1. Home
  2. Electronics Design
  3. Current measurement

Current measurement

It's been quite a few years since my last physics class, but here I am trying to measure a current without putting anything in the circuit being measured. I'm talking about measuring 60Hz AC currents in the range of 0-20A or so, preferrably with close to +/-1% accuracy through the whole range. I ultimately need to get the signal into a DC voltage to feed into an ADC and then to a microcontroller (which can perform whatever calculations are necessary).

Last time I messed with this sort of thing (many years ago), you used a simple coil and passed the current-carrying wire through the middle of it. If I remember correctly, this induces a current in the coil proportional to the current passing through the wire and the number of turns in the coil, right? I don't remember the exact formulas... perhaps it is actually also dependent on the rate of change of the current, so that measuring AC is different than DC. I'm guessing that an AC primary current generates AC current in the coil as well, so I'll need to figure out how to turn that into a DC voltage signal. (I'm thinking maybe just run it across a resistor with one leg pinned to +Vcc/2, then I get a zero reading at +Vcc/2, with negative currents below that and positive above? Need to choose the resistor value right to get the right range, of course.)

But browsing through catalogs these days, I mostly see Hall Effect sensors, which is something I vaguely recall being new, nifty, and expensive a few years ago. They seem to be down into the $20+ range now, but that still seems expensive for what I'm trying to do. Then again, looking at CTs (which seem to be what I described above), those are in the $10+ range -- why so expensive? It's just a coil of wire in a plastic housing, right? Am I missing something?

Also, is there any reason to use a Hall Effect for the sort of application I described? Perhaps the output will be easier to condition for feeding into the ADC? Or maybe the concern is that the coil would have an effect on the circuit being measured -- probably adds an impedance some way or another (boy, I wish I remembered something from physics).

Thanks for any advice.

--
Randall
Reply to
Randall Nortman
Loading thread data ...

Accuracy. Most commercial units are fairly accurate devices. Your requirement would not be easy to meet without reasonable accuracy. Most also include sensor elements in the price. Also remember the isolation requirements and testing if dealing with mains.

Yes. The Hall effect device is easier to condition, and can measure DC as well as AC. Consider a DIY approach. Linear Hall effect sensors are relatively cheap (about $1). Run a wire across one, and read the current from this. You will need to add a precision AC-DC converter, and either an RMS, or peak measurement circuit, depending on which measurement you require. The funny thing is that once you have done this, the price will be similar to the Hall effect assemblies....

Best Wishes

Reply to
Roger Hamlett

I'd use the linear hall sensor, add a ~ 240hz filter on its outpu then feed that to an A2D. Use the micro to sample to the positive negative peeks of the signal and give you an RMS reading based on the assumption that it is a true sine wave or actually calculate the RMS within the micro for a TRUE rms reading independant of the current waveform.

PIC microcontrollers are more than capabile of doing this and even come with internal A2Ds and clocks for $2 - $4. they can drive either LCD display or serial export the data to another device of computer serial port.

doesn't get much cheaper than that.

Reply to
Mook Johnson

It is impossible to measure anything without the measuring instrument or method having some effect on what is being measured.

It is possible, of course, if you understand what you are doing, to apply a correction to the measured value to estimate what the true value actually is when the measurement is not being made.

But it is always an estimate.

--
Reg.
Reply to
Reg Edwards

On Sat, 10 Sep 2005 17:09:33 +0000 (UTC), "Reg Edwards" wroth:

That statement in not entirely true.

One example of a measurement that has no effect on what is being measured, is counting. A pulse counter does not change the total number of pulses a device will deliver when it measures (counts) the pulses.

I'm sure a long list of measurements that do not have to have corrections applied can be generated.

Jim

Reply to
jmeyer

I read in sci.electronics.design that Randall Nortman wrote (in ) about 'Current measurement', on Sat, 10 Sep 2005:

Google for 'clamp ammeter'. For AC-only measurements, they are just a sort of transformer. The commercial ones tend to have a sensitivity of 1 mV/A, so you need to amplify quite a lot before trying to rectify, and a precision active rectifier is required. For measurements down to DC (but often only up to mid-audio frequencies like 400 Hz), a Hall effect sensor is used. Still 1 mV/A.

--
Regards, John Woodgate, OOO - Own Opinions Only.
If everything has been designed, a god designed evolution by natural selection.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk
Reply to
John Woodgate

instrument

number of

===========================

It is the process of obtaining the count, whatever it is, which causes an error. You are suffering from delusions of accuracy as are most engineers and so-called scientists.

--
Reg.
Reply to
Reg Edwards

...at best.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
Reply to
Spehro Pefhany

Just counted my fingers, 10, not 9.9999999 or 10.0000001, just 10.

Think you have been reading some quantummy books, John Gribbin is quite good for laymen like me. Pity Amazon doesn't show the one I mean, must be a measurement error

martin

Reply to
martin griffith

On Sat, 10 Sep 2005 21:46:31 +0000 (UTC), "Reg Edwards" wroth:

OK. Please enlighten me as to how a method to count of pulses can either increase or reduce the number of pulses generated as compared to the number of generated pulses when the pulses weren't counted.

Jim

Reply to
jmeyer

Hall effect sensor - cat # SS-19 $1.00 for 2 Current transformer - cat # VT-5 and cat # VT-27,

75 cents and $1.25 respectively.
formatting link

The Hall effect sensor produces a DC voltage proportional to the magnetic field, so its output is more suitable for your needs.

Ed

Reply to
ehsjr

to the

===============================

For example, a noise spike can trip the counter.

There is ALWAYS noise present. It is a matter of probability whether or not an error will occur in a given time interval.

The probability may be exceedingly small but it ALWAYS exists.

Don't delude yourselves. The measuring instrument ALWAYS interferes with the measured value. No need to go beyond the simple voltmeter to illustrate the point.

Another way of looking at it - the measuring instrument abstracts information - and information = energy.

--
Reg.
Reply to
Reg Edwards

Of course CTs don't have response at DC (hey, they are transformers after all!), but they usually have a fairly impressive wideband response and are not clunkers, like Phil implies. A CT's low-frequency response is limited by the winding inductance (which can be pretty high with high-permeability cores) and the load resistance, and the high-frequency response is set by the leakage inductance and the load impedance (including any capacitance). One usually enjoys many decades of flat CT frequency response between these extremes. Most 50/60Hz CTs work to 100s of kHz, and some even up to 20 to 50MHz, so they do perfectly well showing all the harmonics.

Unless one needs response to DC, or 6 to 8-decade frequency response, there's little benefit to Hall-effect versions.

--
 Thanks,
    - Win
Reply to
Winfield Hill

I read in sci.electronics.design that Phil Allison wrote (in ) about 'Current measurement', on Sun, 11 Sep 2005:

Are you quite sure about that? Why should it be so?

--
Regards, John Woodgate, OOO - Own Opinions Only.
If everything has been designed, a god designed evolution by natural selection.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk
Reply to
John Woodgate

Aircraft '400Hz mains' is moving to variable frequency Speff, to save the weight of the constant-speed gearbox. I've checked CTs as being within 0.1% over the frequency range 350 to 750Hz, and 100 to 375Arms.

--
Tony Williams.
Reply to
Tony Williams

I read in sci.electronics.design that Phil Allison wrote (in ) about 'Current measurement', on Sun, 11 Sep 2005:

You are very eager to pick up and contradict others statements, usually with foul-mouthed comments. Your avoidance of answering my polite questions is no doubt noted by group subscribers.

--
Regards, John Woodgate, OOO - Own Opinions Only.
If everything has been designed, a god designed evolution by natural selection.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk
Reply to
John Woodgate

yess...i *did* forget to take off my shoes to count above 10...

Reply to
Robert Baer

False. A simple current transformer is accurate over a fairly reasonable frequency range. A number of clamp-on ammeters that i have seen are good from below

50Hz to above 400Hz; Tektronics and HP have made probes good to the RF region.
Reply to
Robert Baer

"Randall Nortman"

** A Hall Effect *transducer* ( Lem, Honeywell etc ) will supply an isolated output voltage proportional the actual current flowing through the centre hole. Frequency response is from DC out to typically 100kHz with good accuracy and temp stability.

This will allow almost any imaginable mains appliance supply current waveform to be accurately measured and analysed.

A simple current transformer, OTOH, is only accurate at one frequency ( 50 or 60 Hz) and has no response to any DC component that may exist.

Depending on your application these limitations could rule it out completely - eg most electronic equipment has very non-sine AC supply current waveforms.

......... Phil

Reply to
Phil Allison

"Spehro Pefhany" "Phil Allison"

** Meaning that the iron core may saturate and spoil the accuracy for the AC component ?

Do makers of CTs ever spec the max DC component ?

......... Phil

Reply to
Phil Allison

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.