Looking for ways to measure AC line voltage with a uP.

There has to be a bazzilion. How accurate do you need to be? Low cost? Size? Weight?

One basic method would be to get a reduced-size version of the waveform, then either rectify it and sample it with an ADC, run it through a true RMS chip and sample it with an ADC, or sample it fast enough with an ADC that you can calculate RMS (or whatever) in the microprocessor.

Another would be to rectify the high-level voltage, then drop the resulting DC voltage down and measure it.

You'll need to be isolated, so you either need to have the microprocessor running on the "hot" side and isolate it's communications from the PC, or you need to isolate the AC while getting an accurate copy. The former technique would call for opto- or other isolators between the microprocessor and the PC, it would make it easy to get an accurate measurement, but it would make it harder to work on the circuit. The latter technique would keep working on the circuit easy, but would make accuracy expensive.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google?  See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html

Thanks Tim,

I would like to measure the AC line on a remote heater and water pump.

I can run a RS-485 to the site, but measuring the AC side is where I am having to be creative.

My idea is to isolate the line measuring side from the uP.

What do you mean by "make accuracy expensive" ?

This is a one off project, for our hot tub. :-)

donald

LEM make modules useful for that. Hookup measured voltage on one side get an isolated signal out the other. They do need +/-15V.

They are known primarily for their current sensors though.

Robert

It's poor grammar for "accurate transformers cost a lot".

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google?  See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html

Buy a Fluke with RS-232 interface and don't play the fool.

VLV

Why would the voltage fluctuate at the remote site (long cables. heater and pump in series etc.) ?

A simple and safe solution would be to use any safe transformer, rectify and filter it and load with a resistor. This low voltage and low impedance signal can be transferred a _long_ way (with perhaps plenty of low pass filtering at the receiving end), before passing on to an ADC.

Paul

?????

Transformation is controlled by the turns ratio. It is presumed the turns are counted as they are added. What is there to be inaccurate?

Capacitance, leakage inductance, and core saturation to name three. If all transformers were infinitely accurate, you wouldn't have power transformers and instrumentation transformers.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google?  See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html

If you are measuring voltage, you aren't loading the transformer. No saturation problems. At 60 Hz you need a lot of capacitance to affect the readings. I don't think leakage inductance is an important factor either, but I can't quantify it.

A bigger factor would be diode drops, assuming the result is rectified and measured as a DC value.

--
Chuck F (cbfalconer at maineline dot net)
   Available for consulting/temporary embedded and systems.

"Donald" a écrit dans le message de news: gbednYjt7vdBtPrYnZ2dnUVZ snipped-for-privacy@comcast.com...

Have a look at

in the applications notes for the PSoC's devices. As far as I remember there is a design for a PSoC-based wattmeter.

Friendly, Robert

Since when has the primary taken no current when the secondary is unloaded? There's lots of current but hopefully at 90 deg phase. That current is still capable of saturating the material. Of course that would mean you're using the transformer above its voltage rating.

Well don't do that then...

Peter

Since we are applying AC, not DC.

--
Chuck F (cbfalconer at maineline dot net)
   Available for consulting/temporary embedded and systems.

I was talking about AC - nothing to do with DC. If we apply AC volts to an inductor L, such as the primary of a transformer with unloaded secondary, there will be a current V/(2*pi*f*L). The current can product a magnetic field in the core of the transformer outside the linear region if the coltage is large or the frequency low.

Peter

Do you mean the transient when an earlier load is disconnected or when no load has never been connected?

Are you perhaps referring to a current transformer? Else, the preceding is garbled.

It's done all the time. It's just a matter of calibration.

Analog Devices make some Energy meaurement devices suitable for measuring various energy related parameters of mains supplies. Some of the devices has got serial interfaces with which to interface to a uP.

Regards Anton Erasmus