Propane Conversion With a Regulator It Cannot Hold
A 120,000 Btu/h condensing furnace is field-converted from natural gas to propane. The kit's LP orifices were installed and the new label calls for 10.0 in. w.c. manifold pressure. Measured at the valve outlet with the furnace alone at high fire: 11.9 in. w.c.; when the propane range also fires it climbs to 12.6 in. w.c. Inlet at the furnace is 11.5 in. w.c. static, and when the range and the tankless water heater run, inlet falls to 9.6 in. w.c. and manifold pressure falls to about 8.3 in. w.c. Ignition is delayed 3-5 seconds after the igniter reaches full glow and the burner "thumps," sometimes rolling flame at the burner box; there have been two lockouts this week. Flames are yellow-tipped and lazy; the flue reads O2 2.8% and CO 320 ppm air-free at high fire. Soot streaks are visible at the burner compartment louver. The installer says the kit was complete and the tank regulator is fine because the tank is 70% full.
1. Lay out the diagnostic sequence. Why is the igniter not the first suspect, and what do the pressure readings tell you?
Read the worked answer
Late ignition after the igniter has completed its full warm-up points at fuel delivery and air-fuel ratio, not ignition. The igniter is doing its job; the gas is arriving late or wrong. So the sequence starts at the tank and works toward the burner: tank level and first-stage regulator (ice, vent orientation, a regulator vent plugged by insects or paint), the second-stage regulator's capacity and vent, the line size and length for 120,000 Btu/h plus the range and tankless, the appliance's inlet pressure, and only then the manifold and burner.
The pressures are the diagnosis. With only 11.5 in. w.c. static, this valve cannot reduce to a 10.0 in. w.c. outlet and hold it - the regulator needs headroom above its setpoint, and a service regulator starved of inlet will go wide open and track the inlet instead. That is why the outlet reads 11.9 to 12.6 in. w.c. And when the other appliances fire, inlet drops to 9.6 in. w.c. and the outlet collapses to 8.3 - below the setpoint, so the burner goes lean with flame lift. One appliance, two opposite failure modes, both from an inadequate or incorrectly configured LP supply and an unverified regulator spring.
2. Quantify the effect of a 11.9-12.6 in. w.c. manifold on a burner designed for 10.0, and explain how that produces O2 2.8% and 320 ppm CO.
Read the worked answer
Flow through a fixed orifice varies with the square root of pressure. At 11.9 in. w.c. the ratio is sqrt(11.9/10.0) = 1.09, about 9% overfired; at 12.6 in. w.c. it is about 12%. Add the possibility that the valve is out of adjustment or the spring is the wrong one and the drift explains the soot. Nine to twelve percent over on propane is enough to richen the mixture past the primary air the burner can entrain.
At O2 2.8% the excess air is only about 2.8 / (20.9 - 2.8), roughly 15%, and LP's much higher stoichiometric CO2 (~13.7%) makes that a rich mixture in absolute terms. Rich burning produces yellow, lazy flames, carbon, and CO - 320 ppm air-free is a sooting condition, and the soot streaks at the louver confirm it. Meanwhile, when the other appliances run, the manifold collapses to 8.3 in. w.c., flow falls, the flame lifts, and you get the delayed light-off and thump. Either way the conversion is not correct, and soot on the heat exchanger plus a delayed-ignition roll-out is a CO and fire risk, not a nuisance.
3. What must be verified and corrected before this furnace is returned to service, and what results would you accept?
Read the worked answer
Confirm the conversion is actually complete: LP orifices matched to model and altitude, the LP regulator spring if the valve requires one, the conversion label filled in with the orifice size and manifold setting, and the burner/air components the kit specifies. Then correct the supply: verify the tank's regulators, the line size for the connected load, and the regulator vent termination; set the manifold to 10.0 in. w.c. with a calibrated manometer at steady high fire and re-check it with every other propane appliance running. Leak-test every joint you touched or disturbed.
Accept only stable numbers: inlet pressure that stays within the appliance's listed range under all-load conditions, manifold holding 10.0 in. w.c., and a combustion analysis at high and low fire showing roughly O2 4.5-5.5%, CO2 near 10-10.5%, and air-free CO under 100 ppm, with no yellow tipping and no soot. Verify a supply-air CO reading against ambient before you leave, clean the soot from the heat exchanger and burners, and confirm no flame roll-out on light-off over several cycles. If the supply cannot be made stable, the furnace stays off and tagged - a delayed-ignition roll-out on a propane appliance can scorch wiring and crack the heat exchanger.