Shower water saver

An attempt to recycle the water wasted during the warm-up period at the start of a shower.

Similar projects worth following
Some houses are built in such a way that the pipes carrying the water from the water heater to the shower head are quite long. When you first turn on the shower, the cold water that has been sitting in these pipes has to be replaced with hot water before you can feel any warmth from the running water. In my house, I have to wait almost a minute before the water warms up. This is a minute of running water down the drain that is not being used.

My idea is to capture this water in a holding tank, which in turn fills up the tank used for flushing the toilet.

Not only would this system save water, but it would shorten the time you have to wait for the water to heat up. One reason I have to wait so long for the water to heat up is that I use a water saving shower head. The "water saving" feature is simply a flow restriction. The reduced flow makes it take even longer for the water to heat up. By bypassing the shower head, I should be able to minimize the wait time.

I can think of doing this project several different ways.

Mechanical method: The reservoir that would fill the toilet tank would have a float system just like the toilet tank system. If the reservoir was empty the float would be down and the valve would be open. When the shower is turned on, the open valve would cause the water to go to the reservoir instead of the shower head. When the float shuts off the valve, the water would return to the shower head. The system could be tuned by adjusting the float height, so that only the cold water would be bypassed to the reservoir. The mechanical method would be very robust, but could only handle a 1:1 shower to flush ratio.

Electrical method: By using electronics and sensors, the reservoir could be larger, potentially saving more water. The water could bypass the shower head until a specific temperature is reached. The electrical method would also allow for other ideas to be incorporated (like volumetric feedback etc).

Either method would be an add-on to existing systems, not a replacement. If no saved water was available, the toilet tank would refill in the normal manner.

  • Environmental Considerations

    MechaTweak04/01/2015 at 03:00 0 comments

    Recently, Ryan H commented on this project. He asked why not simply use a hot water recirculation system. As I had never heard of such a thing, I started doing a little research. These systems are basically comprised of a pump (installed at the water heater) and a loop/check valve installed between the hot and cold supply lines at the furthest faucet. Multiple loops and valves are needed for each branch in the plumbing. The pump forces the water through the hot water supply line and back into the heater tank through the cold water side. The user gets instant hot water. No wait time, no water waste. So at first glance, it appears that this system does the things I was trying to do, and probably better.

    I was seriously considering stopping the project, but as I thought about recirc systems, a thought occured to me. I takes a lot of energy to heat water. These systems have to reheat the water everytime they cycle water back into the tank. After some more research, it turns out that the better recirc systems have built in timers. This way, the user can set the times that the pump operates. This helps a lot, but can we do it better? What has a larger environmental impact: reheating the water, or just pouring it down the drain? How do you compare the impact of water vs power?

    I found this really good resource about the carbon footprint of water:

    It's worth the read. One interesting piece of information was that the average power to treat and pump water to a residence ranges from 1050-36200 Kwh/MG. With common values ranging from 1250-6500 Kwh/MG. In my area, we use ground water, and since I have a septic system, my water energy cost (not including heating) is probably around 2000 Kwh/MG. The average power used for heating water (for a shower specifically) is 148,832 Kwh/MG. That's 74.4 times more power to heat the water than to treat it and pump it to my house. Another staggering number is that 58% of all water related carbon emmisions come from residential water heating.

    I'm sure someone could make the case that if you set the timers perfectly on a recirc system, and showered at exactly the same time everyday, you would not waste very much energy. My reality is that I have four young kids. The messes they make are rarely on any kind of schedule, and impromtu showers/baths are a common occurance. If I used a recirc system, I would probably want to have it on at least 12 hours a day. That is a lot of wasted power. Maybe someone could figure out a way to only turn on the recirc pump whenever hot water is needed. I can't figure out a way that would be cheap or practical enough (maybe an app where you could manually activate the pump a few minutes before using hot water?)

    Taking all this into consideration, I think my idea still has merit. I think I can save water and tip the carbon scales in the right direction.

  • Valve Testing

    MechaTweak03/26/2015 at 01:19 0 comments

    My initial pseudo-schematic showed a 3-way valve to divert the water to the holding tank. After a little research, it became obvious that this would be an expensive piece. I thought I would test another method to see if I could save some money. The thought was to simply use a 2-way valve on the back side of a tee...

    My concern was that water may come out of both the shower head and the open valve. I assumed that the distance "d" would have an effect on how much water came out of the shower head. So I ran a little test.

    First I removed the shower head, and adapted the 1/2" npt threads to some 1/2" flexible tubing.

    The tubing ran to a tee fitted with a valve....

    then back to the shower head (yes that is electrical wire holding the head on).

    When the valve was closed, the shower operated normally.

    When the valve was open, most of the water came out of the valve.

    As I varied distance "d", more or less water would come out of the shower head. I had to put the valve 3 feet below the head to stop the flow altogether. This means that my reservoir should be as low as possible (but still above the toilet tank) to save as much water as possible. This testing also varified that a 1/2"valve should work fine.

    The next step will be to get some solenoid opperated 2-way valves and start automating the process.

  • Tank Level Sensing

    MechaTweak03/23/2015 at 04:00 0 comments

    Part of the problem with getting people to care about water saving is that water is so cheap. At less than a penny/gallon you really have to save a lot of water to make any device seem worth while. Of course the ideal scenario would be to have a system that would eventually pay for itself. If this could be achieved, consumers wouldn't have to care about conservation. They would want the system simply because it would save them money. I know that this build will never pay for itself, but the hope to prove the concept. Maybe in a mass production environment, a version of this device could be produced for cheap enough to make economic sense for the average consumer.

    With this in mind, I will be attempting to make this build as cheaply as possible. One of the first parts I started researching was float switches/liquid level sensors. I think that both the holding tank, and the toilet tank will have to have a level sensor of some kind. These will help prevent overflows and let the system know when to refill the toilet. The low end float switches were around $10 (automation direct) I would need two. The low end level sensors were $40 (adafruit). After doing a little digging, I found that some people had been using capacitive touch sensors to read liquid levels. I had an arduino and an adafruit MPR121 on hand so I ran the following test:

    I wrapped a plugged PVC pipe with four 1" strips of tin foil. Each strip was taped on, and roughly placed at quarter, half, three quarter, and full positions. I then soldered wires onto each strip and connected them to the MPR121 inputs.

    I then wrote some code to test the sensor. At first I used the example code from Adafruit, this worked fine for my hand touching the strips, but did not sense water in the column. Even after I set the MPR121 for maximum sensitivity, it didn't register water as "touches". I then set the MPR121 to output the raw data, and noticed that a change was noticeable. All I had to do at this point was record the empty values and full values and write the code to look for these. It worked like a charm. Right now the plan is to use a 4" (maybe 6") pvc pipe as the holding tank. This would have foil sensors wrapped around it for level checks. For the toilet level check, I am going to try to do the same thing, but inside out. A dry-on-the-inside tube with the foil on the inside lowered into the water. Thats all for now....

  • Initial testing

    MechaTweak03/20/2015 at 01:20 0 comments

    I just did some testing to establish a baseline for this project. The results are as follows:

    Time for running shower to reach comfortable (not full) temperature: 45 seconds

    Volume of water used (wasted): 19 cups (1.2 gallons).

    Time for shower with head removed to flow 1.2 gallons: 22 seconds.

    These results are encouraging. It seems this idea should be able to save more than a gallon per shower and half the time I have to stand naked and shivering with my hand under the cold water. My calculated flow rate is 3.24 gal/min which is reduced to 1.6 gals/min with the shower head.

    I would love to hear from any readers what your shower warm up time and volumes are! Testing should be done several hours after the last shower use (first thing in the morning would be great). Run water into a bucket and time how long it takes to reach a temp where you would step into the shower. Then measure the volume. It would also be interesting to hear about your various setups (shower only, tub shower combo, independant or single/mixing valve, etc). My test was done on a shower stall with a mixing valve.

View all 4 project logs

Enjoy this project?



Alex Homebrew wrote 05/12/2020 at 05:31 point

Years ago I purchased and installed a device that pumps water from the hot line to the cold line.  It was kind of expensive but I didn't have time to build my own.  It has a temperature sensor that stops the pump when the water get hot enough, though it almost always cuts off early. It worked for several years then the valve that prevented cold water back filling failed so I installed a manual shutoff. 

IMHO, the architects in the US do a poor job of designing plumbing. In Europe, each place that needs hot water has an instant hot water heater rather than a big tank in the garage and long pipes that are usually not insulated. I believe the recirculating pump is a solution that benefits the plumber, not the homeowner. More pipe to install, more devices to repair, and more energy wasted.  

When I lived in California during a drought, I had a bucket to capture the water, which I then watered the garden with.  Your plumbing would make that easier, in fact you could simply add a drain to the outside with a faucet to water your plants.  The tricky part of filling the toilet tank would be to fill it first with the stored water, then with line water if that's not enough.  Maybe your storage tank could have a float like the toilet tank?

  Are you sure? yes | no

kristine wrote 03/17/2020 at 07:08 point

For me, a hot water recirculation pump is a water saver and it gives instant hot water! I stumbled on this WaterQuick Pro here

  Are you sure? yes | no

Henti wrote 10/20/2019 at 14:39 point

Wow very interesting thank you.

How to remove Kohler shower handle

  Are you sure? yes | no

Frank Vigilante wrote 04/01/2015 at 03:56 point

About a year ago, I was looking for a way to quickly heat water for an outdoor gardening project. I stumbled on this awesome company, which had just crushed it on KickStarter.

  Are you sure? yes | no

MechaTweak wrote 04/01/2015 at 22:33 point

looks like a good system.  Thanks for the share.

  Are you sure? yes | no

Ryan H. wrote 03/26/2015 at 03:24 point

I cannot help but wonder why you wouldn't just use a hot water recirculation pump. They solve the hot/cold water issue extremely well, and do not waste a drop of water.

  Are you sure? yes | no

MechaTweak wrote 03/26/2015 at 15:27 point

I was unaware that recirculation pumps existed.  They seem like they solve the problem quite elegantly.  I have talked to several people around here and no one has heard of them, so they must not be too popular in our area.  Even though you kind of burst my bubble, I appreciate the constructive criticism.  It's always hard to hear that an idea is not as original as you initially thought.  After a little research, here are a few comments.  Recirculation pumps do use energy, not only to drive the pump, but to constantly be reheating the water.  Water takes a lot of energy to heat up.  My system would probably be more energy efficient (and a completely mechanical version could potentially use no energy).  The better recirculation systems look like they run on timers to bring energy usage down, but these systems are more expensive to buy.  Also, since my idea involves supplementing toilet flushes, it would be quite easy to incorporate rain water sequestration and potentially other water saving ideas.  I'm going to let this project continue to evolve, and see where it takes me.  Thanks again for the great comment.            

  Are you sure? yes | no

Mike Szczys wrote 03/30/2015 at 20:53 point

I have heard of recirculations pumps. They installed one on the TV show "Ask This Old House". Pretty neat device. However, like you I have never seen one in use and I don't know anyone who has one.

Whether or not your concept would be more efficient or better than a recirc. pump, I think the ideas is worth exploring and could lead to other interesting concepts. Nice!

  Are you sure? yes | no

MechaTweak wrote 03/31/2015 at 12:09 point

Thanks for the encouragement!  I think my next step is to do some research on energy usage of recirc. systems.  It seems like a lot of water conservation techniques use electrical power to save water.  I wonder which is truly better for the environment?  Both water and electrical infrastructure damage the environment in different ways, so an apples-to-apples comparison will be tricky.

  Are you sure? yes | no

Martin wrote 04/21/2017 at 07:40 point

You could activate the recirculation pump only on demand: Take the  time it needs to bring the water to sufficient temperature at the shower head (e.g. 1 min.) and install a one shot timer which you start like one minute before entering the shower.

I don't like this "water saver" shower heads anyway. I had to experience them even in some hotels. If I want less flow I can turn down the tap , nobody forces me to open it fully. And if they restrict the flow too much it just takes longer to rinse off shampoo or soap, so near zero net savings.

  Are you sure? yes | no

bouni wrote 03/25/2015 at 07:18 point

This idea could be really interesting if you plan to build a new house!
The plumber could install a third pipe back to the basement where a large tank is located that holds all the returned cold water, which could be used to supply the toilet tank. And in case not enough water is returned from the shower it could be filled from the normal water system.

  Are you sure? yes | no

MechaTweak wrote 03/25/2015 at 11:39 point

I agree, a new house installation would be ideal.  My prototype will probably be quite ugly (and annoy my wife).  

I thought about a whole house system.  I think it would be better to put the main tank in the attic so it could gravity feed down to the toilets.  This way you could eliminate the cost of a pump.  For now I only plan to pair one toilet and one shower together.  

  Are you sure? yes | no

Blecky wrote 03/26/2015 at 06:16 point

Just do your business in the shower beforehand and wash it away :P

No need for a toilet in a new house then!

  Are you sure? yes | no

ErikH wrote 03/23/2015 at 08:49 point

Hi! If you have a hot water boiler, you can ignore this post ;) Otherwise: with older gas powered water heaters the first ~20-30 seconds are used to start the system and not a lot of heat is generated. In that case, you could limit the flow to the minimum needed to keep the heater burning, and after 20 seconds increase the flowrate to get the (now hot) water to your shower. Cheers! Erik.

  Are you sure? yes | no

MechaTweak wrote 03/23/2015 at 13:20 point

I do have a boiler, so sadly your idea wouldn't work for me.  Thanks for the comment though.  I don't know anything about the kind of system you are talking about.  Is it a whole house system, or just used for one faucet/shower?  Is it similar to modern on-demand heaters?

  Are you sure? yes | no

ErikH wrote 03/23/2015 at 14:32 point

In The Netherlands gas powered combined heating systems are the norm, as we have quite a large supply of gas available and almost all houses are connected to the gas network. These heating systems heat the closed circuit central heating system and deliver on-demand heat for hot tap water for the whole house. Older ones tend to need quite some time to start though. Modern ones are quicker, or have a small (5-50liter) integrated boiler.

  Are you sure? yes | no

MechaTweak wrote 03/23/2015 at 16:22 point

I'm sure it depends on the plumbing of the house, but it sounds like these older systems you are referring to waste more water on start up than the boiler system we generally use in the US.  Probably more energy efficient though...  

  Are you sure? yes | no

ErikH wrote 03/24/2015 at 15:42 point

On demand heating will always waste more water I guess, as it takes some time to start. However, water is a lot cheaper than (gas/electric) energy. I can buy 250 liters of water for each kWh, and even a small 10L boiler has ~40kWh of losses each year, so for a on demand system I can put 10m^3 (2642 gallons) down the drain... But it would make your problem even bigger ;)

I like your idea. For me it's more practical to compensate water losses by using rainwater storage for toilet flushing. But that's a project for later this year..

  Are you sure? yes | no

MechaTweak wrote 03/24/2015 at 16:39 point

I thought about integrating rain water storage into this project.  I am already using nearly all the components to make it work.  My only concern is that rain water is not treated.  Would it go green if left too long?  Would it wash impurities off the roof that would mess up anything (like the sealing seat on the flapper valve).  I guess as long as it was run through a fine screen and used quickly everything would work out OK.  

  Are you sure? yes | no

ErikH wrote 03/25/2015 at 11:41 point

When exposed to light, the water will turn green, but if kept dark and not too hot, it should be okay. Some filtering is needed though. Think about wind and birds dropping stuff on your roof ;)

  Are you sure? yes | no

MechaTweak wrote 03/25/2015 at 13:31 point

I think the tank would have to be in the attic (so that it could be gravity fed down).  Here in North Carolina, attics can get pretty hot in the summer.  I wonder if mixing the rain water with treated tap water would give any protection.  

  Are you sure? yes | no

ErikH wrote 03/25/2015 at 13:58 point

Would depend on chlorine content I guess. Here in The Netherlands water quality is equal or better than bottled water, so no chlorine needed. In the US it's normal to have chlorine in it I read somewhere? That would keep algae out of your tank I guess ;)

  Are you sure? yes | no

MechaTweak wrote 03/25/2015 at 17:50 point

Must be nice.  We do have chlorine, and on some days the water coming out of the tap tastes like pool water.  I have a water filtering jug that helps get the chlorine taste out.  

  Are you sure? yes | no

David Shamblin wrote 03/20/2015 at 03:24 point

My wife and I have been doing this manually with a cup in order to water the plants. But to use it for the toilet is a great idea. 

  Are you sure? yes | no

MechaTweak wrote 03/20/2015 at 14:01 point

Thanks.  Is one cup enough to save all the cold water?  The shower in my master bathroom uses more than a gallon!

  Are you sure? yes | no

David Shamblin wrote 03/20/2015 at 18:42 point

In our house, the shower is really close to the water heater, so it doesn't take long. It's a large cup, 32 ounce I believe, and it doesn't catch all the water, but certainly enough to water some house plants. I've been thinking about your idea though and ways I would go about it. If you knew the flow rate, you could have it as an automatic timed thing, where it would dump in a toilet tank and then divert to the shower when it was a) hot (timer expires) or b) full (your float system idea). You could even have a tank between the toilet and shower that would fill if the toilet one was full (could be used for plants, fish bowls, etc). 

  Are you sure? yes | no

MechaTweak wrote 03/20/2015 at 19:23 point

I was already thinking of using a tank between the toilet and the shower (see sketch).  I was planning on having the tank automatically fill the toilet after a flush if water was available.  I like the idea of having a faucet on the tank for other uses though.  My goal is to have a fully automatic system that would not require any action from the user.  

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates