Current measurement

On Sun, 11 Sep 2005 07:03:37 +0000 (UTC), "Reg Edwards" wroth:

That would be significant if the measured property were related to energy as voltage or current would be. The original statement I questioned was one that stated that any measurement method affected the very property that was being measured and thus needed to be adjusted to be accurate.

Counting related measurements, including frequency, do NOT change the property being measured.

Jim

Except that in order to count, for example, electrical pulses, you are first measuring voltages, and counting whenever the voltage rises above a set point (or falls below the point). Measuring this voltage by necessity affects the voltages (adds a capacitance and an impedance to ground, which must be driven by the pulse source). However, if you only care about counting, then this is a threshold effect: either the pulse source has enough oomph to drive the counting circuit, in which case the effect of measuring the voltage is irrelevant, or it doesn't, in which case you won't get any counts at all (or it's borderline, which will result in unreliable counts). It is also true that interference, cosmic rays, etc. could result in unreliable counts, but that is a completely separate matter.

So you're both right, and this discussion is rather pointless.

"ehsjr"

The OP would have to find a suitable core to pass the wire through, cut the core and then bury the IC inside the gap.

Then calibrate the the beast.

.......... Phil

"John Woodgate"

........ Phil

"John Woodgate" Phil Allison

Do not expect me to reply to ANY of them !

.......... Phil

............. Phil

JW suggested the OP use the simple CT from a clamp meter, with suitable amplification and rectification.

So I thought it might be instructive to see how one of them (by Hioki ) performed over a range of frequencies either side of the nominal 50/60 Hz.

Using an audio power amp, sine wave generator and resistive dummy load the following readings were obtained with a steady 4 amps rms through the clamp ( ie 32 volts into 8 ohms) at each frequency.

With meter set to 6 amp FSD:

Freq. Reading

Not too crash hot is it ???

.............. Phil

Jeez Phil, what's with the negative ways? You seem to have a problem with what I posted.

The post identifies a site where the OP (or anyone else) could get a couple of CTs and a Hall effect sensor cheaply.

I'm sorry that's not to your liking.

Ed

Whew, pretty poor HF response. I've not encountered anything quite that bad, to my memory. I wonder if this clamp-on contains an R-C filter at ~ 180Hz? Perhaps added in a hope it'd have more accurate readings of 50 - 60Hz current-draw when used with cheap multimeters? Did this Hioki have a built-in RMS meter, perhaps that was at fault?

Passive, yes, of course. But as an CT transformer it has to have at least an internal load resistor, inside its molded assembly, and it may have more, such as a capacitor. That's my speculation.

I have used some of those "asinine" clamp-on probes, and some of their nanosecond region (non-clamp on) probes as well. In fact, i have kept the Tek P6026 (and 131 amplifier). Specs: rise time = 18nSec, 20Mhz at 3dB on high end, 230Hz ar 3dB on low end; less than 1dB from 3.5KHz to 10MHz @2mA/mV with termination. Sure as hell is *not* one-point spec! And one can take apart a filament transformer, unwind the secondary, put it back together, and use it as a current transformer that can be calibrated over a decent frequency range, if terminated properly. Just because *you* seem to be unable to do that, does not mean that most others on this NG cannot.

Amen.

Tested with a constant amplitude sine wave from a bench oscillator ( output Z = 600ohms ) imposed across the CT's coil and over the exact same frequency range ( and beyond) as used in the earlier test - the Hioki meter's response is completely flat.

........... Phil

" ** The Hioki clamp meter is a passive device.

Tested with a constant amplitude sine wave from a bench oscillator ( output Z = 600ohms ) imposed across the CT's coil and over the exact same frequency range ( and beyond) as used in the earlier test - the Hioki meter's response is completely flat. "

The narrrow response range is due to magnetic effects alone - as outlined well by Mr Popelish.

Have you never seen how poor the energy coupling is through via core of a laminated iron transformer where the primary and secondary windings are physically separated on different limbs ?

.......... Phil

I think the description of the "pincers" gives a clue to the lousy HF characteristics. Seems to be they were cheap "bent nails", to be a bit crude in the description.

...

Is this discussion giving you a quantummy ache? ;-P

Cheers! Rich

I think you're being just a little bit extreme here - for example, when I hold a tape measure up to my foot, it doesn't change the length of my foot.

Thanks, Rich

I read in sci.electronics.design that Rich Grise wrote (in ) about 'Current measurement', on Tue, 13 Sep 2005:

I wish it did; I'd measure my waist VERY often. (;-)

But it *does* change the length of your leg...