Harvested rainwater for non potable use in the home
 

I am a technical writer/engineer at a small company that does contract engineering.
i.e.; people come to us with an idea and ask us to make it work.

Technical Services

Our field of expertise is electronic engine controls. We work a lot with MEFI controllers. But that's off topic.

Just the other day the boss dropped a new project in my lap. The concept is to use harvested rainwater to flush toilets.
Have any of you guys seen any technologies that interface a rainwater system with a city water (or well water) supply?
I've been internet surfing like mad, and have found some sites like harvesth2o.com and others that are linked from threads here on ecorenovator, but have not found much yet.

I just don't want to be reinventing the wheel. :p

It's definitely a good idea; grey water systems turn black and are gross to see sitting in the toilet; rainwater seems to be an excellent choice.
Hope to hear someone's used it already. good luck.

Sorry, I don't know anything about it (yet). If you come up with anything you can share, please do!

I haven't seen anything about this type of setup specifically... but...

When I grew up I lived on a farm... it was once a very old farm and there was a wind mill that actually did the water pumping. We had a big cistern that held the water that was pumped so you had it on windless days... kind of like a battery bank I suppose. Something like that may be a good place to start. I would put a second water holding tank in the attic (provided it wouldn't freeze) to allow gravity feeding of water to the toilets. Use a small pump to get water from the cistern to the attic, let gravity do the rest.

Where I live now, one section of our roof drains at one point (we don't have gutters) and we can fill one garbage can with water if there is a good rain. I use it to water the plants when it's dry.

Yup, seems that "Everything old is new again".

I've been surprised how widespread 'rainwater harvesting' is around the globe. It's even a required design element for new construction in Bermuda and the Virgin Islands.

It's such a simple concept. Now I want to install a system at my house... :p

Thanks for the input, wyatt. The attic tank is a good idea...

Rainwater is naturally soft, but corrosive to metal plumbing. PVC and PEX are recommended.

What am I gonna make the valving out of? $tainless? Injection-molded plastic? Hmmm...

Not to throw your thread out of whack....but what's the difference between PVC and ABS plumbing? I'm doing the tomato plant thing in the other thread...and it's asking for PVC pipe to fill the tub with water, I guess because it can take the rain water. I have Schedule 40 ABS pipe...that's not the same thing is it?

Higgy,
I am guessing he used PVC because he could find it easily (one of his criteria). Rain water isn't going to hurt plastic (hopefully). Not sure on the differences.
Tango Charlie,
What kind of valving are you in need of? I would think you could find PVC plumbing valves that would do the job quite nicely. I have seen PVC ball valves for simple "ON/OFF" operation. Do you need something more sophisticated? A major advantage of using what exists is that you don't have to make it yourself.

Well, the goal is to meter in rainwater with the city water going to the toilet, in varying proportions. During a drought, more city water will need to be used, to prevent exhausting the cistern, and to keep things flushing smoothly. :)

That's what I was looking to see if anything was already out there, commercially or otherwise. Haven't found anything, so it looks like we'll be doing this from a clean sheet of paper, as they say. I'm thinking it'll take some sort of actuated valve that will interface the two systems. And it must be automatic, so the homeowner doesn't have to fiddle with it. Initially, it'll probably be used in a 'smart house' that's all wired up into a computer. But we'd also like to make it a stand alone system that can be retrofitted into any house.

I'm learning A LOT on this project. Let me research ABS vs. PVC. I don't think rainwater affects either. It's the metal piping that's the problem, because of corrosion.

Thanks for the feedback, guys!

I would think you would want to keep your cistern cycling so the water doesn't have time to stand and start growing things, so what I would suggest (with nothing to back me up) is that the cistern gets used for flushing the toilets 100% until it's depleted to maybe 10% of capacity. At this point you can have a float in the cistern (an existing technology) that will be tripped and keep the cistern from going completely dry, but also keep it from going completely stagnant. This would make it a stand alone system and make it so there can be computer monitoring, but the computer is out of the loop for all intents and purposes.

What you've described, wyatt, is an auto-fill device. What I don't like about it, is that now you're introducing the nasty hard city/ground water into your wonderfully soft rainwater system.

A big issue I've run across is back pressure and back siphoning from a "cross connected" system (I'm learning the lingo!)
If the city water supply loses pressure, there can be no chance of the non-potable water from being forced by pressure, or sucked in by siphon action, into the potable water supply. I think a toilet already does this, as the valve in the tank dispenses above the water level in the tank. But we need to mix the two sources before it gets to the toilet fill valve...

...The research continues.

hmmm... good thing you are telling me why these things are a bad idea... let's try this one...

http://i378.photobucket.com/albums/o...rntoToilet.jpg

ok, so we have a cistern, a tank in the attic, and a house. Nothing is to scale, PLEASE do calculations if you are going to do something like this, there are probably easier ways to accomplish the same thing, I am not qualified to design these sorts of systems, etc. etc. etc.

Latest idea:
IF cistern is over 50% full, pump 1 operates every time valve 1 is tripped, topping off the tank
IF the cistern is less than 50% full, pump 1 no longer operates, pump 2 AND city water operate every time valve 2 - city valve are tripped, mixing 50/50 and filling the tank (More likely there would be one valve tripped that started both)
IF the cistern is dry (or below a certain level), pump 1 and pump 2 no longer operate, city water operates every time city valve (or common valve) is tripped, filling the tank
IF you get tired of my ideas, just let me know.

Have a good Earth Day!

Wyatt

Why would I get tired of your ideas? That's why I posted here! :D :thumbup:

We could eliminate pump 1 and use a simple logic program to mix the two sources into the attic tank. Would need a level transducer in the cistern, then.

The attic tank would have 8 or 9 feet of head, giving us 2.5 to 2.8 PSI to the toilet. Wonder if that would be enough to operate the toilet fill valve...

I think so, since the toilet starts to fill the instant you flush and doesn't stop until the toilet is full. That binary ON/OFF type thing isn't operated by the pressure of the water if I understand correctly, it's operated by water level in the tank. If I have a few extra pieces laying around I will test out the gravity feed portion, but I will be testing it on the counter next to the toilet first, because if it works there, it will for sure work in the attic. It's possible you will have a thumbs up or thumbs down tomorrow on the gravity feed, but it seems a simple siphon action would fill the tank.

I proved two things, and disproved one!

Experiment:
Attach a siphon hose to the tank inlet, see if gravity feed will fill the tank.
Hypothesis:
Yes, it will.
Method:
No siphon hose available... no hose available... made a hose out of plastic, duct taping the seam, duct tape the "hose" to an old tank fitting. Fill "hose" with water until the tank starts to fill.
Outcome:
Tank didn't fill, even when water was one foot over the rim.
Dis-proved hypothesis!
Proved that "hose" didn't hold water well (leaked like crazy)
Proved that the tank required pressure to activate. How much pressure is unclear.

I was hoping I could get video of it working... no such luck. The tank actuator (at least mine) is pressure activated. A better experiment with better quality components would be needed to determine if a gravity feed system would be able to fill a toilet tank. Sorry, I guess I have to leave that one up to you.

Ha ha! you kill me, wyatt! :p You've officially done more hands-on work than I have with this project! I'm still stuck in my cubicle researching plumbing codes, since we don't know what we don't know at this point.
Our resident construction guy has constructed a mock-up for testing, though;

http://i337.photobucket.com/albums/n...ie/mockup1.jpg

I think he's going to set the toilet today and plumb up the drain.

Today I've got to work up a Preliminary Report for the client, who is coming for a visit next week.

any sweet updates? can you trip the toilet with pressure from the attic? video? I've thought more about your project than I have about most of mine!

No big updates yet.
The client liked my report and our mock-up, so work will continue.
Still researching plumbing codes and trying to wrap my head around backflow prevention valves.

I wanted to post some rough schematics that I've sketched up to show what I've been coming up with, but since this is a paying customer's project, I probably shouldn't. Non-disclosure statements and all that jazz, you know. :(

"It's all very 'hush-hush' and 'hoosh-hoosh', you know"
(Terry Thomas as Lt. Col. Hawthorne in "It's a Mad Mad Mad Mad World")

Three different ideas;
One: Collected rainwater is pumped from a cistern to a small tank in the house (not the attic, due to possibility of freezing). Then gravity fed to a valve that would mix the city and rainwater roughly 2:1 as it goes to the toilet. An RP backflow preventer would be needed on the city water supply line to prevent the possibility of rainwater contaminating it. And they ain't cheap.

Two: Rainwater pumped into a small interior tank as before, but topped off with city water when rainwater is depleted. An airgap (cheap!) at the top of the small tank would prevent backflow/backsiphon. Another pump in the small tank would provide pressure to operate the toilet. Is a second pump cheaper than an RP device?

Three: Junk the small tank altogether and pump the rainwater straight from the cistern to a solenoid-actuated valve that chooses between city water or rainwater, depending on availability of rainwater. Also requires an RP device on the city water supply line.

Hopefully this is sufficiently detailed to get the conceptual idea across, without disclosing too much to get me in trouble. :o

This week I finally got around to doing some hands-on testing. I rigged our mock up to a garden hose and recorded some tank refill times at various pressures. The garden hose gave me around 46 psi static, 40ish while filling. That took about 33 seconds to refill the tank. I got down to 10 psi, and that took about a minute and ten seconds.
http://i337.photobucket.com/albums/n.../mockup004.jpg

Are you saying that below 10 psi the valve wouldn't open? or is that just all the lower you tried?
By the way, 1 psi = 27.7 inches of water, so if 10 psi is required to open the valve, you would need about 23 feet of head to operate the system, which would mean even in the attic you would need a pressure boost for bathrooms directly below.

10 psi is all the lower I tried. It was getting late in the day so I called it quits. It would be interesting to go lower. I'll try that next week. (I'm still on four day weeks, Fridays off)

Here's the latest concept I've been pursuing. This week I'm trying to source the solenoid valves that will select city or rain water. The boss tells me that geothermal systems use electrically actuated valves like this, so I'm off to learn about that.

http://i337.photobucket.com/albums/n...ptdiagrama.jpg

I tried using a simple venturi pump to suck water out of the reservoir and mix in with the city water, but the toilet requires such a low flow, that it just didn't work. So we'll have to use a pump.

Any Updates
 

Just found this thread.

I was voting for option 3 with the solenoid valves as I was reading.

I had thought about putting a tank somewhere and just charging it with rain water instead of well or city water.

I have not calculated the cost of this project.
Replumb toilets for rain/recovered water only.
Pump, tank, cistern large enough to last through the dry months.

The up side is having a water supply if there is a Hurricane. :thumbup:

Sorry for the delay.
No, there are no updates. I was laid off in July, so no work has progressed on this project.
The downside is the cost of the solenoid valves and the backflow preventer. You don't want to go cheap with these components. If memory serves me, (which is a big IF) these three items would run you well over a grand, total.

I miss my job. :-(

I was thinking of this thread the other day when I found the link below.

Rainwater Harvesting and Purification System


Sorry to hear about your job Terry. :(

Thanks, Daox. That's an interesting link.
The problem with roof washers is that they only work as designed in areas that get downpours in between periods of no rain. As the fella experienced, being in drizzly Oregon, they just require a heck of a lot of maintenance to keep them functioning. Even in moderate climates like here in the Midwest, they just don't work so well. It's better to just collect and filter it all.
There's a company out there called WISY that makes a filter that is an integral part of the downspout that filters all the water coming through, so it wastes much less rainwater. It's basically an inverted cone shaped screen that the water would stick to and get diverted to the outer wall of the downspout, and the debris would fall out the bottom. Viola, less maintenance. But again, pricey stuff, which is a concern when you're trying to engineer a commercial solution. We DIY'ers can cobble one together for a fraction of the cost for our own residence, though. :thumbup:

I would go with a pump with a pressure switch on it in the rain water tank, a pressure tank after that to maintain head so the pump is not kicking on and off in 3 second bursts and a check valve to stop back flow, city water would have the same kind of check valve, these two water sources would connect to a tee that connects to your non potable water supply for the house, it would be a good idea to have a pressure regulator/reducer on the city water to keep it under say 40psi, have the pump pressure switch set at 50-60psi and your rain water storage tank has a water sensor like wet basement alarms have to turn the pump on when there is water, turn it off when there is non, otherwise a float valve would work as well.
all off the shelf parts, no risk of back flow (check valves) it would only need one storage tank and a small pressure tank, should work for 20+ years without an issue.

Ah, you'd think so, wouldn't you? ;)
Unfortunately, a simple check valve is not an approved backflow preventer. You'd get away with it in an unregulated area, but it wouldn't be the responsible thing to do.

so toss in a few more check valves, another pressure switch and a electronic controlled that turns the city water off as well.
or float valve in your rain water tank that keeps 5 gallons of water in the bottom from the city water, if it's above the 5 gallon mark the city water is off.
I just like the idea of a single tank and not having more city water then you need hanging out.

I was just reading a pretty current plumbing book (plumbing. for pros by pros) and the author talks about how each house is supposed to have a double check valve to prevent back flow of water from the house back in to the main in case of loss of pressure in the main and this double check valve is all that is needed so the cruddy water in your rusty pipes doesn't back feed.
I don't see why a double check valve like that could not be used in your house as well.

The idea behind slimline water tanks is to use as little floor space as possible without reducing their water storage capacity. That's a great idea.

I just fell upon this thread.

In Europe, there are systems for this already. I think mainly some Northern European countries offer these products.

In general, it is NOT allowed to mix city water with rain water systems, no matter how many check valves you install. The way to do it is to have a sensor in your ground storage tank. When the water level gets below a certain level (just before empty), a valve will open for a set time (say one minute) and refill the tank with an amount of water from the city water. If it goes down to the same low level again, the procedure will repeat itself, effectively keeping the tank from running dry.

As the ground tank is unpressurized, and the city water is pressurized, it is technically impossible for the water to flow the wrong way. Worst case scenario, a valve failure would mean that the storage tank would fill with city water. This is just like the cisterne of a toilet - the water cannot run from the cisterne (unpressurized) back into the city water line (pressurized).

See for instance

YouTube - RewatecUKTV's Channel

Rainwater Harvesting / Rainwater Collection / Sewage Treatment : : Rewatec (the UK website from a German company)


It is important to have the tank underground: Keeps it from freezing in the winter, and from rotting in the summer (depending on your local climate)

A Danish company (no English version) Regnvandstank (4 m³) - køb regnvandstanke til regnvandsopsamling hos Nyrrup Plast - 1 m3 (cubic meter) is almost 300 USG.

Potential plumbing code violation, I suggest talking to a plumbing inspector on what is permitted.

The easiest and probably most likely way to pass code is to use city water to refill the cistern when needed and plumb the toilet to only use cistern water. That way the two systems are never directly connected. Of course your laws may differ from here. I'm also not sure you can even do that here.

With the rain water collection tank in the ground, including a top-up system to fill city water in it when it gets toward empty, is the best way to keep "watertight firewalls" between the two systems.

There is no need for a tank in the attic. There are already pumps made just for the purpose - keeping pressure in the line on demand. They are quite simple, actually. When the pressure gets too low, the pump starts, and then it stops when it gets too high again.A small buffer tank prevents the water pressure from cycling although some pressure and flow variation can be expected if the pump is too powerful. This is mainly a concern if you try making a system from scratch, and not buying a unit made for the purpose.

As city water is still quite cheap many places, ready made units can have a hard time to complete, when you calculate how many years to pay back the system. I am going to install it in any case (eventually) at least for the reason that I think it is utterly foolish to flush toilets with drinking grade quality city water ...

City water may be cheap, but it isn't going to get any cheaper, at least here. Poland has been procrastinating with its promise to build new sewage treatment plants, so it will have to pay huge fines to the EU (or EC). The price of water (and sewage) for homeowners is expected to rise 25-fold within the next few years, while in the Warsaw area by as much as 50-fold!

The Wife and I are amoung the few who are glad about this, since hopefully it will teach people to pay more attention to conservation and will reduce the number of lawns being watered while it's raining. With our seasonal rainwater barrels, and with the planned greywater system, we won't feel the higher water bills as much as people who don't care about wasting.

How true. It's too bad that the city system doesn't have a second set of pipes with non-potable water - for flushing, car washing, garden watering, etc.

I have installed this rain water harvesting tank now:

ECO-Plus package suitable for pedestrian loading

The tank is in the ground, and connected to part of the roof area, but the rest of the installation is still in the works.

I have chosen the currently largest tank in this series, 7500 liters (2000 USG). The price in USD is around $4400 at the moment (weak dollar!). The Graf rainwater tank is a German product, so it is priced in EUR.

It might seem like a lot of money - why not DIY from bottom up? There are so many things to consider, it seemed better (and cheaper) for me to go with a well tested system. I don't want to reinvent the deep plate. I looked at more local products, but they were so expensive - I think because of low volume. In Germany, rain water harvesting seems a lot more common.

In this system, normally the tank itself has no pump built in - a tank-mounted boost pump is only required if the distance from the tank to the point of the external pump is more than 50' long / 10 feet up. As mine is just within this limit, I only need the pump which is built into the control/pump unit, inside the house (must be frost free!). It does not need to be primed in case it has run completely dry.

The water suction-feeds from the tank to the control unit, which in turn pumps it onwards to the appliances and outlets in question (for now, washing machine and toilets). Inside the control unit is a small cistern, which allow mains water to be added if I should run dry - then it tops up inside the small internal cistern automatically.

There is no physical possibility for the rain water to be pumped into the mains, as the mains water feeds the cistern (before the pump).

This satisfies the legal requirements, where as one-way valves and so on, do not. All pipes with rain water in must be clearly marked to avoid future mixup of the two. Mains meter and main shut-off valve should also be placarded "Rain water system installed in this house"

From the tank, there is a floating filter, which ensures water is taken about half a foot below the surface. There could be oil and other items lighter than water floating the top, as well as sand on the bottom, so this is where the cleanest water is.

If it does not damage the tank, I might install a UV-light inside the tank, to run whenever the pump runs. Even though rain water is quite clean, there will be pollution sources such as bird droppings on the roof. But since the tank is well underground, the water should keep a temperature around 40F or so, year round. Also, there is no sunlight at all, so any bacteria there would have minimal growth conditions. In the winter, the water is also kept free from freezing, which ... is kind of nice too!

Winter is where I think I am most likely to run out of rain water, especially in the event of the winters we had the last two years. Lots of precipitation, but no thaw for a long time - everything on the roof is frozen solid ...

I filled the tank half way during installation in the ground. This is required to avoid the tank to "float" in the gravel. I connected part of the roof at this stage, and the remaining 1000 USG has already filled, even before I have had the system connected to the house!

I will post more here, once the system is completely up and running.

http://img.archiexpo.com/images_ae/p...nel-426946.jpg

http://farm6.static.flickr.com/5142/...87ed9995_b.jpg

Good for you!!

Is there some kind of filter system between the roof and the underground tank? If not, is the cistern easy to clean periodically? What kind of roof do you have? Our roof has shingles and I keep cleaning the sand out of our barrels. Plus lots of pollen in the spring, and dust after a few dry days.

Maybe a heating element in the rain gutters? I'm not a fan of wasting electricity, but that may be worth the extra water.

http://www.warmzone.com/images/danfossrx.gif
http://www.heatersplus.com/images/roof.jpg

When I was reading this thread I was trying to come up with a way. All I could come up with was to use the city water flow as a power source to pump the rain water...sort of like a turbo charger if you know what I mean.

I do like osolemio's setup.

As far as filtration for the rain water, how about building a filter box before the water enters the cistern? One could maybe use furnace filters laid in as slats, pull out occasionally to toss the bulk junk away, and replace when the filter box won't flow enough and overflows through a top relief pipe elbow.