To make repairs above the front porch ceiling in our 100-year-old house I had to remove the ceiling light fixture. Afterward I was surprised to find that the doorbell didn't work. The only electrical change I had made was to disconnect the fixture. Thinking the doorbell outage was coincidental, I set about troubleshooting the doorbell circuit.
The transformer was mounted on a junction box in the basement and connected to the single AC circuit inside, which entered the box on one side and exited on the other. I quickly determined that the circuit in the box was completely dead. I checked the panel to be sure no breakers were tripped, then attempted to trace the wiring into the walls, etc.
Finding nothing obvious, I decided to backtrack. I reconnected the front porch light, but used a test socket with a 20W bulb instead of the fixture I had removed. When I flipped the switch the front porch light came on as expected. I turned off the switch and tried the doorbell. It went "clunk, clunk," which seemed OK because that's the way it sounded before. I pressed the button several times and noticed, to my astonishment, that the front porch light came on every time I rang the doorbell! It wasn't very bright, but it definitely came on, leaving me wondering how the low-voltage doorbell circuit could be controlling the 120V light circuit.
After thinking about it, I sketched a schematic of the way the circuit should be so that I might figure out what could be wrong to cause things to act this way. Looking at possibilities, I retraced the wiring and discovered that the junction box mentioned earlier was not on the AC line but on the switch circuit for the front porch light, and on the neutral side of the line at that! The transformer primary was connected across the light switch and thus in series with the front porch light, but only when the switch was off.
When the light switch was on the primary of the transformer was dead-shorted and the doorbell could not possibly work. Nor could it work if the light bulb was burned out. We hadn't owned the house long and had simply never noticed that quirky behavior. Nor had we noticed that the light came on when the doorbell rang, probably because the current was too low to cause a higher-wattage bulb to glow.
To fix the problem, I moved the transformer to an active AC circuit. The doorbell rang with a crisp ding-dong and pressing the button had no effect on the light. Problem solved!
But wait, why did it behave the way it did? Then it dawned on me: reflected load!
The load on the secondary of the transformer appears on the primary as the actual load times the square of the turns ratio, or: Zp = (Np/Ns)2 x Zs, where Zp is the apparent load impedance on the primary side of the transformer, Zs is the actual load impedance on the secondary, and Np/Ns is the ratio of primary turns to secondary turns in the transformer winding.
With the light switch off, the current flowing through the light bulb and the primary of the doorbell transformer was minimal. The secondary of the transformer was connected in series with the doorbell and the doorbell button. When the button was not pushed, the secondary circuit was open, making the reflected load impedance on the primary side very high. When the button was pushed to ring the doorbell, the low secondary impedance was reflected through the transformer, lowering the primary impedance dramatically and causing enough current to flow in the lamp to make it glow.
The incorrectly wired doorbell went "clunk-clunk" instead of "ding-dong" because the voltage across the primary of the transformer was reduced by the voltage drop across the light bulb.
I subsequently used this bizarre example when teaching reflected load to my students at Vermont Technical College. With a standard 40W light bulb in series with the primary of a doorbell transformer and the AC line, I connected a decade resistance box across the secondary. As I reduced the load resistance on the secondary, the lamp grew increasingly brighter. It brought a mathematical concept to light!
Currently a Radio System Test Engineer with Harris Corporation, C. Michael Reeves has held numerous design and management positions in his diverse career on both commercial and military programs. An electrical engineering graduate of the University of Kentucky, he is a licensed professional engineer and a retired officer from the US Navy Civil Engineer Corps. From 1999 to 2003, he was assistant professor and chairman of the electrical engineering technology department at Vermont Technical College.
Tell us your experience in solving a knotty engineering problem. Send stories to Lauren Muskett for Sherlock Ohms.