How much power does it take to go from a "cold" start to heated?
Or even from partially cool back to "running"?

For a guest house the notion is straightforward - turn it off when
nobody's there.

But I see these "energy saving" devices that turn off the heater in
the day. My understanding is you use up that energy getting the
water back to its running temp. That you'd do better to insulate
the heater if it's losing that much.

But when my heater needs replacing, we're going for an
'insta-hot'/on-demand type one. Lower costs in the long run
and infinite hot water.

Phil Sherrod wrote:

On 29-Mar-2004, Edward Cheung wrote:


Nice work on the investigation. My minor comment is that you used a
standard line voltage in your
calculations. How important is this? Probably not much. In my case, I
just checked the voltage change
over one day. It peaked at 123.4V, and the min was 118.6V. If one
assumed 120V, there would be
only about 2% error


The line voltage will make no difference in the total energy usage, it just
changes the duty cycle.

The water temperature is set by the thermostat. When the temperature drops
below the setpoint, the thermostat turns on the heating element which runs
until the temperature reaches the thermostat's setpoint. If the voltage is
below the standard, less heat will be generated per minute of operating time,
so it will take longer to heat the water and the heater will run longer -- but
it will draw less power while it is running. So as long as you have enough
voltage to generate sufficient heat to balance the energy loss (70 watts in my
case) the same net energy will be used.

What it boils down to (no pun intended) is the law of conservation of energy:
to maintain an average temperature in the water heater, the heat going in has
to equal the heat being lost. The duty cycle changes, but the total energy
used does not.