Automation help - trigger when value stays at a certain value for too long

And to add my bit to this thread…

In this case since the suction of the pump is clogged (which wasn’t mentioned at first but only that the pump was “stuck”) a pressure sensor might work depending on exactly where in the system you installed it. For example, if it’s installed too far away from the discharge of the pump there wouldn’t be much pressure at that point if the pump was operating correctly or not since this is an open system.

A better solution would be to use an actual flow sensor.

A flow sensor would always work if any of the probable modes of failure of the pump has occurred - clogged discharge or suction, locked pump impeller, a sheared pump shaft, a loss of power to the pump, etc. We don’t care about the pressure, per se, only that there is adequate flow.

But the OP said they didn’t want to add a new sensor so the point is moot in this case but I wanted to throw this out there in case someone in the future sees this thread and is looking for similar ideas.

1 Like

I already added the tube to the inlet.

To make sure the tube can’t suck on the poolwall again, I drilled some holes in the attached tube.
This way this specific issue should be gone.

@finity Hopefully you are right and this thread will be useful for someone in future, so that it isn’t completely worthless.

Install it near the pump’s outlet (too far away qualifies as ‘bad installation’). My pool’s pressure sensor is located in the valve head, about 1 meter downstream from the pump (1.5 HP Hayward). Anything other than nominal pressure correlates to all the failure modes you listed.

The moral of this story is not to seek complex solutions to simple problems. Detect a blocked pump inlet by monitoring minute temperature differences? Yeah, no.

The greatest irony is that it took more effort to explain the proposed solution’s flaws than to simply fix the inlet pipe.

Except for a blocked discharge pipe. Many times in a centrifugal pump you might not notice much of a pressure rise right at the pump discharge if the pump is dead headed, especially if you’re not exactly sure what “nominal” is supposed to be. In that particular case, a “bad installation” actually helps you see the problem since then the whole line will be equal to pump discharge pressure.

Anyway, one way or another the thread was “interesting”. :slightly_smiling_face:

?
When opening the pool for the summer, just adding the “eyeball” jets to the three returns causes a 2 PSI increase in back-pressure. Pretty sure blocking all three returns would cause more than that.

I could block two of the three returns and tell you precisely how much the pressure increases but I’m not going to risk breaching seals within the valve head just to prove the point.

As for the value of nominal pressure, it depends on the valve setting. Any setting that allows the water to bypass the filter (such as recirculate or waste) will result in a lower working pressure. A drop in pressure, while operating in filter mode, typically means the pump is sucking air because the pool level has dropped (below the intake). Similarly, a failure to develop nominal pressure shortly after startup is a potential sign that the pump can’t develop sufficient vacuum to draw water (intake may be clogged or the pump simply needs priming). A pressure sensor can be used to provide insight into many failure modes.

When I noticed the issue, I already thought about the tube solution (to be honest).

But when I saw the diagram I wanted to be able to detect the obvious difference.

For me it was clear that I won’t need the automation a long time. It was more about how to solve that issue with HA.

What I learned is to always keep in mind that the sensor data isn’t exact (noise).
I already knew that but I did not account that.

I found it also interesting to see different solutions.