We have been using a pair of NCD PR-16-9-10A relay boards for several years to support a system that controls a pump in a water well as a function of the level of water in a large remote storage tank. It had been working reliably and unattended for the last 3 years or more, turning the pump on and off on command about every 2 or 3 days (dependent on consumption, of course).
Recently we noticed a significant increase in the electric bill for the pump. Further investigation revealed that the 1600 gallon storage tank was overflowing. Obviously the pump was not turning off when it was supposed to. An incoming request for more water would cause the relay to close successfully and turn the pump on. A subsequent incoming request to turn the pump off would have no affect; the pump would continure running.
After considerable additional investigation, it was determined that the relay on the NCD board that we were using - #1 - was not opening when the firmware running on the imbedded Particle Boron commanded it to do so. Note that the normally-open terminals on the relay were the ones to which the pump was connected. Note also that the red LED associated with that relay correctly reflected the desired on or off status of the relay, so the only way we were able to verify that the relay was not opening was by using a multimeter to confirm that the contacts were still closed.
Assumng that the relay had “gone south”, we moved the pump control wires to the second relay and modified the code accordingly, only to discover that precisely the same problem exists with it. That revelation is now causing us to believe that there might be a bigger problem with the NCD board itself rather than just a pair of bad relays.
Has anyone else seen this and, if so, is there a fix short of replacing the board (we have a spare). Any thoughts would be much appreciated.
This sounds like it could be a power supply issue.
It is possible that the power supply is providing enough voltage/amperage to light the LEDs and run the, but not enough to power the relays. Try a new power supply and see if that resolves the issue.
Possible Jacob, but unlikely, since closing the contacts (after a reboot to release the relays) works perfectly at all times. Do you know if it takes more power to open a relay than to close it? If so, then you might very well be right. We will try that anyway just in case. I’m sure I have other power supplies lying around (12 volt, right?).
So it takes power to activate the relay and close contact to the NO position (continuity between NO and Common). It takes no power to deactivate the relay and put it in the NC position (contact between NC and common).
If the relay is able to go to the NO position then it won’t be a power supply issue unless this relay controller shares power with an inductive device.
When you activate the pump are you simply closing a contact into the pump to activate it or are you switching power to the pump? Does the relay function as expected with no load attached?
A bit confused Jacob. NO means that the contacts are open - no continuity between them - until commanded to close. They need power to close, right? But when commaned to open, I would think that would require no more power and maybe nothing at all. If I have that correct, then insufficient power should be reflected when trying to turn the pump on, not off. Am I missing something?
Just to reiterate the obvious: this whole thing was rock-solid without any intervention of any kind for several years.
NO just means normally open, when the board is powered off or the relay is not activated there will be no continuity between NO and common. When the relay is activated there will be contact between the NO and Common.
I am not sure of your wiring so I am going off of the relay contacts and their labels.
That is precisely our situation as I described. The problem is that when commanded to open, the relay stays closed. This has been proven to be the case with both relays.
To be more specific: before the problem first occurred, when the storage tank’s Particle Photon sent a “1” to ask for more water (as a functon of a float switch), the Particle Boron at the well, having received the “1”, sent a command to the relay associated with the pump to close and the pump started. When the tank became full, the Photon would send a “0” and the Boron would send a command to the relay to open and the pump would shut off. This all worked perfectly for years.
Repeating my initial description: we verified, using a multimeter, that after a “0” was received, the relay would not open (even though the relay status LED on the NCD board indicated that the command had been sent); there was continuity across the NO and COM terminals. The LED has always correcty indicated whether the relay was supposed to be closed (on) or open (off). The fact that the LED was off but the relay was still closed rather than open certainly implies that there is no direct “hard wired” link between the LED and the relay.
I hope that Travis might have some insight to offer.