Watching a Killer at Work: Part 1

Oct. 1, 2007
by Mark Eatherton We have all had this phone call, and even if its not one of our own installations, we still have a moral obligation to address the situation

by Mark Eatherton

We have all had this phone call, and even if it’s not one of our own installations, we still have a moral obligation to address the situation to the best of our capabilities and, we hope, make a good, lifelong customer, and move on down the road.

The call came from an “inherited” customer. Their orphaned systems, two standalone snowmelt systems, were brought to our attention by the general contractor who had hired the original subcontractor that had performed the installation in the first place.

He happened by a snowmelt system we were installing in the same neighborhood, and said “So that’s how you address long driveways, eh?”

I stopped what I was doing and took the time to answer his questions regarding our hydrology and the use of remote manifolds, which, evidently, his contractor had not yet discovered. In any case, he took a business card and said he might consider using us on a future project.

Fast forward two years. The general called me up and said that of the two systems that had been installed two years earlier, one of them had never operated correctly. He wanted me to go look at it and see if I could make it perform automatically, the way it was supposed to.

At the house, I found one of those wonderful blue box snowmelt controllers on the wall, and the readout for the slab was showing an error code that the sensor could not be found. I went out front, 250-ft. away, and found the suspect snowmelt sensor. I carefully removed the sensor from its well, and saw the -in. PVC conduit leading from the sensor, headed in the general direction of the boiler room, which was located in the basement at the back of the house.

When I returned to the basement, I found where the wire penetrated the mechanical room and connected to the snowmelt controller. The only problem was that the wire that was entering the mechanical room wasn’t the same wire that I saw in the sensor cup outside. There was no junction box visible between the sensor and the house, so I had to assume that the splice was made inside the -in. PVC conduit. As I pulled the wires out of the PVC back to the sensor cup location, the splice between the two cables finally came to the surface.

The two wires were overlapped and duct-taped together. No wonder the controller couldn’t see the sensor — it was never connected. I replaced the sensor with a sensor that had a 300-ft. lead on it, made the connection to the control, test fired the unit (it was warm outside) and left. About three months later, at the beginning of snow season, the homeowner called me up and said, “That snowmelt boiler you worked on didn’t turn on and melt the snow.”

I hate it when that happens. Especially with an orphaned system.

I returned to the home to find that the boiler, a 600,000 Btuh copper fin tube product that was atmospherically vented, had locked out on the ignition safety. This meant that it had either tried for ignition three times and misfired on all three tries, or it had lost the flame confirmation signal in the middle of a sustained call for heat.

A simple hard boot reset cleared the fault, and the boiler fired right up as I was standing there watching it, and took right off. There was the usual short spill period, when the boiler first fires and hasn’t quite established a good draft in the cold chimney, but nothing that really caused concern. I cycled the boiler a dozen times and it caught and took off each and every time. I told the homeowner of my findings, and left it running to finish the job of melting the snow that had fallen.

The very next snowstorm, I got the dreaded phone call again, with the homeowner saying, “That snowmelt boiler you worked on didn’t turn on and melt the snow.”

This time, as I walked down the stairs toward the boiler, I could smell the aldehydes from the boiler. I surmised that the boiler was spilling products of combustion into the closed boiler room, which had more than adequate combustion air provisions. I tested the boiler room for possible negative air pressure conditions using a Magnehelic gauge that can read in 100ths of an inch water column, and did not see anything to cause me concern. This left the open draft relief hood as the only possible spot that could be problematic.

The mechanical room that served this boiler was in the basement, was unheated, and was isolated from the rest of the home. The flue gas stack was approximately 30-ft. tall. I surmised that the slug of cold air that stood in the stack during idle periods was so significant that when the boiler attempted to fire, it could not overcome the plug of cold air. Consequently, products of combustion were spilling into the room from the open draft relief hood, thereby snuffing the main flame as well as the pilot flame, and causing the boiler to go into a lockout condition.

This was going to require me to change to the venting system. I must caution you that you should not make these changes unless you are willing to accept full responsibility for your actions, and you would want to make sure that you have received the proper training from the National Comfort Institute ( prior to making these changes.

I blocked the conventional draft relief hood with sheet metal and installed a barometric damper on the downstream side of the appliance to control and regulate the draft on this natural gas-fired, atmospherically vented appliance.

In addition to this, I installed a manual reset spill switch at the branch inlet to the barometric damper, and interfaced that into the ignition control wiring sequence, such that if the draft hood were to inadvertently spill products of combustion, the appliance would shut down. I test-fired the unit for all combustion parameters and I was satisfied that the unit was in safe operating condition.

Once again, at the beginning of the next snowstorm, I received the dreaded call, “That snowmelt boiler you worked on didn’t turn on and melt the snow,” to which she added, “Are you ever going to be able to get it to work like it’s supposed to?”

This will cause the heart of any good contractor to drop to his knees. I asked her to attempt a hard boot restart and the boiler took off as it was supposed to. I then gave her my home telephone number, and my cell phone number, and told her I didn’t care what day of the week it was, or what time of the day it was, but that the next time she heard the boiler pump start (a pronounced bang due to the pump contactor) that she was to call me so I could come by and observe the goings on within her boiler room in real-time conditions to resolve the situation for once and for all. Tune in next month as we continue our observations of a killer at work. Until then, happy controlled combustion hydronicing!

Mark Eatherton is a Denver-based hydronics contractor. He can be reached via e-mail at [email protected] or by phone at 303/778-7772.