"Input" on an "output"

The NC in series seems very straightforward. Just put 0.1mA or something like that through the loop and detect an open circuit. Add a time-out for instant indication of a broken switch or wire. No problem with different types of solenoids.

The parallel version seems troublesome what with the possibility of pushing the switch when the solenoid is energized and and the relatively low resistance of the solenoid coil. You could probably do something cute taking advantage of the inductance of the coil, but do you want a project or a quick solution?

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward" 
speff@interlog.com             Info for manufacturers: http://www.trexon.com 
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

If you are talking about shorting the 24V supply line to signal it, you bet ter have plenty of replacements ready. It might only work once, or not at all.

No, but cleaner solution:

The proper way is to RF from a handheld controller back to the 24V source, most likely inside your garage.

This controller has numeric keypad. So, you can use "*1" to turn #1 on, "#

1" to turn #1 off, etc. The controller on the right can control eight or m ore relays. The problem is that you have to program it yourself. We only supply you the tools to do it. You need to program both the base and remot e controllers.

http://173.224.223.62/rf/remote.jpg

No, that would depend on what was "upstream" from the "short". E.g., if driven by a current limited source, the button just causes that limit to be engaged.

Already have that (read post). Problem is it is not convenient.

E.g., if you are in the yard and decide you want to use the hose to "fill a bucket", you'd have to run into the garage to "enable" the water at a particular "faucet" (bibb); then, back into the yard to manually open the (mechanical) valve on the faucet; etc.

Much more convenient to push a button *at* the faucet (hose bibb), turn the manual valve on, dispense water and then reverse the process when done. (or, rely on an implicit timeout enforced by the irrigation controller)

Don't you have electric valves already? I.e. 24V solenoid? Typical sprinkler valve work fine with pressurized water. You don't need to open or shut the water source.

Yes. They are upstream from the hose bibbs ("faucets"). They *gate* the flow to the hose bibb.

The manual valve ("faucet") is DOWNstream from the electric valve. It is used to set the *rate* of flow. Including "off", if need be.

Do *you* always use a garden hose with the "faucet" wide open? Or, do you use the "faucet" to adjust the flow to a rate that is appropriate for *how* you are *using* the water?

E.g., if I am washing the exterior windows, I don't need the water to be RUNNING for that entire time. Instead, I turn it on, wet down the windows, then turn it off while I soap them, etc. When they have been "washed" (soaped), I turn the water back on and rinse them clean. When all of the windows are done, I'll rinse out the soap bucket and then turn the water supply "off" for good.

In this scenario, the "button" would be pushed exactly twice: once just before I started (wetting the windows) and once after I had finished rinsing the bucket.

There should only be one valve.

If you build the system properly, it will save lots of walking/running back and forth. But you need something better than the dumb sprinkler controll er in most garage.

At the risk of disclosing too much information too early, someone we are wo rking with is building an RF controlled electronics faucet. Namely, it can adjust flow, rather than just on and off. That's why they need more intel ligent controller at the source (garage), but more convenient remote (hand- held) at the usage site (lawn and yard).

In any scenario, if the button fails, I lose the ability to signal for water -- this is an acceptable loss.

I'm leary of that approach as the button is now a reliability issue. If the button fails, then the function associated with the button should be the only thing that I "lose" (i.e., not being able to call for water on/off).

If the button fails open, I can never activate the valve. Nor can I see if the valve is even present (vs. an opened coil, etc.). I.e., the "circuit" requires immediate attention.

I figure all I need to do is cause a "noticeable" increase in current drawn while the button is depressed. Then, be able to tell that the draw has increased (beyond what it was, previously) or is "at" some particular level (though this doesn't have to be precise).

If the button fails open, I can still actuate the valve, detect the valve as present (not an opened coil, etc.). I can NOT detect that the button is broken, though!

If the button fails closed, I can also continue to actuate the valve, detect an open coil, etc. I just waste more power in the process. And, I can be reasonably certain that an "always pressed" button is probably a defective button and not a user standing outside for "12" hours with his finger holding the button down! :> I.e., the circuit remains functional but degraded.

Trick is ensuring the "difference" between no load (open coil), just button (open coil but button pressed), coil (engaged or idle) and coil with button are all discernible without having to adjust for some "future" coil characteristics.

E.g., I think the coil on the master valve is considerably different than these satellite zones.

No. You're assuming that one valve also does metering. That changes the requirements ($$) for the valve -- should the other

15 irrigation valves similarly be proportional actuators? Even though the individual *loads* they feed are already predetermined flow rates? Should you "stock" two different replacement valves because you won't know which type may fail from calcification, etc.? Will homeowner be able to buy that valve from a local supplier for $20?

How do you connect hose to this "pipe", then? Look for a hose bibb that does NOT have a manual valve as part of that assembly? (!) Or, force that manual valve all the way open and then cut the knob off so it doesn't look like one was ever present? :>

In our case, *two* valves must be engaged in order to dispense water: the master valve gates water into the outdoor plumbing (several manifolds around the yard each with one or more valve assemblies) while a secondary valve feeds a particular "circuit".

This allows us to keep that outdoor plumbing unpressurized unless needed. And, ensures any water standing above soil level drains automatically to avoid frozen pipes (I've added "spitters" to the "load" side of each valve). Also complies with municipal code requirements for "backflow prevention" (so water from our yard isn't ever drawn *into* the municipal water supply).

And, minimizes the chance of a (plastic!) valve failing and flooding the yard (as has happened to two of our neighbors at separate times).

Lastly, ensures the water is gated OFF in the absence of power (each valve can be "manually" opened but that is tedious).

The only "build" decision that saves walking back and forth is having a means of calling for control action *at* the point where you are dispensing water. But, short of *forcing* the user to carry a "control", the only way to do this is to put the control in place without relying on the user to have it in his possession.

Already have "better than" that. :> Yet, with "plain old irrigation valves". What's missing is a way of accommodating user without

*burdening* him to have a control in his possession any time he *might* possibly want water (which could be any time he happens to be outside).

E.g., you can purchase a smartphone interface for your irrigation controller. But, do you want to have to carry your phone so you can turn the water on? How do you make water available to the guy you've hired to paint your house -- give him your phone while you are away at work? :-/

Is your pipework metal or plastic?

If metal then you can use this as either as ground, or if plastic there might still be a small conduction path to ground depending on your water quality.

You'll probably need a second voltage source and monitor current on this to determine if the button is pressed, keeping your solenoid power source isolated wrt earth.

--
Mike Perkins 
Video Solutions Ltd 
www.videosolutions.ltd.uk

For the purpose of your (below) comment, "PVC". Actually, the line *to* the solenoid valve is PVC while the line

*from* the solenoid (to the hose bibb) is copper (for mechanical reasons). [I don't have deep enough pockets to run hundreds of feet of 1" class L copper *to* the solenoid valves :> And, I won't use galvanized under *any* conditions! Corrodes too quickly!]

Note that there are 4 of these "faucets" that need to be individually "addressable" (i.e., 4 "demand" buttons as well).

I figured I already have a common and a "drive" signal at each valve. So, if I can change the load that the driver "sees" (senses), then I can use this to convey information back to the driving end of the circuit.

For example:

- wire has been cut (a connection has failed, etc.)

- circuit has been shorted

- solenoid "open" (coil burned out)

- solenoid "shorted"

- solenoid intact, not actuated

- solenoid intact, actuated

- solenoid intact, not actuated, button pressed

- solenoid intact, actuated, button pressed

- etc. [the last four examples being the nominal cases, of course]

Note that I can't test the button without having a user present. But, I can *probably* make many assumptions with little downside risk. E.g., if the button *appears* to be engaged for more than a handful of seconds, it's probably

*broke* -- people aren't likely to push a button more than briefly... even if the system appears to be unresponsive, chances are they won't *lean* on the button indefinitely!

(My design philosophy calls for alerting the user of problems as soon as you can realistically detect them; especially if you can do so *before* the user stumbles onto them and is, thus, inconvenienced!)

Of course, all of this needs to be isolated from my actual controls (iso-optilators, etc.) to protect them from environmental conditions and "attack". So, whatever is on the "field" side of the isolation barrier wants to be robust and "sacrificial".

OTOH, I *do* have "smarts" available and these are terribly low bandwidth signals.

...

Hi Don

You could use DCC:

DCC (Digital Command Control) - optionally with two-way-communication !

You could use HO-rails, 115/230V wire or coaxial cable to remote control and power e.g. an antenna rotor - with only two wires.

Acually you could remote control more than 100 antenna rotors - or locomotives, lights, railroad switches, (hose) solenoids...

DCC is an open standard:

There are a lot of DIY electronics for DCC:

My MiniDCC© System Block Diagram [uses 16F628 2*16 char LCD display]

Now:

Before: MiniDCC© - A Digital Command Control Do-It-Yourself Project!: Quote: "... DCC MiniStation© ... The MiniStation© has been improved considerably but still uses the same PCB. The new 16F628 with twice the memory offers a pin for pin drop-in replacement for the orginal 16F84A. Now you can add 128 speed step, F1-F4 function control, auto-repeat keystroke and up to 123 step of combined route/turnout sequencing just by replacing the chip. Assembly code for the new chip is available free. A pre-programmed 16F628 is also available. Best of all, there is no price increase for the new chip! See the menu below for details on how to build your own. ... DCC MiniBooster© An efficient and simple to build DCC booster has been designed using the most recent MosFet integrated chip from National Semiconductor®. A double-sided PCB prepared by John Zajdler is also available (see Parts menu). Now you can control up to 4 trains simultaneously, over 120 turnouts and program all your decoders using any one of the 4 NMRA standards at a fraction of the cost of a commercial unit! ..."

Loco-decoder for Märklin-Digital in DIY

-

"Locomotive" "talk-back":

Recommended Practice RP-9.3.2:

(main: )

/Glenn

Ouch! This seems like overkill for the "one bit" I want to transmit. :< It also looks like it would require a bit of "smarts" on the button end of the link.

Note that the button would be located in a rather hostile environment -- temperature, precipitation, etc. So, I am hesitant to put much of anything out there beyond the switch itself (e.g., resistor is probably a safe bet but a *cap* would have problems holding its value, etc.).

[Even the switch is a bit of a challenge to locate suitable candidates that are easily mounted (in a chunk of concrete) and impervious to the elements...]

But, these are excellent references! I will dl/bookmark them. Thanks! It looks like it might be a practical alternative if I redesign the controls to reside *at* the valves. I think the best scheme is a set of *tiny* valves and controls that can be deployed at individual plants from a single "water bus" -- instead of creating a small number of "water circuits"/zones that feed a large number of plants concurrently. This could also open the door for "local sensing" whereby the individual needs of the plants could be monitored.

(all these "tiny valves" would still have to communicate with some central "supervisor" to ensure they don't overdraw the available water supply -- as well as telling them what criteria they should operate under)

--don

you've atcually got one, it's called "ground"

yeah, that'd work too.

diodes are fairly rugged

sounds tricky.

that one's probably slightly easier.

only if it fails open, if it fails closed...

have you seen that video where that guy wires three switches and three lamps in series...

--
?? 100% natural 

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

That's called a "hose union", a fairly common plumbing fitting, probably even more common in domestic irrigation.

that one's too easy.

--
?? 100% natural 

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

Hose union is MGH x MGH. I.e., connect two *hoses* together. More common to find FGH x FGH (to allow the "male" ends of two hoses to be interconnected -- as the female end of one of them is probably terminated at a bibb, someplace!)

That would require FGH fitting into which it would mate. More likely you would sweat a 3/4 FPT adapter and screw a 3/4 MPT x MGH nipple into it.

[ISTM, you can't find an MGH sweat adapter hence the need for the FGH fitting. But, I can't imagine a FGH sweat adapter either! Therefore, the FPT adapter with the MPT x MGH nipple]

When was the last time you saw a MGH connection sticking out of a wall that *didn't* have a (manual) valve attached to it? :>

(IME, never -- unless it was the male end of a *hose* already mated to a traditional hose bibb)

The joys of plumbing!

My hose bib in the back yard started leaking. Got a new washer, turned off the water, and tried taking it apart. An hour later, I gave up! Can't get the stem out, looks like lime build up is so thick that it has made its own washer in there. So, it looks like it is time to replace the hose bib.

Go into Home Depot, talk to gal there, and she gives me something called a sharkbite fitting. You just cut off the old sweated on hose bib and put it on and screw on a new hose bib. Sounds doable.

Next day, turn off water and go to get to work. First, I compare the new fittings and the existing - not the same size! She gave me a 1/2" sharkbite, but OD of existing is 1". Turn water back on...

Go back to Home Depot, looking for pipe with OD of 1". She doesn't know of anything like that. Brings in old guy (my age!) and he thinks about it, and figures it out. Plumber used 3/4" pipe and a 1/2" hose bib, but sweated a coupler across both. Coupler would be flush with stucco wall. So, now I have a torch and plumbing kit, a new 3/4" sweat to male thread, and a new 3/4" hose bib.

So, now I have a new adventure - learning to sweat pipe! Any hints and suggestions? Pipe is painted white - do I need to remove paint first, or just burn it off with torch? How hard is it to sweat off a coupler like this? Any advice will be appreciated!

Charlie

Don't burn the house down would be my advice.

Soldering copper pipe is really easy if the parts are fluxed and bright-clean or already tinned. It's about impossible if they're dirty or there is a slow drip of water.

The trick I learned to deal with the latter is to stuff a plug of white bread (no crust) into the pipe. The bread dissolves later.

I don't get why you have the electric valves. What are they for when your mechanical valve is right there?

About your proposed approach with a current limited driver for the solenoids and a button to activate the current limit, I don't get the problem. The solenoid will draw some current, but it will be for a considerable time. The button will draw a current for a short time. The discriminator will be time, not current.

The PSU will not need to limit the current as much as measure it. The button can have a simple resistor to set the button current draw. If a current change is seen that lasts for the right amount of time it is considered a button push. Currents for other time profiles will be considered the solenoid or a fault. It would even be possible to implement other commands with other time durations. 0.5 second press turns on/off this faucet, 3 second press turns off *all* faucets for example.

What ever you are using for a controller would need to be programmable at an appropriate level to implement this.

--

Rick

Some irrigations controllers are able to sense a shorted solenoid valve.

RTFM, or momentarily short a valve and observe what shows on the display. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.