Detecting AC current with a micro controller

A "clamp" amperemeter around one (or more) of the the power supply lines. That's a device measuring the current going through a ring by the magnetic field it generates.

The "or more" could be a problem. If it's a single-phase device then the current transformer will output a voltage proportional to the sum of all the currents going through the toroid. Unless there's a serious problem with the machine (a ground fault) the current going in equals the current coming out, so the net is zero output voltage!

A current transformer can be used for this task. It will provide isolation. A clamp ammeter is just a current transformer and meter with a core that can be opened to put around the wire. Be sure to put a load resistor on the current transformer output or the voltage can get undesirably high. A Rogowski coil is another possiblility.

Best regards, Spehro Pefhany

Current transformer ot optocoupler with LED with current-limited circuit and phototransistor.

You could convert high voltage signal to low with resistors or galvanicaly isolate it from microcontroller and supply that proportional low voltage signal to the microcontroller pin and sample that pin number of times per minute, and if counted number of samples high were > X then you have power on that line othervise no power.

see

by

Use a hall current sensor from Allegro or LEM or some such. You'll get a voltage output (0...5 V, for example) and almost any current range (up to kA's). The "downside" is that these sensors are too accurate for you, but I hope you don't mind.. :-)

-jm

A company called Solid State Control, SSACI, makes a device that will do want you want. Yuoe set a trip point and a delay time and you have a relay contact output.

Check out :

Click on the ECS Series link.

MR ?????

-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----

-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

Hi One should note that the peak current my not be in phase with the voltage. Using things like a zero-cross detector to synchronize the processor to the line my not give a clear current picture if the load has a lot of reactance. You might need to hunt for the peak. Dwight

Some 10 years ago I did a self-reporting power meter for a municipal utility. This task is not as easy as it seems.

Not only is there a power factor to confuse but loads aren't necessarily even a sine function! Some power controllers use SCR's which chip the sine into a varying number of degrees, other loads like arc welders have a huge initial current spike, etc.

If you want to actually measure power then you'll have to integrate the actual signals before digitization. Some of those spikes can be in the nanosecond and microsecond range, if you can relegate integration to hardware your computational horsepower requirements become much less severe.

S> Hi, I have an application that among other things has to detect the

-- Regards, Albert

---------------------------------------------------------------------- AM Research, Inc. The Embedded Systems Experts

----------------------------------------------------------------------

Hi, one of those little pics will do this easily.All you need is a current transformer/bridege rectifier/burden resistor to provide the signal the ad input of the pic. Then sample continously, store/add/ average enough samples for your line frequency, even TRMS is not hard to do, compare the result to the reference value and drive your output bit.