The recent ATO thread by Les Sanders reminded me that this was on my to-do list. We all know that these ATO clocks run for a long time on a battery. They're efficient and require very little pendulum impulse to run. I'm currently working on one of these clocks. It looks like one of the normal Kundo versions but is labeled "Seth Thomas" I expect it's just a rebranded Kundo. I decided to try and measure how much battery power it actually takes to run the clock. With 1.5 volts applied, it takes an average power of about 53 microamps. With .95V applied (the minimum needed to run the clock) the average current dropped down to 31uA! The pendulum was actually still swinging below .95V but, no longer had a large enough amplitude to engage the pawl. This clock is powered by one alkaline C battery. These typically have about an 8Ah capacity. How long could the battery last? Take 8Ah divided by 53uA and you get about 17 years of life. This is ignoring the fact that the current would slowly drop to 31uA as the battery drained. You won't actually get 17 years from a battery. The typical shelf life for alkaline is commonly listed as 5 to 10 years depending on the temperature. The battery self discharge rate is certainly a big part of the battery life of this clock. The method I used: The clock was powerd with an adjustable DC power supply. A 510 Ohm resistor was placed in series with the power supply. The power supply voltage was adjusted higher to account for the voltage drop across the resistor. The voltage drop across the resistor was measured with an oscilloscope. With 1.5V power to the clock, the voltage drop was 125mV. With .95V applied, it was 75mV. These voltages are during the time that the coil is energized. The voltage across the resistor was used to calculate the curent draw. (125mV/510 Ohms). About 245uA at 1.5V. The clock runs at a 1.5Hz rate (666.6ms period). The coil is energized for about 145ms per cycle. This is a duty cycle of 145ms/666.6ms (21.75%). 245uA * .2175 for an average current of 53uA. The clock will also be drawing a small amount of current when the coil isn't energized. The quiescent current. I didn't bother to measure or account for this. I have several different makes and styles of ATO clocks. When I have nothing better to do, I'll perform this test on them and see how they compare. I also plan on doing this with quartz clocks. On the scope screen captures, the vertical axis is 50mV per division and horozontal is 100ms per division.