Current sensing switch

[jmaja]

I'm trying to control a borehole pump using a current switch. I have a ground source heat pump with an external heat exchanger with ground water as a heat  source. Unfortunately there is no output from the heat pump telling when the compressor is running and thus needing ground water.

The compressor is internally VFD driven and I have a VFD driving the borehole pump. The latter VFD can except a on/off input e.g. as a 24 V/0 V and can source 24 V for that.

I thought it would be easy to put a clamp type current sensing to one of the three phases (400 V, 50 Hz) going to compressor VFD, but at least the first try didn't work.

First I used a clamp on True RMS multimeter and it showed less than 0.1 A when compressor was not running and about 1 A when it was running. The compressor takes about 0.9 kW at minimum load and can take up to about 4 kW.

Then I bought this (NPN-version) https://www.crmagnetics.com/switches/ac-current/ac-voltage/cr9321
and put one phase going to VFD through it. For testing I used a lab power set to 24 V and 0.06 A. With heat pump power, but compressor not running, the lab power showed 24 V and 0 A. When the compressor started it seems to take a while (10 s?) before the lab power started to show 2-3 V and 0.06 A. Later it went to 0.8 V and 0.06 A. But when the compressor stopped, it still showed 0.8 V and 0.06A. Even shutting the power input totally to the heat source pump didn't bring back 24 V.

What am I doing wrong?

I can't quite understand the specs of CR9321. I tried to ask them, but didn't get a helpful answer. I really can't understand what could be the difference between NPN and PNP models. Both have just two wires. What would it matter if the load is on negative or positive side? The CR9321 is floating and can only see the voltage and current between its wires.

Specs say that 0.35 A s rated full-on. It doesn't say anything about full-off. But having power switched off (from the electric cabinet of the house), should certainly qualify.

Is there an another current sensing switch known to work for this application?

Other option would be to use the 230 V used to power the circulation pump inside the heat pump. But that pump starts every now and then when the compressor is not running. Thus it would not be optimal.

[moffy]

The CR9321 has a holding current 0f 0.01A, which means that if it turns on, then the current has to drop below 0.01A before it will turn off again. Since the VFD never drops below 0.1A, it never turns off.
P.S. You should  use say a 4.7k resistor connected to the PSU instead of just current limiting the output of the PSU, there might be output capacitors that are causing the turn on delay.

[jmaja]

The transistors switch a current a DC loop. You get to choose whether the current flow is from - to + or vice versa. If you choose an SSR, check data sheet for max control input current. (Attachment Link)

I'm not planning to use a relay. Just 24 V or not to VFD digital input. Now I just have a wire from 24 V to that input. If I remove the wire, the pump stops. It's pulled low.

I still don't understand at all what's the difference between NPN and PNP models. Wouldn't NPN work in the left picture just the same, if it was put so that red is where PNP has orange?

Isn't relays "+" in the wrong corner in the right picture?

[Terry Bites]

I see loads of these cheapo relays on the market. They claim to have a threshold starting at 200-500mA. https://www.aliexpress.com/item/32968582836.html

[jmaja]

The CR9321 has a holding current 0f 0.01A, which means that if it turns on, then the current has to drop below 0.01A before it will turn off again. Since the VFD never drops below 0.1A, it never turns off.
P.S. You should  use say a 4.7k resistor connected to the PSU instead of just current limiting the output of the PSU, there might be output capacitors that are causing the turn on delay.

Where does it say 0.01 A holding current? I can only see it for the ACA-model, for which it says "Minimum holding current: 10 mA".  I have NPN-model. Also I had the power totally disconneted for several minutes and it still showed 0.06 A/0.8 V. I don't know how low the current to VFD goes when the compressor is not running. The clamp ammeter I used has a resolution of 0.1 A.

I have another CR9321. I'll do some simple tests with that first.

[jmaja]

I did some testing with the other CR9321. Now I used only 10 V from the lab power and had 200 ohm resistance in series. All the tests were carried on with 230 V one phase load. It worked as expected with loads tested. It didn't turn on from combined LED work light (10 W), soldering iron (25 W) and mobile phone charger (15 W). But it did turn on from a heater (650 W) and it turned off when the heater was turned off even with all the other loads still on. And there was no delay turning off nor on.

The CR9321 installed to heat pump seems to be broken. It has very low resistance in both direction. Thus it seems it has been damaged. But what has caused the damage? Could the lab power do that with current limiting? CR9321 is rated for 30 V. The cable is a few meters. Could that cause the voltage to go high enough and destroy transistor/FET or whatever there is inside CR9321.

[moffy]

Sorry I didn't pick up on the distinction between the -ACA and the -NPN.
It is possible the lab power supply with current limiting could have damaged the current switch. Normally there is lots of capacitance on the output of a power supply and the current limiting will tend to switch on after it has discharged. That initial discharge could be sufficient to kill the -NPN switch.

[jmaja]

The other CR9321 seems to work OK now. It's connected directly from VFD 24 V output to its digital input, which is said to have about 4 kohm input resistance. Works just as expected: the borehole pump starts and stops just as the compressor does.

[moffy]

Very pleased to hear that.