The Energy Factor (EF) rating system provides a means of comparing the efficiency of hot water heating systems. Hot water tanks, for example, have EFs ranging from about 0.53 to 0.62 (the minimum energy factor of a 40 gallon gas tank in BC was raised to 0.62 in 2010). This means that only 53% to 62% of the heat generated is actually used, while the rest is lost to standby losses (heat lost while the tank is not in use) and out the chimney (due to the inefficiency of the burners). This compares to the EF ratings of between 0.80-0.98 for tankless water heaters, which should provide energy savings ranging from 23% to 46% compared to hot water tanks (see image).
However, the EF ratings are based on equipment tests conducted in a laboratory setting, and it was unknown how accurately they predicted real world performance.
The Canadian Mortgage and Housing Corporation (CMHC), in cooperation with Enbridge Gas, recently conducted a study on the real world performance of tankless water heaters. In 23 Toronto homes, hot water tanks were replaced with tankless units and the amount of gas used for hot water heating was measured before and after.
The result was an average energy savings of 46%, which is higher than predicted by the EF ratings. However, savings varied widely, from as low as 2% to as high as 66%. Much of this variation was due to changes in the amount of water used. For example, the household that only saw a 2% decrease in energy usage had increased their hot water consumption by 51% after installing the tankless unit, which still indicates a significant increase in efficiency.
On average, homes increased their water consumption by 2% after installing the tankless systems (although this varied widely from as much as a 31% decrease to a 62% increase). Yet the results still suggest that few homeowners greatly increase their water consumption after installing a tankless system (only 2 homes showed more than a 30% increase).
Due to the current low price of natural gas, installing a tankless system would only be expected to save $69 on gas per year. However in some homes the annual savings could be $150 or more. Tankless systems usually cost $2000 to $4000, compared to the approximate $1000 cost of a hot water tank. This means that the payback on a tankless system can be as little as 7 years, but is typically 20 years or more (basically the life of the system).
Homeowners stated that they liked the unlimited supply of hot water, but many did not like the delay in the supply of hot water to the tap (which varies according to the tankless installation, but was reported to average 20 seconds in this study).
Tankless systems also offer several other advantages such as longer life and lack of corrosion (see our tankless water heaters page for details).
The full report can be downloaded at http://www.cmhc-schl.gc.ca/odpub/pdf/67548.pdf?fr=1336131516776
UPDATE: A similar study was performed by the Minnesota Office of Energy Security and produced similar results. The authors suggested that the higher than expected energy savings provided by tankless units could be because the laboratory tests used to determine EF do not represent typical usage in a real home (the testing only involves 6 uses of hot water per day, while usage is more frequent in most homes). The authors estimate hot water tank ratings are overestimated by 23% while tankless units are only overestimated by 10%. ASHRAE is currently developing a new testing procedure for water heaters. You can download the 91 page Minnesota report here.