Remember, the meter should be set to DC voltage.

It’s also helpful to have a extra set of hands, but it is very possible to perform this check by yourself if you hold your tongue correctly (or just use alligator clips). First, disconnect the thermocouple from the gas valve. Then light the pilot. Most gas valves have a turn knob that has to be set from On/Off to Pilot. There usually is a push button that is pressed to manually open the pilot valve, sending gas to the pilot assembly in order to light the pilot. The trick is to light the pilot, and position the meter leads in the proper place to read the voltage. The push button must be depressed through the whole check. With the thermocouple being disconnected from the gas valve for checks, the pilot valve should not stay open and the flame should go out when the push button is let up.

Put on meter lead directly on the gas valve side of the thermocouple. Put the other lead on the copper line as shown by my right hand in the picture above. While holding the meter leads in this position, light the pilot. The thermocouple needs to heat up for 30 seconds to 1 minute in order to obtain a proper reading.

30 millivolts is the desired reading, with a swing of plus or minus 5 millivolts. If the readings are in that range, and you have been having pilot failure issues, more than likely there is some other cause. Dirty pilot assembly/ orifice is the most common other issue I encounter, but it could be down draft/flue or combustion air issues, fuel pressure problems, or a failing gas valve. But as stated above, the thermocouple should be eliminated as a potential issue before moving on with a proper diagnosis. Don’t throw parts at a problem and see what sticks. With thorough troubleshooting, you can save a lot of time, headaches, and maybe the customer a little bit of money and frustration.

— Justin Skinner

I received an email from a podcast listener with some furnace related questions. Based on the nature of the questions I figured it would be better to ask an experienced furnace tech. Benoît Mongeau agreed to help by answering the questions. 


My name is Matt and I am a newer tech (fully licensed this September, have been doing the work for 2ish years) who lives in Northern Ontario, Canada. I really enjoy the HVACR school podcast. I don’t do any A/C stuff but I still enjoy listening and wrapping my brain around it. I have always struggled with the theory behind getting cold from hot. The bulk of my work is residential gas heating, mainly high-efficiency furnaces, and gas fireplaces. My questions for you are, (these are just ideas for your podcast though help is never turned down)

On a millivolt system (runs off of a thermopile)
– How to easily test for gas valve failure, what are the expected resistances across the solenoid in the gas valve?
– What expected readings should we consistently get from a properly working system (voltage of thermopile alone, with gas valve open, with thermostat closed etc)

On high efficiency
– What is the relationship between the pressures in the collector box of the secondary exchanger and the pressure switch?
– How does a clogged condensate trap lead to the pressure switch not closing?

Another Question
– Is it possible to check readings from the circuit board when the wires are in a harness? For example, I troubleshot a gas valve failure. It was either the board or the valve. The wires coming to the gas valve from the board are in a harness. How do I know which to check and what am I checking for. (Given that everything else was working I leaned toward a faulty gas valve and was right, just so you know!)

Thanks for your time and for doing the podcast.

All the best,
–Matt


For the collector box/pressure switch:
During normal operation, the collector box is under a vacuum (negative pressure) when the inducer is running. That vacuum is what the pressure switch checks for. If the vacuum is sufficient the contacts will close and signal the board everything is good. If your condensate trap is blocked, the collector box will still be under a vacuum. That doesn’t change.

However, the pressure switch port (where the tube is attached on the collector box) should be at the bottom of the box, usually near the drain port. The backed up condensate will simply end up blocking that port and the switch will no longer be able to ”feel” the vacuum, the contacts won’t make and you will get an error (pressure switch not closing or stuck open).

What may also happen, but not always, is that the port will block during a cycle and the vacuum will remain stuck in the pressure tube. As your inducer comes off and normal pressure returns, the air can’t go in the pressure tubing because it’s blocked with condensate, and you’re basically trapping that vacuum inside. So the contacts will stay closed, until the next call for heat. When that call starts, the contacts will already be closed before the inducer starts, and that will also give you an error (pressure switch stuck closed).

Now if your exhaust is blocked, this will create back pressure and your collector box won’t be under the appropriate vacuum, and once again won’t close.

For millivolt systems:
Unfortunately, I can’t say what typical resistance values would be for a mV gas valve because I don’t know. I would say however that in three and a half years I haven’t had to replace a fireplace gas valve. They rarely go bad. In most cases the pilot tube/orifice is dirty, the thermopile is too weak, or, if it works with a wall switch, very very common: the switch is bad. Standard wall switches are meant for AC voltage.

Running millivolt DC thru them will work, but as soon as you have a bit of resistance in the switch contacts, that voltage will not get through. If it runs on a thermostat, usually it works better but you can still get the same problem.

For typical readings, I’d say between 450-650mV from the thermopile alone, open circuit. With the valve open (so, closed circuit) around 200-300mV. But this is very general, it may vary a lot between systems.

If your thermopile alone doesn’t produce enough mV’s, check your pilot flame. Make sure it hits the thermopile well. You might be able to adjust it (on some valves) to make it bigger. As I mentioned, the orifice or tubing may be blocked. That is relatively common especially if the pilot was kept off for a long time.
If your thermopile gives enough voltage but the valve won’t open, check your switch/tstat and even the wire itself for any significant resistance or short.

Isolate section by section and ohm it out. If everything is good and sufficient mV’s come back to the valve and it still won’t open, then yes, that valve might be bad. But I’d probably even replace a switch/tstat before I condemn the valve regardless, just to be sure, just because changing those valves in most cases is a total pain in the butt.

For the gas valve/board dilemma:
If your wires are all in a harness with a big fat connector on the board, there’s a good chance you won’t be able to pull it off and diagnose on the board pins, because by removing the connector you remove most or all of the safety circuits.

If you want to look at the gas valve, you need to hook your meter on the wires at the valve itself. If it’s just a standard 24v valve with 2 or 3 terminals (Common + hot or common + low and high solenoids) just pull the wires off (or connector) at the valve and you have to check for 24V on the wire across common and hot. Even with the valve disconnected if your board is OK it will still send 24V in that wire at the proper time in the sequence of operation (i.e. wait until the ignition sequence completes!!). If you don’t have 24 volts, the board is bad. If you have 24V, the gas valve is bad.

If it’s one of those Honeywell SmartValves, then that’s another story entirely. A good portion of the controls are actually inside that gas valve and it will have multiple wires going to it. They are a bit more difficult to diagnose. My best advice is to follow your electrical diagram. If there’s no way for you to disconnect wires at either end (which should never happen as far as I know…) you could always cut the wire and check your voltage in the wire itself. But try to avoid doing that.

— Ben

Scroll to top
Translate »

Daily Tech Tip

Get the (near) daily Tech Tip email right in your inbox!
Email address
Name