Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Waterleak detection display board design

Status
Not open for further replies.

toddmanqa

New Member
I am in the process of remodeling my house, and I wish to install a leak detection system that will shut off the water if leaking water is detected.

Here is the system:

https://www.a-leak-detector.com/leak-detector-installation.php

I will be using wired only.

However, I want to install a lot of sensors, including behind walls (for the shower valve, etc.), and which won't be easily accessible/viewable.

My concern is that the system will shut down due to a leak, and I wont be able to figure out where the leak is coming from.

I feel that I need some sort of display system that will show where the leak is coming from. To that end, I will assume that I'll need to run a separate wire to each sensor, then run each of those wires to an LED bulb display, then to the shutoff controller.

I'm asking for advice on how to design/create such a display. How do I retain the memory of which sensor tripped?

Or, if I'm trying to recreate something that already exists (and I've searched for it on Google focusing on systems like fire alarm annunciator panels and airplane control panel displays), please point me to the right product.

Thanks,
toddmanqa
 
Welcome to the forum!
Judging by the fact that each sensor shown in the link appears to be just two probes a short distance apart, and that ordinary bell wire is used to connect sensors to the controller, I would guess that the principle of operation is simply detecting a reduction in resistance between the probes when they are in water. Can you confirm that? If so, I think you will need to gain access to the sensors periodically to remove any accumulated dust/insects which could affect their operation.
Will you be using the linked controller, or a home-brew one?
 
Here https://www.selinc.com/printProduct.aspx?id=1016, Now you know what they are.

I'd like to make a few comments. In "typical" fire alarm systems, a resistor is placed across an open contact. Thus, the fire alarm panel can determine if the wire has been cut.

I made a simple annunciator out of 3PDT relays as part of a safety panel. They generally operated with a brief momentary pulse and the relay would electrically latch. That's actually fine and dandy unless you want to know when the alarm clears. So, in the case of a fire alarm activation of this home made annouuniator, I added an LED which went into the SPDT relay of the fire panel.

The design was, the red LED indicated that there was continuity between the Common and NC terminals of the fire alarm panel, so it was on when it was not active. The LED would go out, if the fire alarm was reset. The other part of the announciator used IDEC labelable rectangular LED indicators (expensive) and indicated an alarm.

So, I did not have the tradidtional "monitored contact", but the system would know if an entire wire got severed.

The problem with alarms is that you need to latch them and you need to know their current status. In the case of a fire or safety alarm, contact monitoring is employed which is done with a resistor across the open contact or a two level current trip. The trip needs to be latched and ALSO monitored for currrent conditions.

The 3PDT relay basically used the contacts as follows:
1) Alarm indication
2) Latch the relay
3) what I would call the alarm loop. All of these contacts would be in series.

A power fail required the panel to be reset, which could be a nuesance in your applicaton. So, lots of things to think about.
 
I HATE what I see. They don't even tell you how many wires or wire gauge to use or anything useful. It should depend on how far away it is.

The sensors could be simple switches OR they could be two two electrodes that say the sensor sees 4 mA quiescent and say > 10 mA in alarm. Yea, I made this up. The "4 mA powers the sensor". I doubt they are communicating sensors. My GUT feeling is that they are not simple contacts either.

Alec:
Any water sensor worth their weight has to use AC excitation, in order to avoid plating the electrodes if it uses electrodes. They really don;t even show how to connect multiple wired sensors.
 
My concern is that the system will shut down due to a leak, and I wont be able to figure out where the leak is coming from.

Here is how I look at this. The system you are looking at uses a solenoid valve where the water line entering the house gets turned on/off. Before thinking about anything else I would plumb a ball valve around the solenoid valve. Regardless of what system is used leave yourself a manual bypass valve. This way you are never stuck without water during any sort of system failure.

Unfortunately the system you are looking at doesn't give you the benefit of knowing which sensor detected a leak and turned off the water. I also agree with Alec in that the sensors appear to be common generic type sensors where two wires are bridged by conductive tap water to trigger an alarm and turn off the water. One downside, as simple and maintenance free as these sensors are, if they ever do need maintenance you are ripping out some wall just to get to them. Whenever possible try to leave yourself an access panel. Like the solenoid bypass valve it saves headaches later.

The first problem with the system you are looking at is I can't find any pricing information. When I go to this page all I see is Out Of Stock? Then as mentioned to look at each location of a sensor and call each location a zone, short of hacking into the thing there is no easy way, that I can see, to determine which zone has a problem? I also did not see (unless I missed it) anything about that in the FAQ so you may want to contact these guys and ask.

Annunciation panels like the one KISS linked to are fine, however, that is a $3,700 USD panel. Panels like this are great in critical industrial applications because as KISS mentions they will detect a cut wire or failed sensor circuit. I just don't see them as practical for a residential application, unless of course you live in a mansion of sorts.

Depending on how many zones (locations) you want to monitor a residential unit could be built fairly easily and inexpensive. Each zone could be represented by a few LEDs (Red = Problem and Green = Good) or even just using a single Red LED. The system could latch on a fault and require a manual reset or non latch. Pretty much a matter of how simple or complex you want it to be. To answer your initial question, without seeing a drawing of the system you linked to I am not sure how you could enunciate where a problem was.

Ron
 
See, **broken link removed** Useless too. Especially the Watercop WIFI. It tells you nothing about whether it can report individual sensors.

Might as well design your own.

I actually have two water sensors, but for water infiltration, not for plumbing.

One alarm is local and sits on a towel on windowsill. If water gets in, the towel gets wet and the alarm sounds where you don't really hear it.

The other alarm is wireless and supports 4 sensors, The sensor uses a CR2032 battery. That sensor sits in a "french drain" and the "announciator" sits in the bedroom with the CO/Methane detector;

I'd like to have two or three methane detectors in the house, One old stove has no pilot for the oven, It MUST be lit with a match everytime and no safety cut off either. The stove has pilots and so does the upstairs stove.

The whole point is, for my water detection system I have to be home to remove water manually with a wet vac and, in the other case, I have a bilge pump in the window well. Right now, it is not turned on.
Again, not good. Very, Very Very rare use. Permanent solutions are slow coming.

Water is funny. You could totally miss a water heater tank leak if you don;t have a drip pan. And you might miss a Temperature and pressure valve leak, since that may just drain into a sump.

And you might like the system to work during a power failure. At least locally.
 
Last edited:
You could go this route **broken link removed** . It will give you a 16 unsupervised or 8 supervised sensors and an annunciator. Now, you jut interface it, to your favorite water valve system.

Something like this http://www.alarmsaf.com/apdptc.html , but way less expensive can give you 12 V backup power. I used something like this and it had to be UL listed and everything Class 2 or power limited. This particular box needed a separate battery box. One I used from this company a long time ago just was a 3A power supply and battery with the batteries self-contained.

There are other companies that make them too, typically for camera power.

Although, this is a water alarm, you could consider fire cable which is red and typically uses a PTFE jacket. Problem is, I've never seen it sold by the foot.
250 ' was easy to use up in a work environment.

There is lots of neat stuff here: http://www.grisk.com/security_products.html under Liquid Detection.
Note that their water valve can be operated remotely. So, with a remote contact have at it.

If it were me, I'd love to have SMS notification on a sensor basis. There are systems that can do that.
 
Last edited:
Mr. KISS mentions:
Might as well design your own.

For what it's worth that is what I would do and keep the thing simple. Buy a bag of 2N3904 transistors (two per channel) and a bag of 1K resistors (two per channel, and a bag of 560 Ohm resistors (one per channel) and a bag of Red LEDs (one per channel), and a bag of small 12 volt coil relays, and a bag of sensors like these. I don't see anything costing much. If, depending on needs, as KISS mentions include a 12 Volt battery backup. I wouldn't care as the whole house is on generator backup, but in case of power loss a battery backup would be a nice to have. Making the thing wired makes for simple as long as the wires can be run easily depending on the building. Just about any common doorbell wire or thermostat wire can be used as wire resistance won't matter much. I shoved two meter probes in a glass of drinking water in my house and see about 500 K ohms, your mileage may vary, won't matter much anyway. Likely, the most expensive part would be a good quality solenoid valve for the mater mains supply line. Without working up a bill of materials I don't see the entire affair running over maybe $200 USD (rough guess).

The more fancy and more features added the more the cost increases which is pretty much the same with buy or build. If we want a system, for example, that will call us on our cell and send a text or email then we can add to the cost. If we want the system micro-controller based with a display then, again we add to the cost.

I guess we just hang around and wait for the original poster to return and maybe provide more information. :)

Ron
 
reloadron said:
the most expensive part would be a good quality solenoid valve for the mater mains supply line.

You don't really want a solenoid valve. You want a motorized valve. Too much power dissipation.
 
You don't really want a solenoid valve. You want a motorized valve. Too much power dissipation.

The supply line to my residence is 3/4". I have a common everyday 3/4" 24 VDC coil continuous duty solenoid valve sitting here. Happens to be a Parker manufacture but Asco and others make the same valve. It's a 24 VDC 11.5 watt coil. So at 11.5 watts this thing draws less than 500 mA. The 11.5 watts isn't much of anything. If continuous duty is a concern then I would just use a magnetic latching type which is what I suspect is used in the system originally posted. I say that because then mention is the system experiences a power failure the valve will remain in the last state it was in.

I just do not see a need for a motorized valve or where 11.5 watts is too much power dissipation. If I were concerned with power dissipation I would just run with a magnetic latching solenoid valve but remain with a solenoid valve design. Valves like this typically run in the $50 to $150 USD for a quality brand name like Asco, Parker Skinner and others. I am sure by now there are Chinese imports for much less. I would just use something like one of these. Likely McMaster Carr Part Number 48765K76.
Granted not cheap but I would use that or similar. What motorized valve would you use?

Ron
 
Alec: Any water sensor worth their weight has to use AC excitation, in order to avoid plating the electrodes if it uses electrodes.
While that's definitely the way for frequently/constantly immersed electrodes, it might be a tad OTT for a leak sensor, since one would hope that the sensor would never actually get wet :).
 
While that's definitely the way for frequently/constantly immersed electrodes, it might be a tad OTT for a leak sensor, since one would hope that the sensor would never actually get wet :).

Wet = Bad :) I agree in that these sensors are not made for immersion. They are generally a few stainless steel probes. There really isn't much to them. During a normal existence they should never even see getting wet.

Ron
 
Status
Not open for further replies.

Latest threads

Back
Top