I have a capilar thermostatic switch which turns on Grundfos Alpha2 circulating pump in heating circuit. I need to get info if this pump has power. Maybe it will be better to buy flow meter and measure water flow…
A word of warning - even if the pump is powered, it might be stalled, displaying two dashes on the LCD. I recently experienced this situation when the room temperature was decreasing because of the stalled pump.
Better use a flow meter or optical sensor attached to the LCD. Grundfos has a product called “ALPHA Reader” but I think it works with ALPHA3 only.
Hi stevemann,
can you share scheme and coding how you’ve done it using nodemcu? I would like to identify current for safety critical consumers such as oven, heating table etc for alarming if it is ON without human attendance in the kitchen.
Have you built these PCB’s yet, or have you put the circuit on a breadboard? I built this on a breadboard yesterday and am not having very much success.
The best signal change I can get to the output is from ~2.3V when mains is OFF to about 1.8V when mains power is ON. If I remove the sensor wire entirely, the output does pull up to the 3.3V supply. But when I have the sensor wire connected (without 10k resistor present in series) the output changes to ~2.3V.
Watching Raplh’s video he is clearly having much better success. But looking at his video when he is showing it functioning, it appears there is a ceramic cap between the sensor wire and the base of the first transistor?
Sorry, that was a tilde, I meant “around 2.3V”. I played around with it more today, and it’s very finicky. I was able to get a more stable response and get down to about 0.35V when detected (and obviously 3.3v when not detected) by changing the 4.7k resistor in the charge pump to around 80k and also changing my power source to a battery. I’m using a breadboard power supply which converts 9V down to 3.3v but originally I was using a 9v wall wart and now I’m using a 9V battery adapted to a dc barrel plug.
The change to 80k seems to have made the circuit shift far enough, but what’s really confusing is that the circuit reverses behavior depending on if I’m using the battery versus the wall wart. With the battery I get 0.35v= detected, 3.3v = not detected, but with the wall wart I get the opposite. I assume there is some voltage ripple on the rail when powered by the wall wart and that is causing some issues but it is still baffling. I don’t have an o-scope but I do have a good multimeter (fluke 87v) and using the min/max functions I’m not seeing much for spikes in the 3.3v rail with either supply method.
Anyway, if you end up getting these boards and testing them out I’m interested to hear your results.
Another note is I’m trying to detect 110VAC instead of 230VAC.
The output is supposed to be low true.
The weird response could be from the wall-wart. They are generally made for charging batteries and you’re lucky if there is much smoothing in them. Try adding a 20uf capacitor to the 3V3 line.
In my opinion the simpliest and cheapest way now is to buy one of those cheap china or IKEA etc. Socket switches eith power meter. This way you can switch on/off any device and also check if this device is running because if YES, there will be power cinsumption. This way you can secure some important system devices like boiler circulation pump etc.
I mean you can create script that if the device is ON and the actual power measured is lower than XX Watt it will create ALARM.
