How to Integrate Leak Detection Cables in a BMS System

Thread Starter

DJ_AA

Joined Aug 6, 2021
490
I'm currently working on a Building Management System (BMS) that integrates various types of sensors. I'm exploring the possibility of adding a leak detection feature, particularly for use near heating systems or boiler rooms.


I've come across sensor cables like the AT-Eco-SC Datasheet.indd , which seem to be designed for this purpose. From what I understand, these cables are laid around the perimeter of a room or underneath critical equipment, and they detect water by altering the signal in a specific pattern.


Could someone please advise on what type of circuit is typically used to interface with these leak detection cables? Are they based on resistance change, capacitance, or another method?


Any insight or resources would be greatly appreciated!


Thanks in advance.
 

prairiemystic

Joined Jun 5, 2018
425
Leak detection cable requires a module to interface with a BMS, PLC, alarm system etc. EcoLeak cable seems to be #30 AWG conductive teflon jacket, wow dries within 15 seconds. I see mostly resistance sensing is used by manufacturers, it depends on the cable construction, and note their (EcoLeak) included end-of-line resistor is 82Kohm. So you see 82K at rest and then the impedance drops when water is on the cable. I'm not sure how conductive the plastic is, i.e. wet on 1" is detected or not. Capacitance sensing is more complex but more sensitive with some types of leak detection cable.
This all pays for itself the first time it saves equipment, datacenter from water damage.

EcoLeak does offer alarm "panels" (modules) to go with their cable, just three dry contact outputs for AC Fail, Water, or Loop Break. I'd suggest just adding that into the system for least hassle.
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
490
Thanks, I am developing my own controller for other sensors, so I was aiming to add the circuit for such a sensor cable and include in one unit. I think there are multiple companies making such cables.
 

MisterBill2

Joined Jan 23, 2018
27,366
You will need to know and understand the specifications of the cable that you would be using. Probably it is not as simple as it sounds to be. Reliability is always that second issue!!
 

Sensacell

Joined Jun 19, 2012
3,780
Seems simple on the surface, but...
For this to be RELIABLE you must consider a lot of details like:

Galvanic corrosion (NO DC on the sensing wires please)
RF interference immunity
Common mode leakage voltages - conducted.
AC mains capacitive coupling

(help me out here- what else have I missed?)
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
490
Well, in principle, if I have two wires, I could connect one wire to GND and the other to a pull-up resistor. With proper filtering on the signal (pull-up) wire—along with debouncing to ensure that it’s detecting a true short connection—it should ideally be pulled low for a duration of around 15–30 seconds.

As for the resistive test on the other two wires, could this be done using multiple ADC samples and calculating a rolling average?
 

MisterBill2

Joined Jan 23, 2018
27,366
The simple scheme that avoids the hazards listed in post #6, and also avoids the possible shock hazard from unknowing fools biting the leak detection cable to see it it is live, is a well shielded, well insulated, low voltage, transformer with shielding between primary and secondary. The actual current sensing can be done with a sensitive scheme similar to a GFCI system, that can trigger on a very small current. Really, an actual GFCI device could be used, EXCEPT THAT THE LOGIC IS REVERSED.
Our leak sensor needs to switch ON when it senses any current.
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
490
But what would happen if we were to use the cable with a termination resistor, as one half of a voltage divider circuit?

So, is it best to use a low-voltage AC signal on the sense wire compared to DC? Is there a reason?
 

MisterBill2

Joined Jan 23, 2018
27,366
I suggested one scheme to avoid the list of possible issues. CERTAINLY there are other schemes that can sense a change in resistance between two conductors in a cable. I suggested one that I thought was simple and reliable and stable. I make no claim at all that it is the best.
 

MisterBill2

Joined Jan 23, 2018
27,366
So your method, can it also detect the point of leak?
I NEVER made that claim, that it would pinpoint the location of a leak. My sole claims were that it was simple and very reliable. ALL of the leak detection systems that I have worked with covered a small enough area so that a complete inspection of the entire protected section would take less than two minutes.

I would certainly expect that a leak detection system's main purpose would be to alert a human to investigate farther.
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
490
Yes, I would agree it simply there to alert management to look into mater. I saw some controller mentioning they could pin point the location,so wanted to investigate how, therefore mentioned it.

Do you normally use a 2 pin or 4 pin cable for your detection projects?

And what is usual length of your cables
 

MisterBill2

Joined Jan 23, 2018
27,366
The systems that I have used are supplied as complete assemblies with watertight sealed housings. Those include batteries and a wireless link to the reporting portion of the system. So I do not have access to any circuit information at all.
 
Top