deltasmith
New Member
I find myself living in a new home that has two problems with the domestic hot water system, and I’m hoping for either encouragement or else alternative suggestions on my ideas about solving these problems.
We moved in less than a year ago, and what we’re not liking is that the on-demand hot water heater seems to add more time and gallons to the amount that we need to flush from the hot water pipes in order to get hot water in the kitchen sink, shower, etc., and that it delivers a “cold water sandwich” if we’re in the shower or shaving and try to conserve water by cutting back the flow for any period of time.
We’re in drought-parched southern California and shaving or showering with taps wide open doesn’t seem like the right solution. On the other hand, gas is cheap and this particular house comes with a small amount of “free” electricity (maybe 100 kWh per month – long story, don’t’ ask unless you want to hear my rant about Solar economics). I’m inclined to minimize my energy consumption, but I want to do that in a “smart” way and the current on-demand technology seems to have a questionable degree of “smartness”.
We recently moved from a rental home that had a conventional hot water tank, but it was a modern tank (very well insulated) and it was fired by natural gas, which is very inexpensive in this part of the country. My gas bill was approximately the same, but we didn’t have cold water sandwiches. We did still have the problem of having to flush 1.5 gallons to get hot water in the kitchen sink. It’s up to 2 gallons in the new house and the on-demand system is probably accounting for about 0.5 gallons (ballpark guess) by insisting on a good flow and a good exhaust draff before it will ignite and then additional time before it reaches max temp output.
Some people claim to have solved the “cold water sandwich” problem that is unique to on-demand heaters by adding a small storage tank and a circulation pump. Other people claim to have solved the problem about needing to flush cold water from the pipes between the heater and the sink by installing a system that circulates hot water throughout the house by momentarily using the cold water supply line as a temporary hot water circulator return line. The Watts 500800 is an example.
A competitor to the better known Watts solution is fasterhotwater.com. They lack the reputation that Watts enjoys, but they have an appealing idea that I might borrow from in my own final solution for my home: they use a flow switch to start the circulator pump. It means that the pump is normally off, but I can engage it from any hot water tap in the house by simply opening the tap. This might not be optimal for everyone in the world, but people in my household could easily get used to the idea of “open the hot water for a few seconds then shut it off and wait for 60 seconds, then open it again and the water will be hot”. This would be an improvement over the current system of “open the hot water, fill all of the empty jugs that are at hand and once those have filled to full, let the water continue to run down the drain until it gets hot”.
I’m tempted to try a solution that involves a small (less than 10 gallon) storage tank that is in close proximity to the on demand heater. A circulator pump(maybe the one that comes with the Watts kit) causes the on-demand hater to maintain the temp of the storage tank. If I add the thermostatic “bridge” valve under the kitchen sink, and if I also position the pump such that it is downstream of the heater outlet (and upstream of the tank inlet, the heater inlet, and the bridge valve under the kitchen sink), then when the bridge valve is thermostatically closed, the system would reduce to a simple on-demand heater with a small storage tank. When the bridge valve is thermostatically opened, the pump would be pushing water through two distribution loops instead of one, and water would flow in inverse proportion to resistance, but I don’t know now to accurately calculate the resistance. Would this design work, or should I invest in two separate circulator pumps in order to solve the two different problems?
Part of me feels like I’m building a Rube Goldberg contraption. Surely there’s a better way?
Creative ideas (and also conventional thoughts on this problem) appreciated. Thanks
We moved in less than a year ago, and what we’re not liking is that the on-demand hot water heater seems to add more time and gallons to the amount that we need to flush from the hot water pipes in order to get hot water in the kitchen sink, shower, etc., and that it delivers a “cold water sandwich” if we’re in the shower or shaving and try to conserve water by cutting back the flow for any period of time.
We’re in drought-parched southern California and shaving or showering with taps wide open doesn’t seem like the right solution. On the other hand, gas is cheap and this particular house comes with a small amount of “free” electricity (maybe 100 kWh per month – long story, don’t’ ask unless you want to hear my rant about Solar economics). I’m inclined to minimize my energy consumption, but I want to do that in a “smart” way and the current on-demand technology seems to have a questionable degree of “smartness”.
We recently moved from a rental home that had a conventional hot water tank, but it was a modern tank (very well insulated) and it was fired by natural gas, which is very inexpensive in this part of the country. My gas bill was approximately the same, but we didn’t have cold water sandwiches. We did still have the problem of having to flush 1.5 gallons to get hot water in the kitchen sink. It’s up to 2 gallons in the new house and the on-demand system is probably accounting for about 0.5 gallons (ballpark guess) by insisting on a good flow and a good exhaust draff before it will ignite and then additional time before it reaches max temp output.
Some people claim to have solved the “cold water sandwich” problem that is unique to on-demand heaters by adding a small storage tank and a circulation pump. Other people claim to have solved the problem about needing to flush cold water from the pipes between the heater and the sink by installing a system that circulates hot water throughout the house by momentarily using the cold water supply line as a temporary hot water circulator return line. The Watts 500800 is an example.
A competitor to the better known Watts solution is fasterhotwater.com. They lack the reputation that Watts enjoys, but they have an appealing idea that I might borrow from in my own final solution for my home: they use a flow switch to start the circulator pump. It means that the pump is normally off, but I can engage it from any hot water tap in the house by simply opening the tap. This might not be optimal for everyone in the world, but people in my household could easily get used to the idea of “open the hot water for a few seconds then shut it off and wait for 60 seconds, then open it again and the water will be hot”. This would be an improvement over the current system of “open the hot water, fill all of the empty jugs that are at hand and once those have filled to full, let the water continue to run down the drain until it gets hot”.
I’m tempted to try a solution that involves a small (less than 10 gallon) storage tank that is in close proximity to the on demand heater. A circulator pump(maybe the one that comes with the Watts kit) causes the on-demand hater to maintain the temp of the storage tank. If I add the thermostatic “bridge” valve under the kitchen sink, and if I also position the pump such that it is downstream of the heater outlet (and upstream of the tank inlet, the heater inlet, and the bridge valve under the kitchen sink), then when the bridge valve is thermostatically closed, the system would reduce to a simple on-demand heater with a small storage tank. When the bridge valve is thermostatically opened, the pump would be pushing water through two distribution loops instead of one, and water would flow in inverse proportion to resistance, but I don’t know now to accurately calculate the resistance. Would this design work, or should I invest in two separate circulator pumps in order to solve the two different problems?
Part of me feels like I’m building a Rube Goldberg contraption. Surely there’s a better way?
Creative ideas (and also conventional thoughts on this problem) appreciated. Thanks