Irrigation Pump Controller Project

JSWrightOC

I know this is a long post, but I wanted to make sure I explained it well for those interested in assisting me!!

I've got a project that I'm working on that involves controlling some irrigation pumps fed from water storage tanks. The controlling of the pumps is easy - but I am curious as to what might be the best way to measure the water level in the tanks. I want to avoid anything mechanical or involving routine maintenance (float switches, electrodes, float on a screw shaft driving a potentiometer, etc.). The water level displacement is about 6-10 feet. So far I have come up with the following; feel free to suggest new, improve upon or shoot down these ideas.

-My first thought was to use some kind of ultrasonic range finding device. A Parallax PING))) would probably be ideal, in fact I think they mention this kind of application - but at the same time, not guaranteeing that it will work right because of moisture. That's my biggest concern, is moisture damanging the transducers. Also temperature and humidity affect the speed of sound, though I have not researched if this would throw off my measurements significantly or not. (On a side note - anyone ever considered sonically measuring really long distances to say, read the water table in a well?)

-My second thought spawned from fixing an old top-loading clothes washing machine. It uses a vinyl tube that goes down into the washing tank behind the clothes basket (hidden from view) and a barometric switch on the other end. The pressure exerted on the air in the tube from the water filling up around the tube moves a diaphragm in the barometric switch, and a knob adjusts how much displacement is required to trigger the switch, controlling how much water is required to terminate the fill cycle. I thought I might extrapolate on this idea, and instead of using a barometric switch I would use a linear pressure transducer, and read its value. I can get away with this method because the pump controller electronics box will always be above the tank, which is buried in the ground. This would put all of the sensitive electonics and signal wiring inside a box, and I could run the tube out of the box into the tank easily. The only problems I saw with this idea is if develop a leak, there goes the system (it will now perpetually read empty), and possibly the air in the tube gradually dissolving into the water, causing the water level in the tube to rise (and again, causing a lower-than-actual measurement). That could be offset by a small air pump that continuously pumps a very small amount of air into the line (aquarium pump perhaps?) and it should not dramatically affect the pressure measurement in the tube.

I suppose I should start testing some of these plans, but I wanted to know if anyone else had addressed the same application before I started spending money. As Johnny 5 would say, "Need Input!"

Bruce Bates

JSWright -

I suppose this falls in line with your "barometric switch", but why not just put a pressure sensor at the bottom of each tank, as the greater the volume of water, the more pressure will register on the sensor. I have no idea how linear the results are from such sensors, but if need be, you could always use a table look-up for interpolation.

Both Motorola and Honeywell have a large selection of pressure sensors from what I understand.

Regards,

Bruce Bates

Post Edited (Bruce Bates) : 9/15/2007 2:49:28 PM GMT

Shawn Lowe

On kronos robotics web site, I saw a project where he used wires on the outside of a container of water that acted as a variable capacitor. The water level would change that capacitance between the wires. Then he hooked these into a 555 timer circuit and read the frequency output with his controller. As the capacitance changed, so would the frequency of the multivibrator circuit. It would require calibration, but it should be fairly accurate

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Shawn Lowe


Maybe I should have waited to do that......

JSWrightOC

Bruce -

Thanks for the quick reply. Yeah, that sounds like what I was going to do, except I would be remote mounting the sensor and running the pressure to it, versus mounting the sensor in the water and providing a "static" air supply to it. There are ups and downs to each design. AFAIK, the pressure exerted on the transduscer would be linear with respect to the water level - any nonlinearities would be in the transducer itself.

Shawn -

The capacitive wires sounds very interesting - however the tank is concrete and buried. Might it work if I were to place the wires in a piece of PVC pipe lowered into the tank, or perhaps two pipes next to each other in the tank? Also let me note that the control box is located 3-5 feet from the tank, and I wonder if electrical interference and insulation resistance (when it's really humid outside) might be an issue. That's definitely worth a shot. Heck, I have the parts to try that out right now!

Shawn Lowe

It might work, like you said, it's worth a shot!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Shawn Lowe


Maybe I should have waited to do that......

JSWrightOC

What I would really need is some plate electrodes with a dielectric around them. That would give me a higher capacitance value, which would allow me to reduce the resistance in the RC circuit, making it less sensitive to noise. It would also lower the frequency somewhat, so the inductance of the wiring would be less signficant.