Backyard Aquaponics
A place to discuss aquaponics

Here's a thought. I'm sure it's been done a million times before. Any way, it could be made from simple & cheap materials without much mechanical knowledge.

The rod would be a threaded bar, available from hardware stores.

The grey discs would be large washers (large enough to fit inside the chosen standpipe. They would be held in place with appropriate nuts. Fully adjustable.

The orange cylindrical float would be either a standard ball float or something similar (maybe even a large fishing float?). The float would need to be sufficient to overcome the weight of the bar and washers.

The switch would be a standard electronic switch, but it would have to be easy to slide, so that the weight of the bar and washers could throw the switch, and also so the net buoyancy force of the float could throw it the other way.

The switch would then go through a relay to power the pump. The pump would switch on as soon as the growbed (or tank) was drained, and switch off as soon as it was flooded. This would mean the pumping energy was minimised.

that's brilliant The only problem I see is if you have multiple grow beds.

An excellent point. Bummer.

I guess the growbeds could be hydraulically linked (pipes connecting them at the base) so that they fill and drain more-or-less uniformly.

Or you just have one pump for each growbed :-)

It's a nice feedback system. I only see 2 failure points, and 1 down side. The down side is simply the multiple pump issue which has already been discussed. The failure points are 1, if the float doesn't rise due to weight (already discussed), and 2, if the washers aren't quite flush with the pipe they may miss the switch (make the switch a fairly long lever I guess).

Effectively, you've externalized a standard float switch so that it requires a much smaller cross-sectional diameter. Nice.

jimmy,
I thought of something like this, but your solution is more elegant than mine.
Why I stopped designing though, is because it has been done before.
However in quite a different implementation.

if you get polystyrene or other (more robust) solid float. Drill a hole down the middle. Then insert the rod, and add locknuts so the ball can slide along it. then hook the rod onto the switch.

This means for 1 switch, you can have a large hysterisis, rather than 2 switches and a contactor (which is what I did cos I'm lazy)
if you want I can add a diagram if it doesn't make much sense

Why not use magnetic switches (or is that what you said)

Correct me if I'm wrong KP - with your idea you'd need to ensure that the switch remained in the "up" position with the weight of the rod+nuts hanging off it, and you'd have to ensure that the weight of the float (when the water got low) was then sufficient to move it to the "down" position. Is that right?

no I didn't.
Advantage of the large switch with hysterisis, is that you can run 240V at decent current through it. Magnetics generally wont take such a high voltage or load, you then need electronics to back it up. It gets too complicated and expensive.
If going that route, I'd suggest getting an ultrasonic probe. They are almost infallible, no matter what is going on in the tank, but can get a touch pricey

Yeah they generally have a decent spring to ensure this. I am describing something that is commercially available, it is not my idea or design (unfortunately, I could be making money if it were)

This design should be highly affordable with off-the-shelf parts from ordinary hardware and electronics shops. The only one that I've seen online so far had an asking price of $170 !!!

If the growbeds were in a cascading arrangement, one of these floats installed in the bottom bed could control the whole system.

An alternative for multiple growbeds, would be to have one float in each bed, and each could be used to control an electronic valve (as used in automatic garden sprinkler systems), but you would then need to ensure your pump had a pressure-activated cut-off so that when all growbed valves were closed, the pump would stop.

I think the combo of float-switch and electronic valve would end up being similar to the autopot smartvalve system.

You sure you want to run 240v to the switch? Quite frankly I would use it as a signal running low voltage.

As would I in many circumstances.
Check out Steve S's thread. He has some nifty floats (of which he built me one) and I am quite happy to trust this to 240VAC near water. It's an excellent design.
I assume the 240VAC + H2O = dead APer is your worry? 240 can be made quite safe if you're careful.

Just thought I'd post my electronic version of the water level sensor.

It uses a MPX2010DP differential pressure sensor. This component is the size of a fifty cent piece. You connect a piece of flexible black airline to the pressure sensor and the other end is fixed near the bottom of the GB. As the GB fills up the pressure of the air increases in the tube. Using a simple circuit you can compare the pressure in the tube with that of the open air and then determine how full the GB is. A few calculations could actually tell you the depth of water in centimetres.

Pumps or valves can then be controlled or alarms set if any issues arise.

Yes KP that was a concern

Tim
How much is the MPX2010DP and how simple is 'simple' for the circuit. Wouldn't changes in air pressure stuff you around a bit ? eg High's Low's and varn's caused by temps.

10 meters of water is roughly equivalent to air pressure.
So 300mm is 3% of atmospheric pressure.
So I guess a change in atmospheric conditions of 1% would be significant to this experiment.
Does anybody know just how high a high is and how low a low is?
I suspect however it's more likely that it will be 0.1% or less variance in earths atmosphere.