New water system questions

[Alternety]

I am putting together the design for my new water system. I would just like all the expert eyes to tell me if I am doing something wrong.

Residence on well. The basic system is a 3 GPM well with deep pump, 1500 gal unpressurized storage tank, pressure pump, whole house filter for indoor water with seperate unfiltered branch for irrigation.

Filter will support a maximum flow rate of 11 GPM and a continuous rate of 4.5 GPM. Pressure drop is 5 PSI @ 4.5 GPM. House has 3 full and two half baths. Pump will provide 25 GPM. I have two WX-302 tanks (used manufacturers sizing chart) and one 80 GAL tank (currently used with well pump for temporary water).

My thinking is to put the two large tanks between the pump and filter. This would also feed the irrigation branch. Pressure control goes here. This optimizes pump cycling. I will also need a check valve in this configuration to prevent backflow through the filter. That check valve should be between the filter and the first two pressure tanks.

I was going to just tie the filter into the house supply but I am now thinking that I should use the third tank on the house side of the filter to level out pressure variations caused by the drop across the filter.

Any thoughts?

 

[Bob NH]

The system looks good. The third tank on the house side of the filter lets the filters work after the pump reaches shutoff on the first two tanks.

If the filter has 5 PSI pressure drop at 5 GPM when new, then you don't have a big enough filter, or it is a very fine filter such as one of the ceramic types. The pleated or wound filters will handle 25 psi differential at end-of-life so you might want to run the pre-filter tanks at a fairly high pressure. That pressure will end up as your house pressure so that is the upper limit unless you want to install a regulator on the outlet of the third tank, which I don't recommend.

If filter life is a problem, you can get longer media life and more gallons per cartridge if you put more filters in parallel. The lower flow rate per unit allows you to collect more dirt per unit.

 

[Sponsor]

 

[Speedbump]

What are you trying to remove with this "whole house filter" (there should be a law against calling one of these things whole house)?

Do not use a check valve anywhere in the home.

I would get rid of the WHF filter. Then have your water tested to see what you have that may need to be dealt with. The WHF will not address any of the things that 95% of home owners buy filtration for. These things are hardness, manganese, iron, sulpher, bacteria, nitrates and low/high PH. There are many others, but not as common.

What pump are you using going into these three tanks?

bob...

 

[Raucina]

Seems to me all the tanks need to be ganged near the pressure switch and booster pump... I would increase to a high pressure for the general system, then regulate it down just after the house filters. I would use a "spin down" filter first, with a flush valve, then the cartridge filter with a backwash system. Tek supply sells filter set ups with backwash valves installed. Also their filter cartridges are amongst the lowwest in cost. I have never seen pump or tank mfg's suggesting a "remote" tank, and it seems to complicate ones maintenance. A regulator is far simpler and cheaper.

 

[Bob NH]

Since you already have three tanks, and you want to use a filter, there is an advantage to putting part of the available tank volume after the filter. That advantage is that your 25 GPM pump can fill the first two tanks on pressure control, and only about 1/3 of the pump flow is going through the filter during the pump-on cycle. Since the flow rate through the filter while the pump is running is only about 1/3 of the pump flow, the pressure drop is about 1/3 of what it would otherwise be, and you can collect more dirt in the filter before the terminal (end-of-life) pressure drop occurs. The tank after the filter is adequate for household needs.

After the pump shuts off, the pressure differential pressure between the two tank systems forces water through the filter until the pressure in the tanks is equalized. Since the first two tanks have twice the drawdown capacity of the third tank, there will not be enough drawdown to restart the pump until there is some damand on the system.

I have used that setup when I install giardia/crypto filters in surface water systems that already have multiple tanks. That kind of setup increases filter cartridge life.

There is no problem with a check valve between the filter and the third tank (which is where I would put it) because it is after your pressure switch, so the first pair of tanks will keep the system and the pressure switch pressurized.

The 1500 gallon tank that you already have will collect anything that a spin-down filter would collect, IF YOU PUT THE INLET TO THE TANK ON THE SIDE OPPOSITE THE OUTLET. Since your well pump is only 3.5 GPM you will have very low flow rate through the tank and lots of time to settle out the sand.

 

[Alternety]

Thanks for the replies. I really don't want to start another thread of arguments about one filter or the other. I have the filter. It is what wll be used. I am completely aware of my water contents. The filter fixes it very nicely. It is the hydraulics/system I am trying to validate.

The pump is a 1.5 HP DC powered motor with an integrated multistage centrifugal pump. The controller is capable of variable speed pumping. I plan to look at using variable speed control later after I am in the house and can tinker (do NOT let my wife see this sentence).

The reason for the third tank is a buffer to provide higher peak draw than the filter alone will pass. As pressure goes down in the third tank the filter will supply water from the main storage tanks. Flow would start off slowly and increase as the pressure in the third tank lowers. The system would try to bring the pressure in the third tank up to the pressure in the main tanks minus the drop across the filter. The lower the third tank goes the faster the system will replenish it.

An effect of this approach will be to cut off the filter if the main pressure is less than the filtered tank. This is probably not an issue as the main pressure will be restored fairly quickly by the pump.

With a pressure regulator instead of the tank, all of the demand must be satisfied by the filter as it occurs. This approach would see more pressure drop for peak loads and be limited to the max flow of the filter.

Under static conditions the third tank should wind up at about the pressure of the main tanks.

Check valve - why do you say no check valve in the house? I really don't want uncontrolled backflow through the filter. If irrigation water is being used, the third tank pressure could exceed the main tank pressure. The filter will then try to backflow. This will increase the concentration of contaminants in the irrigation water. I don't want that to happen. It is also probably not a good operating mode for the filter.

I also believe there must be a check valve in the main pressure tank side of the system to prevent backflow through the pump. So you would have the pump – pressure control – check valve- pressure tanks and irrigation takeoff–filter–check valve. Or the filter check valve could go on the main tank side but it just feels better on the filter output side.

 

[Valveman]

Valveman

I was trying to stay out of this one but, you have a variable speed pump. Why would you have ANY pressure tanks. The variable speed pump uses a pressure transducer not a pressure switch. The transducer makes the pump hold a steady or "constant" pressure. If the pressure is always at 50 PSI, you never put any water in a tank or get any out. You have to have a pressure on/off bandwidth like 40/60 for the tank to accept any water or express water when needed. If your system always stays at 50 PSI constant, what water is in the tank stays in the tank and gets old and contaminated.

 

[Alternety]

valveman - good point. Simple answer; the controller is a variable speed control. The pump system is not. It lacks the feedback of pressure and the learning capability of a PID controller. I do not have the time to work out an alternative control system. I am trying real hard to get an occupancy permit. An issue with no buffer tank is the pump starting and trying to compensate for a trivial use of water.

Future possibilities include a speed controlled by the rate of pressure change. Faster use/faster pump. My current situation makes it real unhandy to experiment even if I had the time.

I have had one aborted attempt at a pressure controlled pump speed in the heating system. The contractor provided a differential pressure sensor and thought he was getting a controller. Same problem, me having bandwidth to work on it. The variable speed pump also caused unacceptable noise levels.

This reminded me - I am trying to find some parts. I need some level sensors that I can install on some sort of support (e.g., a piece of PVC) along the verticle axis of the unpressurized tank. I need tank full, tank low, and tank empty sensors or a sensor with hystresis that would do the tankfull/tank low function to control the well pump. Power will not be run through the sensors; only low voltage controll signals. I could also use as an alternate approach, a pressure sensor that could be mounted through the side of the tank that allowed multiple outputs for these conditions.

On the pumping side I would like to get more information on the Amtrol'R control than I can find on their site and a place to buy one.

 

[Bob NH]

The system that you described will work fine. Running one pump without a tank to provide water whenever it is demanded is not a good solution for a home that requires anything from zero flow to full capacity of the pump.

With respect to control of level in the unpressurized tank; there are level switches that have adjsutable start/stop levels. You need one (close on low) to control filling which will be near the top of the tank. You can set it at a range of 6 to 8 inches between start and stop.

You need another one (close on high) to prevent the pressure pump from running when the unpressurized tank has insufficient water.

I use switches with SPDT action so they are interchangable. I use the weighted switches (you don't need a pipe) that I get from USA Bluebook. You can find them at www.usabluebook.com and search for 29226 through 29234. The 29231 model has a weight and 20 ft cable. It costs a bit less than $40. The range is adjsuted by moving the weight on the cable, and the height is adjusted by adjusting the total height.

 

[Valveman]

Valveman

With a PID loop and pressure transducer you have “speed control by the rate of pressure change. Faster use/faster pump.†The set point would be a constant pressure of maybe 50 PSI. Some of the newer VFD controls have a pressure bandwidth built into their sleep mode. Sleep at 50 PSI and a preset minimum pump speed. Then not wake up until pressure is lowered to say 30 PSI. Without this pressure bandwidth a tank does absolutely no good. Again if the pressure is always at 50 PSI, no water enters the tank and none is expressed. If you use the control as a glorified soft start and run it with a standard pressure switch, then you would get full draw down from the tanks. I don’t think I would use the Amtrol electronic pressure switch for this situation. If I understand correctly from what little is in the brochure, this switch looks for rapid cycle, dry well, low/high voltage, etc. It would probably see the soft start as an under load and shut down. I don’t think you can use a differential pressure transducer like on a heating/cooling system either, unless you look for differential pressure across the filter. And even then you need some kind of sleep mode with a pressure bandwidth to utilize the tank storage. As for the plumbing, “pump – pressure control – check valve- pressure tanks and irrigation takeoff–filter–check valve†and I assume the third tank goes here. I would put the pressure control (whatever kind) after the first check valve or the pump will come back on the instant this check valve closes causing a rapid cycle problem. The float switches in the un-pressurized tank are a must, but I would use a standard pressure switch and forget the variable speed control. This system is already complicated enough without adding that in. The simpler you keep the system, the more likely water will come out of your faucets when you open one. And even with two big tanks on the control side, I would always run 25 GPM on EVERY sprinker zone or the pump will still cycle too often.

 

[Alternety]

Bob - I have two of those exact float switches. I bought them with the pump. Before installing I decided to test them. One behaves like a clean relay the other has a closed circuit resistance randomly ranging from a couple of ohms to thousands of ohms over many switching transitions. The switches will switch only a control circuit. No significant current to clean the contacts. Relay contacts should be designed to match load expectations. A small reed relay with wetted contacts would be better. I have seen some floats that use this but they are harder to mount. They are designed to go through the tank wall and they do not appear to be watertight where the wire exits the relay.

Vendor won't do anything about bad float switch. Promised pump curves for my pump. Didn't do that either. I can assume I have no support at all on anything from them. I sure hope I do not have any real problems with the pump.

I have also had erratic behaviour on the tank pressure switch. May be the same issue of low current through contacts. Used a bit of paper to clean the contacts and it worked again.

All I am trying to do is get reliable sensors. That is why the Amtrol looked interesting. Solid state, if properly designed, is more reliable than mechanical things.

Valveman - I should avoid letting loose wild hares. But I never learn. Anyway - I see what you mean about that first check valve.

I was not thinking of using the differential sensor in the water system. Just relating an experience (i.e., wild hare). In the heating system I was trying to develop constant pressure across the manifold that feeds the zones. This makes balancing simpler and minimizes energy use. Excess pump capacity in the hydronic systems is usually dumped through what amounts to a pressure regulated bypass valve. Pump always runs full speed even if only one small loop is actually calling for water.

I also understand what you are talking about with the variable speed drive. I have not thought out this approach. The variable speed is there because it is inherent in the control sold with the pump. It will also let me tune the pumping rate to the season or something if it turns out to be useful. I was thinking in general terms of running the pump at lower speed for light loads and higher for heavy loads. I can see how that could wind up essentially taking the tanks out of the equation. That whole thing is a much longer term question. Maybe never. If it tried something, maybe moving the tanks to the other side of the filter and running the pump as a constant pressure system might be useful. This would let the pressure difference between the pump and filtered water storage adjust flow rate in the filter. Possibly a small tank on the pump to create a dead zone for the pump and a little bit of hystresis.

I don't believe I would try to control the pump based on the pressure drop across the filter.

First step - I just want reliable water so I can move in. It is one of those silly old requirements from the county

 

[Bob NH]

You can use float switches with mercury. Always have good contacts. It is well encapsulated. Just that some people get paranoid about the word. If it ever got into the water it would sink to the bottom.

 

[Raucina]

SNAFU - or: how about a cycle stop valve?

Great reading and good tech info, BUT If you really want reliable water put a windmill on the well and raise the water tank on a tower. flush the crud off the bottom once a month. Toss all the other crap in the overworked over-thought bin.

All this about pressure drop across a filter? C'mon guys. Its a sediment filter anyway - it seems - so put it under the kitchen sink. As someone said - use the big tank as your settling basin, create a swirl and a raised outlet. Even then, I dont know anyone in America from a city under the age of 45 that takes water from a tap anymore. POISON! If its not from Figi or France or under a buck a pint and in a cute plastic tube - no one drinks it. We complain about gasoline costs as we swig on some water 5,000 miles from its home at 6 bucks a gallon. Half of that is the cost of the fuel to move it into our houses.

Exception: the guy with mercury switches in his tank needs high tech custom filters just in case Murphys law comes into play

 

[Speedbump]

I couldn't have said it better myself Raucina, maybe I'll go out and reinvent the Amateur Radio. They have gone from analog to digital, I'm sure with a few float valves and a whole house filter, I can get the RF flowing to the ionosphere quicker.

bob...

 

[Rancher]

Alternety,

Here's a solid state relay that's easy to use:

http://www.ssac.com/standard/ff-llc1.htm

I use them on a 3,000 gal storage tank with an electrode every 8"

I have some extras if you need a cheap price.

cow_rancher@yahoo.com

Rancher