Sensors greater than 200A


#1

Hi. I really appreciate your work it is very good job, congratulations.
Can you help me with the next questions:
1.- Is it possible to add current sensor greater than 200A to IoTaWatt?.
2.- What changes needs to be done in order to it works?.


#2

Yes.

The maximum rms current from the CT with the 24ohm burden resistors is 50ma. So if the CT output would exceed that at your maximum current, the burden resistor would need to be replaced with a lower value. The burden resistors are 0805SMT resistors. If you resign yourself to doing that, you should be able to find some CTs that will do the job.

In looking for a higher capacity CT, the prime consideration should probably be large enough opening for the cable. If you can tell me the diameter of the cable and the maximum current you want to measure, I might be able to recommend one.

You will also need to configure the CT as generic, but that’s just a few keystrokes.


#3

Thank you for your soon answer overeasy.
When you choose in the IotaWatt config page, different current sensor for example SCT-013-000 and SCT-019-000, they have different output current, 50mA and 33mA respectively.

  • In config page, when you choose SCT-019-000 model, the value of burden resistor doesn´t change. Does IotaWatt make some adjust in firrmware in order to get the right measurement?.

  • When you choose a generic current sensor, for example SCT024TS, it is 400A input, 100mA output. Is it needed to change phisically burden resistor and to modify the firmware in order the IoTaWatt can read the right measurement?.

  • I would like to measure the power consumption in my work, but the cable diameter is big, it is 4/0AWG and they use around 8 for each phase. I was planning to use a flexible current sensor (Rogowski coil), this device delivers an output voltage around 333mV for 600A. Whats your opinion? is it hard to implement in IoTaWatt?.


#4

Yes it does. The “turns” ratio as well as the phase shift of each CT type is defined in the TABLES.TXT file, and IoTaWatt uses that information to determine the correct calibration factor as well as to correct the phase difference between the VT and CT.

You would need to adjust the burden resistance to handle input current greater than 50ma. For the SCT024TS that you describe, you would need to make the burden 12ohms instead of 24ohms to be able to use the CT above 200A. You can do this by replacing the onboard burden, or by adding another 24 ohm burden externally so that the net burden is 12ohms. After you adjust the burden, you would specify the new value in the SETUP->DEVICE->SPECIFY BURDEN RESISTORS config menu and specify the CT as “turns” 4000. IoTaWatt will calculate the correct calibration factor.

I’ve not used a Rogowski coil, but I think it could work somewhat. Bear in mind that the IoTaWatt input channel ADCs are 3.3V and designed to read 1VAC RMS at full scale, so a 333mv device will have less resolution than a 1V output device. That said, I think it will give decent results. A voltage input would require removing the internal burden resistor on that channel, and the cal would be 1800 (1800A with 1V input). If you post a link to some documentation on the Rogowski coil, I could probably give a better response.

I think it’s also possible to put individual CTs on each cable and connect them in parallel. My understanding is that the output current from each adds so if you had 4 200A:50ma CTs, the output would potentially be 200ma, so the burden would need to be reduced to 6 ohms. I’m no expert on this end of it, but there’s a lot of information out there for dealing with high power. I’ve also seen smaller CTs clamped on the outputs of larger 5A output CTs to get an indirect reading. It’s best to get someone trained in this area to design the CT.

I don’t know how times have changed in Mexico. I worked in an office in Mexico City 45 years ago, and watched an electrician changing a fluorescent ballast. His idea of a voltage meter was to have his muchacho touch the wires and nod if they were live. I gave him a small light bulb with a couple of wires soldered on to use instead. He was unimpressed, but his muchacho was very grateful.


#5

Thank you overeasy.

1.- File TABLES.TXT only can be written by the IoTaWatt config page?.

2.- I have not found, valuable information about Rogowski coil, only commercial one. Here is a supplier and the part number that I think could be used.

3.- As far as I know, if we are monitoring a sinusoidal current signal using Rogowski coil, at the output of integrator we have a rectangular pulse in the half cicle and another rectangular pulse in the other half of cicle. We have separate pulses, does this affect the phase detection in IoTaWatt?.

4.- The electronic industry in our country has developed enormously.
Some companies that are here: Intel, HP, IBM, Continental, Flextronics, Tata and many more. More than 80% of its employees are Mexican. It is not only maquila, it is also made electronic design. I’m glad you already know Mexico. I recommend you visit Jalisco.


#6

The tables file is distributed with the firmware release. It describes all of the CT and VT that can be configured by menu pick. You can also specify a generic device and provide the specs yourself. The tables file is an input table. When you configure your IoTaWatt, a config file is updated with your changes. It is a json file that can be edited manually, but that is not recommended. The configuration utility that runs in a browser is used to edit the configuration. You can get an idea of how that works with the somewhat limited demo version at iotawatt.com.

I looked at the rogowski coils. In particular I was looking at this unit from Magnalab:

It’s pricey ($400) but is actually three coils for a three phase install like yours. When I like about it is that you can select the current range, and specify 1V output which is full scale for IoTaWatt.

What I’m less clear about is exactly what the output looks like and how much calibration is required. IoTaWatt requires a representation of the current waveform. I think that’s what the integrator does, but I’d like to see a graphical example that shows voltage, current, and integrator output.

I didn’t see anything about a rectangular pulse, unless you are talking about the direct output of the coil, which is the derivative of current and looks somewhat like a square wave. Mathematically the integral should produce the original wave, but I’m not certain of that.

I have no doubt that technology has progressed in Mexico. When I was there I worked for Bufete Industrial who were engaged in building a number of nuclear power reactors outside of Mexico City. I was not directly involved in that, but supported the effort by programming their IBM 360 mainframe running.