best way to control city water pressure valves

Using an Uno to build a aquarium auto fill system. Can anyone suggest reliable method of controlling city water pressure on and off? I worry about reliability (floods) and water hammer from a solenoid valve like in a washing machine. Best bet is to use 2 valves for redundancy.

To minimize water hammer, one should use a vertical stub at the valve per standard plumbing practice.

If you are using solenoid type water valves, commonly used in things like washing machines and dishwashers, the valves are very reliable , and fail safe in the sense that the water pressure keeps the valve off with no power applied to the solenoid. You only need to make sure that your controlling circuit cant fail in a manner that keeps the valve energised.

If your worried about water hammer add a throttling valve inline with the solenoid to slow the flow rate down. Even with the valve nearly 100% closed it should still be plenty to keep up with the evaporation of most fish tanks.

Also are you filling the tank directly from tap or are filling an overflow filter or refugium that is out of sight? If it's the later, why not just a toilet fill valve? You can pick one up for under $15 (U.S.), and very reliable, which means a lot considering if your device fails you could facing thousand of dollars of water damage to your house.

I plan on using a pressure reducing valve on the supply side. Also plan on using a floor mounted water sensor to look for overflow conditions. I wanted to use 2 valves inline for redundancy and flood protection. Another consideration is the water level sensor. I don't want to use a float, but some kind of immersion probes - stainless steel probably. Or maybe a capacitive or optical sensor. That's the next consideration.

My primary concern is 99.99% reliability on the water valves. I've considered using both a solenoid (i.e. washer machine type) with a motorized ball valve run by a servo.

My aquarium is fresh water, with a under tank canister filter 55 gallons. It is used to raise catfish who thrive in warm water. I run the water at 79.5F that leads to much evaporation, hence my need to top it off all the time.

Actually writing the software is the easy part. Getting the required control and sensor devices together is the challenge on this one. Thanks for any input.

Well if you do plan on using a motorized ball valve,make sure you pick one up that is power open and spring closed.

As far as water level sensing what a about an diaphragm pressure switch calibrated of inches of water column? If you invert a tube vertically into the water, I'm thinking the type of tubing used on an under gravel filter, attached to a length of air line tubing to the switch, there might be enough displacement to trigger the diaphragm switch. You can get adjustable switches, that have a range of 0-3"w.c., and a lot of them are SPDT.

http://www.grainger.com/Grainger/CLEVELAND-CONTROLS-Air-Sensing-Switch-4XB70?cm_sp=IO-_-IDP-_-RR_VTV70300505&cm_vc=IDPRRZ1

Solenoid type water valves need a minimum water pressure to work. If you dont have enough pressure, the valve will never turn on. Its likley that adding a pressure reducer before the valve will simply mean that the valve wont turn on.

The solenoid could always be installed before the regulator, it's not the pressure that will cause the water hammer it's the flow rate and how quickly it's stopped.

Measure how much water is required to top up after your sensor indicates a low condition, or easier, how much time it takes to provide that water. See how long it takes before water is needed again. Use the second interval to cap pumping: if you already topped up in that timeframe and the tank is asking for water again, don't pump any and provide an alarm suggesting that the level sensor has gone bad. If it's top up time, limit the pumping to the known top up amount, irrespective of what your level sensor may say.

In any event, also cap how much water can be pumped in 24 hours to reduce the potential damage.

WildBill this is an excellent idea. Limit water to specific quantity per day and include some kind of alarms. This would have further benefit of not delivering so much water that the temperature within the tank would drop leading to risk of shocking the fish.

Looking more at valves I am wondering if those valves used for sprinkler system irrigation would be a better option. They are 24VAC not 120VAC.

I also have to figure out a way of using a carbon filter to remove chlorine and chloramines. I think a refrigerator carbon filter would do the trick.

As for the water level detector I have narrowed it to probes or using a clear acrylic tube with a opaque floating ball inside and led/phototransistor to locate the ball. I wonder if the small amount of AC on the probes would bother the fish. I worked at Sprague Semiconductor years ago as an intern and learned you must use AC with water probes so that you don't get electroplating like you would with DC.

I guess I could wrap up the design with a temperature sensor/alarm too. Integrated with the in tank heater. I once had a heater fail due to welded contacts and fried my fish about 20 years ago. I lost many fish in a 75 gal. tank.

Still the biggest design decision is on the water valves as they have the most potential for property damage if they should malfunction.