When deciding what size CT is needed for a particular setup, can it be sized for the normal max load of a circuit instead of the max breaker limit?
For instance I have a panel for my pool that is fed from a double 60A breaker, so 120A max load. The loads currently in this panel are the pool pump (max 17A), a 15A breaker for the controller, and a 20A breaker for a couple outlets. Max draw should be ~50A, but realistically closer to ~20A under normal conditions. I would think a 50A CT would be OK, a 100A CT would be plenty of headroom.
My concern is the panel could have new breakers added to use the full 60A per leg for a total of 120A in the future. Since the CT would be in a different panel than the breakers, if someone didn’t know about the CT in the upstream breaker they could damage something.
Assuming that when the voltage of CT reaches the TVS diode level, the usage graphs will flat line at the top and you will not have a correct power measurement anymore, I just wanted to make sure this use case would not damage anything in the short term. How long can it take an overcurrent condition without damage, and how much overcurrent potential is OK?
I can add thresholds in my monitoring software so if a CT gets to say 90% of its rating, it sends me an alert so I would know I need to address the issue short term and upgrade the CT long term.
Likely if I ever move out of this house I would remove most of my home automation and monitoring equipment anyway so these scenarios are unlikely for the future homeowner. I just wanted to understand any potential future issues as I install this.
Breakers are supposed to be sized for 125% of the expected load. Your 60A breakers are expected to have a maximum working current of 48A. The 50A CTs should be fine.
That said, if you are still nervous, seems to me simply using 100A CTs is easier than programming a home automation watchdog and also a reliable passive solution.
I am just trying to understand what happens to your electrical circuit in an overcurrent condition. I have found posts on the forum that say some overcurrent will not hurt the system, but depending on the design that can mean a lot of different things.
I would assume since you have TVS diodes to clamp the voltage, the weak link in an over current situation would be the clamps themselves as you would need a lot of current to overcome the TVS diode and damage the ADC behind them, is that correct?
As far as I know, of the many thousands of units worldwide, nobody has cooked the ADCs. The TVS diodes in the CTs are there to protect the CTs. Without them, if disconnected from the IoTaWatt while clamped to a current source, they can develop very high voltages internally and damage themselves.
The IoTaWatt has its own protection. First in line are the burden resistors. They are 20 ohm 1/8 Watt. 1/8 Watt would be 158mA input - about three times the 50mA design limit. Somewhere above that they would start to heat up.
At 158mA, they would develop about 3.16 volts rms. That would be +/- 4.5V. Sitting on the 1.65V bias that would be +6.15V/-2.85. There are rail-rail protection diodes on that signal that would limit those voltages to within the absolute maximums for the ADCs.
Which is probably why nobody seems to have had any problem despite a variety of mismatched CTs, both current and voltage type, having been deployed by adventurous users. The end result has always been just inaccurate measurements.
Sounds like since the only measurement I need to be really accurate is the 200A clamps on the mains so I can make sure my usage matches up with my bill, and that should not go out of spec because of the 200A breakers, I should be fine there.
My other circuits are diagnostic to see where my main power usages are, so if someone added more power to the downstream panel later and went above the current limit of the clamp, the worst that should happen is that measurement would no longer be correct, but it wouldn’t damage anything.