Recently the gas furnace in my rental home in Winnipeg stopped, and it was freezing out. I knew the previous renters replaced the furnace motor at my suggestion for a rent reduction. It had an electronic spark ignition for the pilot light. I had a wire beside it to sense the flame temperature. The sensor wire was free to rotate in the bracket. I surmised that the wire moved out of place. If the wire were out of place, it would stop the furnace from going from pilot to operating mode.
My first move was to use adhesive, which could secure the rotating sensor wire in place even at high temps. Then I bent the wire so it sensed the outer cone of the blue flame, rather than the center. Voila, the sensor heated up quickly, and the main flame started a few seconds later. The furnace appeared to be fixed, and the wire shouldn't drift with vacuum service.
It was wise that I had used insulated-handle pliers to adjust the flame sensor wire as it tends to arc from the igniter to flame sensor when metal fills the gap and the pilot arc starts. Otherwise you have to hit the furnace kill switch, adjust it, then turn it on and wait 45 seconds or so until the arc starts on the pilot gas.
So it was fine except that half the time the furnace motor wouldn't cut out after the flame stopped. Yet recycling the power on the slow-speed switch ON reset the motor to the off state. Hmm. Perhaps excess dust was insulating the hot air sensor now, and the reading tool was offset on the high side from stiction with hysteresis on the threshold. Recycling the power seemed to fix it.
On checking the spring thermostat on the furnace hot air duct, I found it just needed to be raised on one of three settings, namely the one for flame-off low-temp fan cutout. I adjusted it from 90F to 100F, and that fixed the problem. Obviously the spring had gone out of calibration somehow.
Then, two days later, the thermostat stuck on the ON position even when disabled. A low battery indicator was on the digital thermometer. Normally this would not leave the furnace stuck ON, so I bought three new AA batteries and hoped that would fix that issue. Removing the panel from the wall stopped the furnace even when set to the OFF state. I lucked out. I didn't really want to buy a new $5,000 furnace now after spending $5,000 on materials and tools to rent this house.
This entry was submitted by Anthony Stewart and edited by Rob Spiegel.
Anthony Stewart started life as an EE design engineer in 1975. He has worked in telemetry design, aerospace design, and nuclear robotic inspection design. He has also designed T1 test equipment.
Tell us your experience in solving a knotty engineering problem. Send stories to Rob Spiegel for Sherlock Ohms.
I agree with "thechas" that adjusting the fame-on sensor is a bad idea, not because it will likely break but because the flame characteristics rather than the sensor position are the problem. I have seen this many times especially on LPG/NG fired furnaces - obstruction(s) in combustion air intake; excessive fuel; damages/obstructed burner and/or flame shaping assembly can independently or combine to change the flame characteristics. Simply moving the flame-on sensor defeats the purpose of it being in the correct OEM position and will often result in subsequent secondary failures.
If the flame characteristics are not correct and the unit is forced to continue operating such as described here, the three most common subsequent failures will be damage to the burner assembly. the fire chamber and/or the heat exchanger assembly. A flame running too high above the burner assembly not only wastes fuel but more often than not will eventually cause the heat exchanger to rupture allowing deadly and dangerous combustion gases to enter the domestic air plenum.
Flame characteristic issues are not limited to gas-fired units, I ran into a similar situation with a waste oil fired furnace being used to heat a commercial paint booth. An experienced plumbing & heating technician was called when the unit began shutting down before the paint booth was up to temperature. Said technician "adjusted" the flame-on sensor which forced the unit to run with the incorrect flame characteristics. Said technician was called back a few hours later when the unit again shut down before set temperature was reached in the paint booth. Technician determined the problem to be a malfunctioning high-limit t-stat on the heat exchanger assembly which he "corrected" by using a couple flat washers to create an air-gap between the t-stat and heat exchanger housing. The furnace continued to operate in this condition for approximately two months until the paint booth suffered a fume flashover explosion.
Post incident investigation discovered the "corrective actions" taken by the alleged technician allowed the flame to directly impinge on the heat exchanger resulting in the structural damage that eventually allowed the combustion flame to penetrate the heat exchanger.
Since the furnace did not have a standing pilot, the flame senor most likely does not sense temperature. Depending on the age of the furnace, the flame sensor either senses the resistance of the flame, or checks the diode properties of the flame.
Either way, cleaning the sensor, the insulator, and perhaps the flame hood would have been a better fix than bending the sense probe. These flame probes are usually made of hard and brittle materials to stand up to the heat of the pilot flame. Very lucky that the sensor did not break.
Just to qualify, years ago I managed the test equipment department at a company that makes furnace ignition controls. In the early 1980's, we switched from raw flame resistance to the more reliable flame rectification method of sensing the presence of the pilot flame before turning on the main burner.
As to the plenum sensor for the blower motor, the standard for years was a coil wound bi-metal spring that operates the on and off switches. The coil does stretch out over time, causing the exact problem you had.
Electronic thermostats are another issue. Some of the electronic thermostats draw power off of the 24 VAC that runs the furnace controls. If the power supply circuit draws enough power, the furnace control will turn on and run the furnace until the overtemp switch (part of the plenum sensor) opens the circuit and shuts down the furnace until it cools. The overtemp shutdown usually ends up with the house temperature being in the upper 80's to low 90's.
Of course, all of this is why you should have your furnace serviced every fall as regular preventive maintenance. Newer furnaces that use a hot surface ignitor to directly light the main burner and have powered exhaust are even more critical to have regular service.
I did have an interesting furnace service problem this past fall. A less technical co-worker who was renting a house called and said that the furnace would not run. The flame would come on, then go out after a few minutes, but the blower would not run. My first service guess was that the unmarked switch for the blower, mandated by the safety code, was in the wrong position. That was not the problem. When I opened the service cover of the furnace to look at the problem, I observed a very cheap switch installed as an unneeded blower safety interlock. The concept being that some fool would plut their hand into the spinning blades of the blower, and be injured. Unfortunately the metal of the housing had flexed just enough to allow the switch to open. I bent the metal back the other way, and very carefully replaced the service cover. This time, a short while after the flames came on, the blower started, and my co-workers house began to warm up. So the rental agent did not need to be called, and the problem was solved. At least until metal creep released the switch tyhe next time.
It seems that quite a few of the problems are related to those two thermostats that monitor the temperature of the heat exchanger. The interesting part is that a few years earlier there were far fewer problems, using switches of an earlier design. The very most reliable switches used a mercury switch bulb as the switch element. Those switches never failed under most conditions, since the mercury contacts were new for each operation. The mechanical life was much longer as well, since there were no stressed elements needed to create a snap action to quickly open and close the contacts. But then somebody came along and pointed out that mercury is toxic, and must be banned from all places forever. The result is switches that fail, both open and closed, usually at inconvenient times. How many of those steel cased mercury switches ever released their little blob of mercury into the envirronment? None of them that were burried without being crushed, and my guess is that even when most furnaces were mashed down, the steel structure of the switch housing protected the glass capsule. How long does it take a glass capsule to leak when it is buried underground? Can anybody answer that one?
Fairly simple appliances? Front loader washing machines with microcontrollers in place of mechanical cycle timers? I actually bought one with a mechanical timer as they tend to be more reliable than the electronic controllers that get zapped by power line spikes. But even my "simple" front loader has a variable speed DC motor controller for the drum with some sort of logic. Fingers crossed that it survives as long as the mechanical timer.
I recently replaced Honeywell mercury thermostats and furnace zone and aquastat controls with their computerized system to squeeze out some more savings on heating oil. Although the old T-stat wiring was only two wire for each zone, Honeywell makes smart replacements that establish 2-way communications and power on the same pair. Every room thermostat displays the outdoor temperature as well as the room ambient temperature, set point, icon for occupied and unoccupied setback, icon for calling heat and an LED backlight. The controller down cellar was easy to install but for one quirky design element. The controller has built-in relays to directly operate 120VAC zone pumps, but nothing for the oil burner! You have to connect a conventional Honeywell zone relay box to the low voltage control terminal if your oil burner does not have its own low voltage control loop. And most don't.
Anyhow, household appliances are becoming more owner unserviceable, just like automobiles, as their complexity expands due to availability of "cheap" silicon.
I agree, Jennifer. These stories get particularly good when the problem is so unlikely it's amazing the Sherlock writer figured out the problem at all. Here are two particularly good ones:
The level of expertise in dealing with the problems detailed in our Sherlock Ohms and Made by Monkeys blog never cease to amaze me. Oftentimes, it just shows what one can accomplish with a little patience and common sense.
Or, in this case, the joys of just being responsible for a house. At least the applicances that we have to deal with are fairly simple and pretty reliable. You certainly had your hands full with this sequence of problems! Nice job.
From Dell / Intel® New Paradigms in Design Work Scott Hamilton, vertical market strategist for Dell Precision workstations, 5/2/2013 5
Early in my career, I worked as a draftsman and remember the days of drawing on vellum with numbered pencils and Mylar with plastic lead. This was a fun experience in the sense that I ...
I've been using workstations for more than 10 years and love finding ways to get more performance from my system. With demanding professional applications that require more power each ...
A lasting memory from my first job as an engineer in an auto assembly plant is standing on hard concrete at six in the morning, vending-machine coffee clutched in hand, listening to ...
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This radio show will show what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.
To save this item to your list of favorite Design News content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Tweet This
7 
