I've finally gotten serious about finishing my rainwater harvesting system that I briefly described two posts previously.  As I explained, provision for a below grade cistern was built into the ICF foundation of the house at minimal added cost.  I just recently poured a 4" reinforced concrete floor in the enclosure with a sump in which a 10 gpm submersible well pump will later be situated.  The floor called for 1 cu. yd. of concrete, and was accomplished for a total out-of-pocket of $240, DIY, including the rental of a 2 cu. ft. wheelbarrow type mixer.  The pictures below show the interior of the cistern.  Note how the rainwater inlet is currently connected back to the overflow line.  The capacity of the tank (3500 gallons) is limited by the height of these lines which was determined by the slope of the underground drainage lines from the downspouts.   Also, the maximum depth of the water in the tank (approx. 6') limits the pressure on the containment walls to minimize the risk of cracking the 8" thick reinforced concrete walls.  The next step is to seal the walls and floor of the tank with three coats of rubberized paint (Blue Max from Ames Research).  The rainwater collected will be used for landscape irrigation.
--Clark

Clark,

It is nice to see a project coming together.

A submersible well pump is probably overkill and not the most efficient choice. But it will work.

Clark,

Does your cistern have a separate foundation to support the walls and floor?  Is there a cold joint between the walls and floor?  Will the waterproofing be inside or outside the cistern?

Posted By jonr on 03 May 2012 05:24 PM
A submersible well pump is probably overkill and not the most efficient choice. But it will work.

What do you recommend, jonr?

Posted By Alton on 03 May 2012 05:38 PM
Clark,

Does your cistern have a separate foundation to support the walls and floor?  Is there a cold joint between the walls and floor?  Will the waterproofing be inside or outside the cistern?

Yes, the walls are 8" ICF on 16" footings.  The floor is a 4" poured slab over 4" of crushed limestone with rebar reinforcing.  The cistern walls are part of the foundation for the house beneath a covered rear porch.  Normally a builder would fill the cavity under the porch with earth before pouring the concrete porch floor.  I built a form within this space, instead, to support a 6" reinforced concrete slab with a 20" x 20" access opening (see picture)  and later ripped out the forms.

Yes, the floor intersects the ICF wall foam for a cold joint.

The waterproofing is applied on the inside of the cistern walls and floor to totally contain the rainwater.   Ames Blue Max is a flexible rubberized foundation sealant rolled or sprayed on in 3 coats each 10 mils thick.  The floor-wall joint is reinforced with a special contour tape in case the join opens up as the concrete shrinks a bit.

Clark 

I would use a non-submersible booster/sprinkler pump designed for efficient operation at lower heads. Will also be less expensive.

Okay. You're suggestion is well taken, however, would a booster pump work in my case? The pump must lift the water, albeit only 8'. I could use a jet pump, but my concern is the potential problem of keeping the pump primed. I can buy a 110VAC 1/2 HP submersible pump for $258. That's more than a typical centrifugal pump, but maybe more reliable in the end?

I could use a jet pump, but my concern is the potential problem of keeping the pump primed.

Can you place a non-submersible pump at or below the level of the water in some other (basement) location? That would diminish the possibility of losing prime.

I use a floating collection hose with a siphon. This allows the intake end to be suspended below the surface where it is less susceptible to sucking debris from either the surface or the bottom of the tank. The pump is located in the basement slightly below the level of the siphon intake which is 90 feet away in the cistern. The nice thing is that the pump stays dry and you can easily work on it or replace it.

Some of the available pumps are self priming and all can lift > 8 ft. In all cases, I would do something to prevent running dry (which often ruins the pump).

The good news is that, when I was purchasing pumps for this system, I noticed that quite a few of them now seemed to have run-dry prevention features built in.

Yes, the floor intersects the ICF wall foam for a cold joint.

Did you consider removing any or all of the interior foam before applying the interior coating or pouring the floor slab?

That foam/concrete interface will always be a weak link for you, particularly in terms of water sealing. I'm looking at the same place in my basement where curing and shrinkage of the concrete slab has opened up a small space all around the perimeter. It's not big, probably less than 1/8" over a distance of 25 feet or so, but mine doesn't have 6 feet of water over it either.

Posted By ICFHybrid on 04 May 2012 12:58 AM

Can you place a non-submersible pump at or below the level of the water in some other (basement) location?  That would diminish the possibility of losing prime.

The floor of the cistern is at the same level as the basement floor.  To minimize the risk of any leakage into the basement, my pump's outflow water line will penetrate the wall above the overflow outlet of the cistern.  Maybe I'm being too cautious?


Is your experience that the tank accumulates a lot of silt/debris?  What steps have you taken to filter the rainwater before it enters the storage tank?  Do you have a filter on the line from the pump to the inside bladder tank?  Since I'm using the water for landscape irrigation, I only need to avoid particles in the water supply that would clog the heads of the sprinkler system.

Clark

Posted By ICFHybrid on 04 May 2012 09:29 AM

Yes, the floor intersects the ICF wall foam for a cold joint.

Did you consider removing any or all of the interior foam before applying the interior coating or pouring the floor slab?

That foam/concrete interface will always be a weak link for you, particularly in terms of water sealing. I'm looking at the same place in my basement where curing and shrinkage of the concrete slab has opened up a small space all around the perimeter. It's not big, probably less than 1/8" over a distance of 25 feet or so, but mine doesn't have 6 feet of water over it either.

I did consider removing the foam.  Doing so would eliminate the risk of a leak at the floor interface.  It also would increase the volume of the tank a bit.

I chose not to remove the foam in the end because I like the idea that the tank would be insulated and could survive the winter without being emptied. 

I also became convinced that the Blue Max sealant in combination with the rubber tape reinforcement at the floor-wall boundary could handle a small gap opening up as the concrete cures.  Ames states that the coating remains flexible and will stretch a good amount to maintain the seal.  I plan to give the concrete 30+ days to cure before sealing the cistern, so hopefully there will be little movement at this join thereafter.  The maximum water pressure at the this joint will be about 3 lbs/sq in which the high density ICF foam can handle with no significant compression.  That's my hope anyway.

Maybe I'm being too cautious?

Heh heh. What is the definition of "too cautious"? ;-)

I think it is a good idea. Regardless of the height the line rises to, most of the issues with losing prime will be related to where the pump sits. If the pump itself sits below the level of the intake, it will be much less likely to lose the prime.

Accumulating debris and silt is nearly unavoidable without lots of filter and maintenance gymnastics. I have first flush diverters which collect the first (dirtiest) water received during a rain event. If enough water comes down, the first flush diverter eventually fills and (cleaner) water continues on to collection. With a first flush diverter, I think you do away with much of the heavier than water silt and such, but a fine layer still accumulates. There is a flocculent layer that settles on the bottom that can be reduced but not eliminated and of course, some floating debris, which I suspect of being in the transition stage to the flocculent layer.

Just having the intake floating with the end positioned a few inches beneath the surface helps keep the incoming stream away from any of the debris I mentioned. I can't remember if the intake itself had a nylon screen on it. It had a cage to prevent bigger particles, but I can't remember about the screen. Rain events generate some turbulence which you can reduce with some sort of an inlet device that encloses the rushing stream, takes it down to the bottom and directs it upwards so as not to disturb any layers there.

I like filters. I use a gross (coarse) filter with a serviceable screen before the pump to protect the system and two filters following that to protect the bladder tank, plumbing and fixtures. We see the water in the toilet bowls and in the washing machine rinses, so having it come in clean is preferable.

I also became convinced that the Blue Max sealant in combination with the rubber tape reinforcement at the floor-wall boundary could handle a small gap opening up as the concrete cures

I think your tape reinforcement was a good idea.

Posted By ICFHybrid on 04 May 2012 11:57 AM


Accumulating debris and silt is nearly unavoidable without lots of filter and maintenance gymnastics. I have first flush diverters which collect the first (dirtiest) water received during a rain event. If enough water comes down, the first flush diverter eventually fills and (cleaner) water continues on to collection. With a first flush diverter, I think you do away with much of the heavier than water silt and such, but a fine layer still accumulates. There is a flocculent layer that settles on the bottom that can be reduced but not eliminated and of course, some floating debris, which I suspect of being in the transition stage to the flocculent layer.

Just having the intake floating with the end positioned a few inches beneath the surface helps keep the incoming stream away from any of the debris I mentioned. I can't remember if the intake itself had a nylon screen on it. It had a cage to prevent bigger particles, but I can't remember about the screen. Rain events generate some turbulence which you can reduce with some sort of an inlet device that encloses the rushing stream, takes it down to the bottom and directs it upwards so as not to disturb any layers there.

I like filters. I use a gross (coarse) filter with a serviceable screen before the pump to protect the system and two filters following that to protect the bladder tank, plumbing and fixtures. We see the water in the toilet bowls and in the washing machine rinses, so having it come in clean is preferable.

My plan was to use a 1/2 HP 10 gpm @ 40 psi submersible well pump, 8.4 amps continuous.  No problems with loss of prime and no noise inside the house when the pump turns on.  I incorporated a sump in the floor of the cistern into which the pump will be suspended in a vertical orientation so as to maximize the effective capacity of the cistern.  The pump would be placed inside a 4" PVC flow tube to ensure adequate motor cooling.  I'm not sure if that is necessary, but I figure it can't hurt.  I would elevate the flow tube 1' off the bottom of the sump to minimize the uptake of silt that accumulates there.  I have a 35 gallon bladder tank which is large enough to prevent excessive on/off cycling of the pump.  My largest irrigation zone demand is 9 gpm at 45 psi.  The others are around 6 gpm.  Does this plan sound right?  I know a deep well submersible is overkill for this application, but the cost is not much more than a decent self-priming jet pump with the same performance.

Clark

Sounds pretty good. Now, all you have to contend with is curious neighbors when you present green landscaping during periods of water restriction.

Clark - you could build a 2 to 3" berm around your sump pit to stop the flow of sediment into the pit and then clean out the cistern as needed. Remember that you will be going into a "confined" space and that the debris can give of gasses. Vent well prior to entry!

Posted By FBBP on 05 May 2012 10:28 AM
Clark - you could build a 2 to 3" berm around your sump pit to stop the flow of sediment into the pit and then clean out the cistern as needed. Remember that you will be going into a "confined" space and that the debris can give of gasses. Vent well prior to entry!

That's an idea I hadn't considered.  Should the "berm" be solid or porous?  I'm thinking maybe a fine mesh screen encircling the sump to block the silt, but still allow water to enter the sump?

How often do you guys have to clean out your cisterns?