Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Knightrider]

It's kinda like my pool fill thing.  I use an old washing machine solenoid valve attached to an appliance module.
You can get these from just about any appliance repair shop for just about next to nothing.  You can also use the water valves from dishwashers and ice makers.  They need a custom designed enclosure and a GFCI.

You are dealing with 120 and water which don't mix well, so be careful.  Also, if you use the washing machine valve, you'll need to cap off the output somehow and wire up both valves to open.

Just be careful and safe!!!!!

I'd also check that you have a valve designed for the voltage you are using.  I have yet to see a washing machine valve run on anything other than 120, but times change.  I also had to isolate mine with a relay to keep the valve from triggering the local sensing circuitry, but a night light would do the same.

Keep it dry and protected!!!!!!!   This is not a hack for a novice.

[DChisholm]

OK, this thread has been dormant for over 4 months so maybe I'm in line for some kind of "Thread Necromancer" award.

I recently hacked one of those Orbit p/n62035 solenoid valves in conjunction with a small consulting project.  The details are in my notes (back at the office tonight) but I have verified (by measurement) some of the speculation in this thread.

1.  The valve's water flow is unidirectional.  The water outlet can NOT be effectively exchanged with the water inlet.  Also, the valve's elastic diaphragm apparently uses the supply pressure to provide some kind of "servo action".  Consequently, determining the "open" or "closed" status of the valve by using your mouth to blow through the thing may only give you a headache and some inconclusive results.  It needs some supply-side water pressure before you can actually observe results!

2.  The valve actually has TWO solenoid coils.  One is pulsed to open the valve; the other is pulsed to close the valve.  (I don't recall which is wired to the tip of the phone plug and which goes to the ring.  A quick experiment with a pair of 9V batteries can answer this question.)  However, I was able to CLOSE the valve by applying a reverse-polarity pulse to the OPEN coil and vice versa.  I don't know if this is an intended mode of operation, or if it could eventually damage the valve, etc.

3.  The two coils have very different electrical characteristics.  The "Valve OPEN" coil has about 4 ohms of DC resistance and a few dozen millihenries of inductance.  The "Valve CLOSE" coil has only about 1 ohm of DC resistance and a few millihenries of inductance.

4.  The Timer/Controller box operates the valves by charging a 2200 uF capacitor up to about 17 volts, then discharging it through the selected solenoid coil for 20 mSec. 

5.  The 17V is created by some kind of switching supply shortly before being discharged through the solenoid.  You can see the saw-tooth ramp on the 17V rail.

6.  When the Timer/Controller turns a valve OFF, it first operates the "Valve ON" coil, then recharges the capacitor and operates the "Valve OFF" coil.  This all happens in about 100 mSec, so an end-user wouldn't be aware of the multiple pulsing.   I don't know why the "OFF" sequence is done like this.  I could speculate that the diaphragm might take a "set" while the valve is OPEN ("ON"), and the initial "Valve ON" pulse is intended to dislodge the "set" and ensure a good seal when the "Valve OFF" pulse is executed.

[Brian H]

  Thank you for the detailed tests.
They will be a great help to many with thoughts of sprinkler controls.

[DChisholm]

Here's the electrical details.  Note that what I previously remembered about coil inductance is off by an order of magnitude:

The "Valve-OPEN" (Water-ON) coil connects to the phone plug's ring terminal and opens the valve when the shell is positive with respect to the ring.  (As I noted above, reversing this polarity will CLOSE the valve but I don't know if that's recommended.)  I measure 4.0 ohms of DC resistance in series with 4.5 mHy of inductance.  The measured duration of the actuation pulse is 22 mSec.

The "Valve-CLOSE" (Water-OFF) coil connects to the phone plug's tip terminal and closes the valve when a current pulse flows from shell to tip.  (Again, reversing this polarity will OPEN the valve.)  I measure 1.2 ohms of DC resistance in series with 37 uHy of inductance.  The measured duration of the actuation pulse is also 22 mSec.  However, there appears to be some kind of current limiting in the Timer/Control box that keeps this current under an amp or so.

The freewheel (kickback) diodes are located in the Timer/Control unit, not the solenoid valve itself.

[Brian H]

Thanks again for the updated details.

[DChisholm]

The bad thing about this valve, for the project I was considering, is there is no electrical read-back of its present state.  There's no (electrical) way to tell if it's OPEN or CLOSED.  Even if you deliberately pulse it to one state or the other, there's no confirmation that the operation was successful.  It would help if there was an auxiliary contact, or a pressure differential switch, to tell you which state the valve is latched into.