Hot water re-circulation: two questions

I did a major edit after I answered one of my own questions, so it's no longer two questions...


Purchased this house with existing propane fired hot water tank with a circulation pump and dedicated return line. It works great, but is just too wasteful of propane to run the way it's set up. I'm on a well and septic, so I understand the rationale for having it, but I want to minimize the cost of running it. It looks like, originally (20 years ago when the house was built), the circ. pump was wired into a timer. Somewhere along the way, the pump was disconnected from the timer and wired to an on/off switch in the kitchen. So, when we bought the house, the procedure was to go to the kitchen and turn the pump on whenever you needed hot water. Not especially convenient if the object was to wash your hands in the upstairs master bath. At the time, I had a lot of other things occupying my time, so I just added another switch upstairs (i.e., three-way switch) so it was a bit more convenient, but we still have the issue of forgetting to turn the pump off, especially if we are running the dishwasher after everyone has gone to bed, for instance. Also, from upstairs, it's impossible to tell if the pump is running if the house is noisy, so the 3-way switch is problematic because you never know if you're turning the pump on or off. So, I'd like to change the set up.

I would like to add some sort of control. Timers, temperature sensors and on-demand switches all have drawbacks in my situation. Programmable timers seem inefficient because you're simply running the pump for larger blocks of time, a little better then 24/7, but not much better. Using a temperature sensor to keep the water at a pre-determined temp. also seems too wasteful. I wouldn't mind an on-demand set up, but I'm not sure it would work for the dishwasher example above, since they appear to only heat for one cycle after you push the button before shutting the pump off (our dishwasher runs for two hours). I think this push button, on demand set up would work for a shower, because it sort of primes the hot water which keeps coming after the pump shuts off, but the dishwasher is intermittent, as is the laundry machine. So, without someone to keep pushing the button, the dishwasher or laundry would receive cold later in it's cycles. I understand that I could also add a timer, but I'd like to keep it simple, as convenient as possible, and I'd like to run the pump as little as possible for efficiency. The electricity cost to run the pump isn't the issue. It's the cost of the propane to keep re-heating the water as it cools in the pipes.

So, I found a product that may work, and I would like to ask if anyone can see drawbacks. Basically, you get a flow sensor switch and a control unit. The flow switch is placed at the cold water inlet to the hw tank, and my existing pump is plugged into the control unit. The control unit is then plugged into the wall. Wires run from the flow sensor switch to the control unit, and whenever it senses a flow of water into the hw tank, it turns on the pump. When the flow stops, the pump turns off. I think this would be ideal for my situation. Currently, if I turn on the hw in the distant, upstairs bath, and then walk over and flip the switch to turn on the pump, I get hot water within a few seconds, almost instantly. I understand that this slightly defeats the reason some people install these circulating pumps. That is, some people install them for the convenience of instant hot water. And having to wait even a few seconds goes against the whole idea of having the pump. However, in my case, without the pump, the wait for hot water would be a few minutes, so this seems like a good compromise.

Am I missing a fatal flaw in the set up?


Thanks in advance,
Colin

p.s. The return pipe is not insulated, which I plan to correct in any case.

Recirculation by definition "wastes" some energy. A small amount in the cost to run the pump, and a little more in the heat loss. Minimize that with lots of insulation. The system SAVES all the water that would otherwise go down the drain while you are waiting for it to get hot. Depending on where you live, the cost of water is starting to rival the cost of the energy. Then of course you factor in the convenience factor. Using a flow switch will not get you hot water much faster than just letting the water run!

Thanks for the reply. Much appreciated, since I'm really having trouble sorting this out. I will definitely insulate as well as I can, but I think this thing is currently costing me quite a bit in propane. It's hard to determine exactly, because my furnace and cook top also run off the propane, but we still use a lot of propane in the summer, when the furnace isn't even on. We used a 500 gallon tank of propane in three months, with no furnace, only cooktop and heating water. With the furnace running in winter, we would typically go through two tanks in the same time frame. I don't think cooking uses much, so I guess it's the water heater. I'm making a fairly wild estimate, but I think it might be costing around $400-500 a year to run. That seems excessive. That said, even though I don't pay for water (I'm on a well), I can understand the advantage of not pumping thousands of gallons of water into my septic system. So, I definitely understand the reason for having it.

I just tried an experiment where I waited until I had cool water in the lines, after the pump had been off for a couple hours. I turned on the kitchen faucet and immediately started the pump. You're absolutely right. It took about two minutes for the water to heat up. Better than the 8 minutes without the pump, but still way too much waste of water to run that long. I will investigate the traditional timers combined with over-ride buttons and/or motion sensors.

Thanks for getting me back on the right path.

Colin

It doesn't sound like you know how much propane is being 'wasted' by the recirculation. What you know so far is that you are using more propane that you would like for hot water and cooktop (summer usage). It is possible that the efficiency of your hot water heater and the total amount of hot water used may be much more of a factor than the recirculation line.

I don't understand the 8 minutes you are reporting. Are you saying that it takes 8 minutes for the hot water to travel from the hot water heater to the kitchen faucet? This seems far too long for any typical system with a hot water tank.

Yes, it takes a long time. The time seems to vary. This morning, I tried turning on the hot when the pipes were cold, without the pump running, and it took about five minutes. I confess my tests have been pretty casual up until now. I suppose I need to test in a variety of conditions, but I just haven't had the time. I was hoping to learn a bit before I start taking wild stabs at this. So, this discussion is helping. The reason I began with the idea that the circulation pump is wasteful is that the propane guy told me he had "gone around" on this issue for years with the previous owner, who, presumably, was complaining about high propane usage. Propane guy is convinced circ. pump is "culprit." I began googling, and turned up all sorts of sites that talk about how (energy) wasteful circulation pumps can be. I also noted how the previous owner had disconnected the pump from the timer and hooked it up to a switch, which seemed like an unreasonably inconvenient thing to do if there weren't some financial incentive. On the other hand, the previous owner didn't insulate the pipe, which seems like an obvious place to start. I just now went to the crawl space, and felt pipes. There is a 3/4 inch return pipe which is 200 feet in the crawl space alone, all un-insulated. I'm not sure how much insulating that pipe would change the equation. But, until coming to this forum, I guess it hadn't occurred to me to question my basic premise, that the circ pump is wasteful of propane. The water heater is 14 years old. Am I chasing around the minor issue while not seeing a major one?

p.s. Also, regarding it taking a long time to get hot water without the pump turned on, I assumed that this was the reason for the pump in the first place. I mean, if I got hot water in a reasonable amount of time, I wouldn't need a circ. pump. Am I missing something there?

p.p.s. I'm not sure if I was clear, but I also understand that you're saying maybe there isn't a "culprit" at all, since maybe there is no crime. In other words, I can insulate, and I could add a timer to add some efficiency, perhaps even upgrade the hw tank, but maybe this is just how much propane it takes to heat the water we use in my house?



Another edit:

I just did a stopwatch/thermometer test. I started out with pipes that were room temp, after not running hot water for about 2 hours. After 8 minutes running the hot in the kitchen sink, without the pump, the water was still only 105 degrees (it took 6 minutes to get to this temp). After 9 minutes, I gave up and went ahead and flipped the switch to turn on the pump. The temp of the water coming out of the tap instantly jumped to 125 degrees (the setting on the wh). I only ran the pump long enough to watch the temp rise and stabilize, maybe 20 seconds. When I flipped the pump off, the temp quickly dropped back down to 105 degrees. I then went down to the crawl space and felt pipes. The 3/4 inch pipe that previously felt warm, when the circ pump was running was not warm. I think this is the return. I went to the hw tank. The 3/4 inch return pipe leading to the pump and then into the t attached to the outlet at the bottom of the tank was not warm. On top of the tank, the hot water outlet and the COLD water INLET were both very warm. I have no idea what this means. Obviously, I need a qualified plumber, but it would be nice to hear some guesses about what is going on, and advice about further testing, so I can describe somewhat intelligently to the plumber what I have observed.

200 feet is a long way. How big is this house? How cold is your crawl space right now?

If the hot water at the kitchen sink is only 105 after 8 minutes, that may indicate that a lot of heat is being lost on the way to the sink. The pump increases the temperature because water is running through the supply pipe much more quickly, and doesn't have as much time to cool off. You would probably see a big improvement by insulating both the *supply* and return pipes.

If the distance between the hot water heater and the point of use is so great, you might also consider adding a second hot water heater closer to where you need the hot water.

I hope someone else chimes in here with suggestions....

The crawlspace is currently 55 degrees, which is probably about what it was yesterday when I did those tests. I arrived at 200 feet not by actually measuring the pipe but by noticing that the pipe made a complete, horizontal circuit of the crawlspace, and the house is about 25 feet wide and 50 feet long...which I just realized is actually 150, not 200. In fact, the pipes are in a ways from the foundation wall, so total is probably more like 125 feet. This is the main pipe I can see, though. Obviously, there is more pipe running vertically to connect to the fixtures (kitchen, bath and laundry on first floor, 2 more baths on second floor). I don't know much about science, but while I can certainly see that the water would cool over a long run of uninsulated copper pipe, it seems unlikely that the water would cool off so quickly. That is, my thermometer dropped within seconds of turning off the pump.

So, I am slowly beginning to realize, after lots of reading, that our circulation system isn't functioning like others. From what I can gather, the way these generally work is that the pump will cycle on, move the water in the pipe for a couple minutes, and shut off. If you do this every 30 minutes or so, that is sufficient. You can then put the system on a timer, so you don't bother to run this cycle in the middle of the night. However, OUR circulation system does not provide hot water at all, unless the pump is actually running. As soon as you shut off the pump, no more hot water. This indicates a problem, right?

As I said above, it seems unlikely that the water could release its heat so quickly, simply by giving it off to the copper pipe, which releases it to the crawlspace. It seems logical that the only way you would get cooling so quickly is to mix in cold water. I did more reading last night, and found out about the phenomenon of cross-over. To test for this, I shut off the cold water intake to the water heater, and opened the hot taps. I got no leakage, which indicates no crossover. But I'm still not sure if this rules out cold water mixing. Again, I'm guessing here, but it seems like running the pump might create a path of least resistance that vanishes when the pump stops running. In other words, to form this as a question, is there a way cold water could be entering the system BESIDES through a broken faucet?

...or maybe it's the other way around. Is there a way hot water could be pouring off into the cold? Again, this is a closed system, with a dedicated return. I have never noticed any problem with getting warm water out of the cold tap. The only other symptom I've noticed is occasional pipe banging when filling the upstairs bath.

As a side note, I'm definitely going to end up hiring a plumber, but I feel like I have to have some sense of what I want before I do that. If I bring someone in before I understand the basics, I feel like I wouldn't be making an informed choice about the things I would be offered. Been there. Done that.

I also have a long wait for hot water. Just a few thoughts:

What size are the pipes? In my current house, the previous owner ran 3/4 copper all the way to the sinks. Long runs of 3/4 hold a lot of water. In my previous house, the main line was 3/4 but all the sub feeds to the sinks were 1/2 which made a big difference.

They do make circulating pumps ( such as Laing LHB08100093) that will only turn on when the temperature drops to a certain point (like 80). They can mount under a sink or where your current one is. Some also combine this feature with a timer. Might be a good option for you.

Thanks for the suggestion. I will consider it once I determine whether something is actually broken with current system, or whether it just needs something more. Since I don't ever get hot without the pump, I'm thinking broken at this point. It is 3/4 pipe, with 1/2 inch sub feeds. I found some web pages with calculations about how much water this represents, and, while it is a fair amount, I think I should still be getting hot eventually. That's what I'm trying to determine at this point: whether there is something fundamentally wrong with the design, or there is some missing or broken element. Problem is, I don't know what I don't know.

Do you have a dedicated return line, or are you using the cold line for a return? You also need a functioning checkvalve, otherwise, you might be mixing the top (hottest) with the bottom (coldest) water in the tank. Did this ever work right? It might not be installed properly.

Dedicated return. This was the setup when we bought the house, and, never having had a house with a circulating pump, I didn't know what to expect. We have always just turned on the pump whenever we wanted hot water, and turned it off when we were done using hot. I've been assuming that we needed to do something, like add a timer, and it has seemed like we have used an unusual amount of propane, but I haven't had the time until now to trouble-shoot. Until now, I wasn't sure that it isn't normal to never get hot without the pump turned on, though I suspected that this wasn't normal. Now I'm almost 100% convinced that the plumber coming tomorrow will tell me what you just told me, that it needs a functioning check valve. I don't really understand where the check valve goes, or how cold water gets into the hot, but it's the only logical explanation common sense provides. Presumably, you are talking about a one way valve at either the place where the dedicated return attaches to the water tank, or a one way valve where the hot water exits at the top of the tank? Or perhaps this is a problem inside the tank?

Edit:
I just stumbled across a diagram of installing a hot water circulation system in a two story house with a single hot water tank, and I noticed that the diagram includes balancing valves in the pipes. Is it possible a malfunctioning balancing valve would cause the problem I am describing (no hot without pump running)? Would my previous (negative) test for crossover also rule out a malfunctioning balance valve in the pipes?

Replace the 3 way switches with 2 "twist" timers in parallel. Either one, or both will turn the pump on but it will only run until the last one turns it off.

jadnashua said:

Do you have a dedicated return line, or are you using the cold line for a return? You also need a functioning checkvalve, otherwise, you might be mixing the top (hottest) with the bottom (coldest) water in the tank. Did this ever work right? It might not be installed properly.

Click to expand...

Just following up to say plumber took one look and confirmed the need for a check valve. He didn't have one on his truck, so I haven't had it installed yet. I'm still sorting out how I want to set this up (timers, etc), but I do have a question about the check valve. If I understand correctly, what is happening is that, when the pump is off, cold water is traveling backwards down the hot water return pipe. This cold water meets the hot water (coming from the top of the tank) at the faucet and mixes to create lukewarm water coming out of tap. Is that right? If that is correct, then am I also correct in thinking that this wouldn't really account for high propane use since the pump creates enough pressure to keep this from happening? Still hypothesizing here.... The problem begins when I turn OFF the pump. I notice that the return pipe fairly quickly grows colder, the cold creeping along (backwards). I'm guessing this is because the hot water wants to rise, and with the pump off, the path of least resistance is backwards through the return, which delivers cold water to the top of the tank, through the hot outlet. So, with the pump off, is the water gradually circling backwards, cooling the hot water pipes? If this is correct, it means that turning off the pump actually increases the propane use, while running it 24/7 would decrease the propane use. I'm really curious to know if that logic makes sense. Of course, the answer is to install a check valve and to insulate the pipes, but I wonder if all the effort to save propane by turning the pump on and off has actually been the opposite of what we should have been doing.

Depending on exactly how your pipes are arranged, water could be circulating in the loop all on its own without the pump. This will draw heat out of the tank especially if the pipes are not insulated. It needs a checkvalve and it's prudent to insulate as much of the hot piping as you can to retain that heat you've paid for. On my system, I run it on a timer so that for about 8-hours at night, it doesn't run. If for some reason I really do need hot during that timeframe, it just takes longer to arrive and I live with it as it's too much trouble to go to the control, put it in manual on mode, then remember to turn it back off to auto again. Doesn't happen often enough to worry about. Somewhat depends on how regular your use patterns are.

I think that explains everything. Without a check valve, my uninsulated pipes have been functioning as a giant radiator all this time. I'll correct this immediately. Thanks for the help.