Getting the Damn Thing to Light
Most gas valve failures aren't what people expect. You'll pull the panel, stare at a dead boiler, and immediately assume the gas valve is fried. It usually isn't. I spent years on service calls where we replaced three valves before someone finally checked the thermostat anticipator setting, and the whole system was operating correctly. Here is how you actually approach a gas valve problem. Start at the weakest point in the chain and work backward. Not at the valve. That gets you nowhere fast.
Gas Valve Troubleshooting Guide
The valve itself has two critical things going on inside it: the solenoid coil and the diaphragm. The coil either energizes and pulls open a small pilot valve, which releases pressure on the diaphragm, which then allows main gas flow. Or it doesn't. When it doesn't, you need to figure out why before you touch the valve. First thing I do on every call is measure voltage at the valve terminals while the board is demanding heat. If you're reading 24 volts AC and the valve isn't opening, the valve is bad. If you're reading zero volts, the problem is upstream. This single test eliminates about 60 percent of unnecessary parts swaps right there. There is a specific kind of Honeywell valve that causes genuine headaches. The RD80 series, specifically the RV8405Q variants that show up in older Lennox and Goodman furnaces. The internal seat degrades in a way that isn't obvious. The valve will spark, the ignitor will fire, and it will attempt to ignite. Then you hear the characteristic double-bang of a delayed ignition event. The valve is partially stuck open on the pilot side but not on the main side. I've seen people swap ignitors, clean sensors, and check gas pressure for three visits before I pulled the valve apart and found the pilot orifice clogged with a residue that looks like baked-on varnish. Standard cleaning solvent didn't touch it. I used a 0.020 inch drill bit, gently worked it through the pilot port, and the problem disappeared permanently. The valve itself was fine. The port was just choked from natural gas impurities and age.
Gas pressure measurement is another area where people make mistakes. You need a manometer, not a multimeter. I know that sounds obvious, but I've personally encountered technicians who were confident their valve output was correct because they measured 24 volts and assumed the valve was passing full flow. It wasn't. The manometer showed 3.5 inches of water column when the valve's nameplate called for 7. The valve coil was within spec, the transformer was fine, and the control board was sending the right signal. The regulator spring inside the valve had simply lost tension over twelve years. Replacing the entire valve assembly brought the manifold pressure back to 3.5 W.C. immediately. You cannot diagnose this without a manometer. Period. The control board deserves a mention here because it is the most commonly blamed component after the valve itself. A bad relay on the board can mimic a dead valve perfectly. When in doubt, jump the R-W terminals at the board with a paperclip and see if the valve opens. If it does with the jumper, your board is the culprit. If it still doesn't open, the valve or its wiring is the problem. This test takes thirty seconds and saves you a diagnostic trip. Wiring issues are underrated. I once chased a no-heat call for two days. Every test pointed to a bad valve. Voltage was present, gas pressure was correct at the manifold, the transformer was solid. The actual problem was a cracked wire insulation on the valve lead running behind the furnace blower compartment. The vibration from the blower motor worked the fracture wider each cycle until it finally opened completely. Heat shrink and a crimp connector fixed it. The valve was never the problem.
Get the Full Details
Safety note: If you smell gas and the valve won't shut off, close the manual shutoff at the supply line and evacuate. Do not attempt to repair an energized gas valve in an atmosphere with visible or detectable natural gas. This is not a troubleshooting scenario. It is an emergency shutdown scenario. Valve replacement procedure. Turn off the gas supply at the shutoff valve. Disconnect power. Remove the old valve by noting which wire goes where and which tube connects to the manifold versus the venturi. Photograph everything. Gas valves are not universal. The orifice sizes, manifold pressure ratings, and inlet/outlet configurations vary between manufacturers and even between models from the same manufacturer. Putting a Rheem valve on a Ruud furnace might physically fit, but the gas characteristics will be wrong and you could create a dangerous combustion condition. Match the valve exactly or consult the manufacturer's substitution chart. When you reinstall the new valve, hand-tighten the gas connections first, then torque to specification. Over-tightening a brass fitting on a gas valve will crack the body. I've done it. It happened on a Saturday afternoon at 2 PM, and the replacement valve cost $187 plus a trip back to the parts store. Use a torque wrench if you have one. If you don't have one, use two wrenches and stop when you feel resistance, then another quarter turn at most.
After installation, you must perform a leak test with soapy water on every connection before restoring power. Then restore power and observe the ignition sequence. Watch for proper spark timing, pilot lighting, and main burner ignition. Measure manifold pressure again with the manometer to confirm it matches the furnace rating plate. The whole process from start to finish should take about 45 minutes for a standard residential gas valve replacement if you already have the right tools and the correct part. Sometimes the simplest explanation really is the simplest explanation. A loose screw terminal on the valve, a tripped limit switch, a clogged filter that starved the inducer motor. Check those first. They are free to check. The gas valve is not free to replace.
