I have had no problems with this 15 year old furnace, until this heating season. This happens about once a week: The unit sees the call for heat, and the ignition sequence usually completes normally... inducer, igniter, gas, blue flame. But the blower motor won't run after the normal delay time, say only once every 6 or 7 days. With that flame on continuously, the heat exchanger gets very hot, and a burning smell flows out of the nearby ducts. The diagnostic LED is happy - it is on steady in this condition. The fan has always kicked on at the next cycle as soon as I cut and restore power. What would be the likely cause for the blower to be intermittent like that? Also, shouldn't a limit switch cut the gas in this condition? It has constant flame to the HX but no air flow. It must have been like that for hours as the ambient temp was 4 degrees below the thermostat setting. I have come home from work and seen this twice now in the past 2 weeks.

Does the system also have A/C? If so, will the blower run continuously when the thermostat fan selector is set to "ON"? If so, you may have a problem with the circuit board - either a burned path, bad relay, or melted solder. If the board is original, you may have to open the plastic casing to view. In any event, the main limit control should be activating to shut down burners during overheat. I hope it hasn't been bypassed, or replaced with one that is too "hot". Post complete model # to allow parts look-up.

Thanks. It is a Bryant 90+, Model 350MAV036040AAKA.

The limit switch is in series with the flame rollout switch, and neither appear to be bypassed. The unit is upflow, equipped with A/C, and the fan does run when I turn tstat fan switch to ON. But this is an intermittent problem, last occurrence was 10 days ago until yesterday. Does that prove anything?

I've included the wiring diagram/schematic of the aftermarket control board available for this furnace. What would you suggest next for this problem that seems to happen only once a week? New board or capacitor? Could it be the motor?

http://www.icmcontrols.com/downloads...90furn-lab.pdf

http://www.icmcontrols.com/downloads...90furn-lab.pdf

The listed main limit control for this unit is part # HH12ZB280. That is about the hottest limit made for these units, so unlikely a replacement is over-temp. Check part # of yours. After running for a while, does the blower motor become too hot to touch (careful)? If so, motor and/or capacitor may be the problem. If not, motor control (circuit board) is the likely culprit.

Thanks again very much. I suspect the fan relay on the control board first. Then the heat side if the fan hi/lo relay contacts. So I'll replace the board. Carrier/Bryant issued a field upgrade kit, and the control board was relocated to inside the blower compartment about 10 years ago.


The unit is upflow, and is located in the middle floor of a 3 story townhouse. With the blower off, I think that there is just enough natural convection to prevent the limit switch from tripping.

Can you advise the part number for the blower motor? Just in case I need to change it.

The OEM is pretty expensive; go with an aftermarket. If 1/3 HP, seek out TOTALINE part# P257-8585. Matching capacitor is part# P291-0503.

It happened again today while I was out shoveling snow. I turned the fan switch ON, and set the tstat to HEAT, a few degrees under room temperature. The fan was running when I went out. When I got back, the Tstat was calling for heat, and I had lit burners, but no blower. I made a fist and knocked on the top of the blower compartment a few times - the blower turned on.

Check the circuit board as previously described. Look for melted solder joints, etc.

Thanks, you were correct. I found a burned solder joint on the load side of the fan relay, at the PC pad common to the L1 terminal. I will replace the relay and repair the board. Meanwhile, I put an ICM aftermarket control board in there, which is working fine. So I will have a spare board after repairs. I also put in a new 5 uF capacitor while I was in there.

Good deal. If the old board has a burned path(s), good luck repairing it. Keep in mind that this circuit is subject to more current than most every other on the board, so good continuity is at a premium.