@overeasy, I realize the following:
But I still feel due to cost, size and low current accuracy, there must be a solid core CT that are a good fit for 15A and 20A circuits.
I’m located in the US and willing to send you some CT to test that you feel have good potential.
Thanks,
Automate
I’ll probably never be able to get the message across that there is no practical advantage to using a 15A or 20A maximum CT over a 50A or even a 100A CT. While that may be true with some of the low sample rate monitors out there, it is just not a factor with IoTaWatt that uses much higher sample rates and 12 bit ADCs. The best inexpensive solid core CTs that I have seen are around 50A maximum. Those are the HWCT-004 and DL8CT2 (I believe those are both made by the same OEM manufacturer).
In the context of IoTaWatt and other monitors with burden resistors in the 20 Ohm range. Those CTs are made for 10 Ohm burden. I’m working with a 10 Ohm burden option in the ESP32 prototype. Echun has a slightly better 25A small split core that I will be testing with that burden. It was not acceptable at 20 Ohm burden.
The offer stands. Understand that I will keep the two samples of each as there may be a need to retest as parameters like burden resistor change.
Since it is AC, do you need a pair of TVS diodes, one going each way? Is there a through-hole style part number you can recommend?
Good question. You can get bi-directional zener diodes. I have a strip of them that I bought a long time ago, so don’t know the part number, but looking up using Digikey parametric search there are a few dozen thru-hole parts. I’d probably go for this one as it’s the least costly:
@overeasy Doing the math on the AZ-0500 with a 500:1 turns ratio on a 20 amp circuit
20 A / 500 => 0.04 CT amps RMS
0.04 amps * 20 ohms => 0.8 volts RMS
0.8 volts RMS * 1.4142 => 1.1314 volts peak
I thought I remember seeing your inputs can handle up to 1 volt max. Would this be a problem?
Short answer is no, 40mA would not be a problem.
The best way to look at the IoTaWatt inputs is the rated current rather than voltage. There are various versions, all are designed to accept 50mA output current transformers. On the V5 unit, the burden is 20 Ohms, so 50mA RMS yields 1V RMS across the burden. The ADC voltage reference is 3.3V. 1V RMS is 2.82V p-p, so centered with the bias that would be 0.24V - 3.06V max, so it works fine.
The reason I stress using the input current is that I have versions of IoTaWatt that use 10 Ohm burden resistors, which is the value that many CTs are designed to. One of the problems with low end, low capacity CTs is that they don’t produce the power needed to develop the needed voltage across higher burden resistors. Case in point is the SCT006, which saturates well below it’s rated capacity into a 20 or 24 burden.
Even CTs that have the ability to drive the 20 Ohm burden will exhibit lower and more consistent phase-shift with a 10 Ohm burden. So I’m moving in that direction, but as long as it’s designed for 50mA CTs, all CTs that work with the V5 will continue to work the same in a newer 10 Ohm model (maybe using different phase shift ladders).
Got some of the AZ-0500 (25A) in today. You can see their size compared to the 50A split core AcuCT-H040-50. I’ll be sending 2 of them to @overeasy
