How do I connect PTs and CTs for 400A 3ph service

On Sat, 14 Feb 2004 10:17:08 -0500 BFoelsch wrote: | If you are truly worried about it, you disconnect the load from the shorting | bar, check it with a meter, reconnect it and remove the short!

If I'm going to do that, I might as well de-energize the whole thing, and remove the short before re-energizing.

| CTs and shorting bars are used safely everyday, everywhere. A CT is NOT | going to set fire to the curtains in a millisecond, it just can't deliver | that much power. It can produce hazardously high voltages, but not at high | currents. Take a look at the saturation curves of some CTs for a real | education. In practice, you start to loosen the shorting strap, and if you | see or hear arcing, you tighten it back up!

Something that can't deliver more than 1 ma at expected current draw might be safer to handle. Fire isn't my first worry, though it is certainly a valid one.

| For your application you should perhaps investigate current transducers | rather than CTs. Small, Hall effect devices that typically have better | frequency response than CTs and eliminate this whole (non)issue of secondary | voltage.

Any particular good keywords to google for to find these?

| I find it kind of amazing that you are designing a product like this from | scratch; there are any number of UL listed assemblies available with all | kinds of network connectivity. Check out the offerings by Allen-Bradley | (PowerMonitor)and Square D (PowerLogic). I would bet that either of these | would do everything you need and more.

I've seen some around, but none have a usable output interface. The output but be optically isolated no matter what method of measurement is used, and a standard and/or documented data path and format (no proprietary software). One I did look at simply provided minute by minute watt/hour usage, which is way off from what I'm doing. Another seemed like it might have more data being handled, but required the use of their software (for Windows, which immediately rules it out), which negates the flexibility of designing my own logic.

|> | For your application you should perhaps investigate current transducers |> | rather than CTs. Small, Hall effect devices that typically have better |> | frequency response than CTs and eliminate this whole (non)issue of | secondary |> | voltage. |>

|> Any particular good keywords to google for to find these? | |

It appears they only sell the raw sensors in large bulk. Maybe Digikey has them in single quantities.

It might be helpful if you told us what you are trying to do, not how you want to do it. So far you have not mentioned what you want to accomplish. You have posted how you plan to do it (isolated input to a computer, sampling rate, etc).

If we know what you are trying to accomplish we may know of instruments that can provide what you need without the need for all of the custom work that you are proposing.

If it has to do with harmonics monitoring and or harmonics compliance, there are many products available.

Just a thought.

Charles Perry P.E.

| It might be helpful if you told us what you are trying to do, not how you | want to do it. So far you have not mentioned what you want to accomplish. | You have posted how you plan to do it (isolated input to a computer, | sampling rate, etc). | | If we know what you are trying to accomplish we may know of instruments that | can provide what you need without the need for all of the custom work that | you are proposing. | | If it has to do with harmonics monitoring and or harmonics compliance, there | are many products available.

I want to get the voltage and current waveforms into a Linux based computer for the purpose of processing it in ways that I want to be able to program. The software I will write will be analyzing the data for things that I want to be able to customize without having to know them all today. At first it may be to simply measure load usage. But then I will add on other analysis, including harmonic, reactive, noise, RF, BPL, and then perform logging and remote control of devices.

My objective isn't to do a specific analysis, but rather, it is to make this data an input so I can explore what analysis and controls I can do. Once the data is in the computer in real time, then it has moved from a realm I am not an expert in (hardware) to one that I am an expert in (software).

I want to keep the hardware minimal, though I also want to protect computer hardware from added risk of any new connection to power. The computer _may_ even run entirely isolated, using a battery, or if not that certainly a UPS.

I'm not interesting in buying an instrument that does the analysis for me, nor am I interested in using someone else's software that does the analysis (although if it an open source version, I may be curious to look at it to get ideas).

Right now my first thought is to feed an analog waveform into a sound card and explore from that. Eventually I see doing multiple channels of data in digital form (still to be decided), and multiple channels of device control.

One long term objective is to manage load from selective devices based on the level of other uncontrolled loads. If other loads are low, more of the managed devices can be on at the same time. If other loads are high, fewer of the managed devices can be on at the same time. That objective would be to maintain a controlled peak load, especially when the power source is not from the power company.

But that's not the only objective; it's just an example. I won't even be able to envision all the possibilities until I get started and see how much data I really get and what I can analyze from it.

So basically, the first objective is to just get the data into the computer.