Hey guys,
I'm trying to find out if there is a valve that can be electrically opened and closed that would be suitable for water. If anyone has ever heard of such a thing or can suggest to me how I build one, please let me know 
If you have a reasonable pressure, ie mains, there are all manner of solenoid valves available at your local hardware or irrigation shop.
If you don't have much pressure it's a lot harder.
Rob
Graynomad:
If you have a reasonable pressure, ie mains, there are all manner of solenoid valves available at your local hardware or irrigation shop.If you don't have much pressure it's a lot harder.
Rob
Why is pressure necessary? I was hoping to use a small pond pump. Are they automatically closed by default or open?
More information required :
a) temperature of water
b) pressure across valve when closed
c) flow rate through valve when open
d) power supply used to operate valve (low voltage 6/12/24DC or mains 110/230AC)
jackrae:
More information required :a) temperature of water
b) pressure across valve when closed
c) flow rate through valve when open
d) power supply used to operate valve (low voltage 6/12/24DC or mains 110/230AC)
Above freezing, less than 30C.
Very little pressure. Maybe 2 psi tops.
A little pond pump. Maybe 50gallons per hour.
If possible, 5vDC or whatever I can get the arduino to max out at with out using a relay.
Quite often pond pumps are centrifugal pumps which don't always need to be closed when turned off.
Turned on they pump water up, once you cut the power gravity will let the water above the pump flow back through the pump into the reservoir.
Why is pressure necessary?
Because the cheap solenoid valves have a spring to keep them closed. The solenoid needs help from the water pressure to open against the spring. 2PSI won't do it I think.
5vDC or whatever I can get the arduino to max out at with out using a relay.
Whatever you find you will need a relay and/or transistor, the Arduino is not capable of driving things like motors and solenoids by itself.
Rob
Simpson_Jr:
Quite often pond pumps are centrifugal pumps which don't always need to be closed when turned off.
Turned on they pump water up, once you cut the power gravity will let the water above the pump flow back through the pump into the reservoir.
Yes, I can attest to this. I do not want the water to flow back through.
If I must use a relay that is ok. I just thought I'd check and see.
I've seen a lot of thread on this and other forums where people wanted a cheap low-pressure valve.
I have yet to see anyone actually find one. If they did they didn't report the fact.
There have been all sorts of methods proposed to bodge something up, but nothing I've seen that's reasonable. One of the problems is that ball valves require a huge amount of torque to turn them
OTOH if you are pumping and there is no pressure when you stop then you can just turn off the pump and have a non-return (or check) valve.
What's the exact application?
Rob
If you want a low pressure "solenoid valve" then go for a motorised central heating zone valve. These operate from mains AC but more importantly are readily available and reasonably cheap. In fact you should be able to find one at your local scrap dealer for pennies.
Alternatively have you considered simply turning off the pump
Graynomad:
I've seen a lot of thread on this and other forums where people wanted a cheap low-pressure valve.I have yet to see anyone actually find one. If they did they didn't report the fact.
There have been all sorts of methods proposed to bodge something up, but nothing I've seen that's reasonable. One of the problems is that ball valves require a huge amount of torque to turn them
OTOH if you are pumping and there is no pressure when you stop then you can just turn off the pump and have a non-return (or check) valve.
What's the exact application?
Rob
Basically I'm building a water ballast tank. Which is used to control positioning. Ahem, I refer to it as poor mans hydraulics. I want a fairly heavy object to face the sun during the day time. So my plan was to use 2 axis positioning by having counter weights perpendicular to each other (the object is a large rectangle pivoting on a CV Knuckle looking object in the center of the back of the rectangle). On the opposite sides of the weights, I would have containers that could hold water. I would use multiple photo resistors to track where the sun is, and according to a difference in readings from the sensors I would turn a pump, and open 1 of 2 valves to fill either of the tanks (or drain them back through the pump) to match the counter-weight.
I realize it's all possible with servo's and motors but I'm pretty limited for this project financially and I figured big servos and motors to move such a big object would be expensive and I also don't require the speed of a servo/motor. The sun doesn't change position very fast. The system also needs to be able to with medium wind speeds. I figured the ballast-counter-weight system might allow it to wiggle here and there.
jackrae:
If you want a low pressure "solenoid valve" then go for a motorised central heating zone valve. These operate from mains AC but more importantly are readily available and reasonably cheap. In fact you should be able to find one at your local scrap dealer for pennies.Alternatively have you considered simply turning off the pump
Thanks, I'll google those and see what I come up with. I have a friend who works with Heat Pumps, perhaps I should ask him.
Turning off the pump unfortunately will allow water to trickle back through the pump.
If anyone has ever heard of such a thing...
You don't indicate where you live but I assume you have seen a washing machine at one point or another. If you follow either of the hoses you will find that it is threaded to an electrically operated water valve.
Don
I can see at least one problem with this approach, what they call "sticktion".
In other words it takes more energy to start something because it's "settled" in one place and the grease, dirt, general friction etc takes effort to overcome. Once it is overcome things run more freely.
So the problem is you pump water until it starts moving, then it moves too fast and you have to pump back again. If you pump too much back it may stop and you're back to square one. Don't pump fast enough and the thing runs away, pump too fast and it goes the other way.
This assumes that you have feedback and actually know what directing things are pointing. Even if it works perfectly you still need feedback.
If the above happens I doubt your pump will be able to move water fast enough to stop it, so the panel (or whatever it is) will slam to the end position.
If this is a panel or something with a large surface area then it will be subject to wind loading and once again without feedback you won't know it's moved. By definition the system is well balanced so therefore easily upset by wind, a bird or just about anything.
All in all I think you're cruisin' for a bruisin' with this approach, an object sitting on a CV is inherently unstable, once it gets just a tad out of balance it will take off and your system doesn't sound like it will have the response time to correct problems.
EDIT: If you just want to follow the sun you can usually do that with a motor, some threaded rod and a nut (for one axis admittedly, double for two). Very cheap.
Rob
Graynomad:
I can see at least one problem with this approach, what they call "sticktion".In other words it takes more energy to start something because it's "settled" in one place and the grease, dirt, general friction etc takes effort to overcome. Once it is overcome things run more freely.
So the problem is you pump water until it starts moving, then it moves too fast and you have to pump back again. If you pump too much back it may stop and you're back to square one. Don't pump fast enough and the thing runs away, pump too fast and it goes the other way.
This assumes that you have feedback and actually know what directing things are pointing. Even if it works perfectly you still need feedback.
If the above happens I doubt your pump will be able to move water fast enough to stop it, so the panel (or whatever it is) will slam to the end position.
If this is a panel or something with a large surface area then it will be subject to wind loading and once again without feedback you won't know it's moved. By definition the system is well balanced so therefore easily upset by wind, a bird or just about anything.
All in all I think you're cruisin' for a bruisin' with this approach, an object sitting on a CV is inherently unstable, once it gets just a tad out of balance it will take off and your system doesn't sound like it will have the response time to correct problems.
EDIT: If you just want to follow the sun you can usually do that with a motor, some threaded rod and a nut (for one axis admittedly, double for two). Very cheap.
Rob
I think you are correct. Can you explain this rod/nut motor thing? I have troubles visualizing stuff with out an image. I have a fairly strong programming background, but I'm kind of weak when it comes to physical engineering. I hope to improve
Threaded rod (aka allthread) is just a very long bolt, if you put a nut on it and turn the rod the nut moves linearly along the rod.
So by mounting the nut to the thing that needs moving and turning the rod the "thing" moves.
So it's just a matter of mounting things in the right place.
It may help if you tell us what needs to be moved.
Rob
Graynomad:
Threaded rod (aka allthread) is just a very long bolt, if you put a nut on it and turn the rod the nut moves linearly along the rod.So by mounting the nut to the thing that needs moving and turning the rod the "thing" moves.
So it's just a matter of mounting things in the right place.
It may help if you tell us what needs to be moved.
Rob
The object needing to be moved is a passive solar heater. Basically a big wooden rectangular flat box, with a bunch of aluminum painted black inside that absorbs heat from the sun.
So how would I attach the metal threaded bolt to a motor? If I wanted the solar heater to move on two different axis's would I need two motors? How about attaching the nut to the panel? Any tips with that? Thanks for the advice man! I really appreciate it. My official project thread is over here http://arduino.cc/forum/index.php/topic,54992.0.html
Attached is a quick drawing of the general idea.
As the rod turns the nut moves up (or down) and moves the panel with it.
Note that there are three pivot points (the red dots), obviously the panel, but also the motor and the nut have to pivot as well. This means that both the motor and the nut have to be in a mount of some kind that allow them to pivot. Pretty easy to do.
This is essentially a "linear actuator" done on the cheap.
This can be mounted several ways and may have to be different depending on the clearances, but this was quick to draw.
Another thing to think about is the need to rack in two axis. One of them (the altitude?) only changes over weeks and months an dit's not normally worth the effort to dedicate control gear to that. Either leave the whole apparatus at the optimal angle for you latitude or have a manual system that you adjust every few weeks.
Another thing to think about, you will need an RTC or some method of knowing the time/date etc.
Rob
Graynomad:
Attached is a quick drawing of the general idea.As the rod turns the nut moves up (or down) and moves the panel with it.
Note that there are three pivot points (the red dots), obviously the panel, but also the motor and the nut have to pivot as well. This means that both the motor and the nut have to be in a mount of some kind that allow them to pivot. Pretty easy to do.
This is essentially a "linear actuator" done on the cheap.
This can be mounted several ways and may have to be different depending on the clearances, but this was quick to draw.
Another thing to think about is the need to rack in two axis. One of them (the altitude?) only changes over weeks and months an dit's not normally worth the effort to dedicate control gear to that. Either leave the whole apparatus at the optimal angle for you latitude or have a manual system that you adjust every few weeks.
Another thing to think about, you will need an RTC or some method of knowing the time/date etc.
Rob
Why would need an RTC? Isn't the point of having a solar tracker so you set it and forget it?
I've been thinking that it probably isn't worth the extra effort to make it pivot on the second axis. But I thought for a challenge it might be fun, because eventually I want to create a solar parabolic stirling generator and that requires the 2 axis tracking style system.
Thank you very much for the diagram! It certainly helped me make sense of it all. I suppose I could get an electric motor that has a threaded drive shaft off of it and I could use some hardware to bolt them together. But how do you attack a metal nut to a piece of wood? I could see me screwing a piece of metal to it and maybe welding it? But then again it has to pivot right ... Hrm...
Why would need an RTC?
Maybe not because I guess you want to use LDRs to track the sun.
But how do you attack a metal nut to a piece of wood? I could see me screwing a piece of metal to it and maybe welding it? But then again it has to pivot right
Yes it would almost certainly require some welding if you want to do it on the cheap. Not hard but does need some metal fab skills.
You could probably bodge something up, for example a hose clamp can be used to hold the nut, the clamp could be bolted to a plate that is screwed to the wood. The bolt is your pivot point. A search through a junk yard may give inspiration.
I suppose I could get an electric motor that has a threaded drive shaft off of it
I doubt you will find a motor with the rod already attached, normally you couple a motor shaft to the rod.
There's quite a lot involved, even for a "simple" application like this.
Rob