Go Down

Topic: ACS712 problem when sensing an inductive load. (Read 2 times) previous topic - next topic

be80be

#5
Dec 21, 2012, 08:13 pm Last Edit: Dec 21, 2012, 08:17 pm by be80be Reason: 1
Your doing your math for DC load to use this with a motor you'll need to run it threw a R-M-S converter

RIDDICK

what if u take 1000 samples per second?

what is the minimum?

what is the arithmetic mean of the samples of 1 second (i think that is what the DC amp meter measures)?
http://en.wikipedia.org/wiki/Arithmetic_mean
-Arne

jackrae

Firstly, what makes you think your test meter is telling the "truth"

You refer to a resistive load - is this with a DC supply or the AC supply.

If the former then using a light bulb (resistive) repeat your tests with an AC supply and see what you get after applying RMS sampling.

If there's a difference between meter and transducer then I respectfully suggest your test meter is telling porkies.

With respect to differences between resistive and inductive current, there is none,  the "difference" is in the phase angle between the current flowing versus the system voltage.  An amp is an amp, irrespective of what its flowing into.  Its "shape" in non-DC type circuits might change but that's where RMS correction comes into play (to determine the equivalent power or heating capability)


Darksaga28

Actually, there's a difference between current on resistive load and current on inductive load other than the phase angle. There are some non-linear devices that can make your ''Current curve'' different (Not sinusoid), so in case of nonlinearity; RMS and Peak values are useless when you talk about heating capability or DC equivalent power. Google ''True RMS''.

jackrae

No-one said it was the same.  You will note that I referred to the "shape" of the curve and the need to factor in RMS correction.   There only is 'true-RMS' that you refer to and that is simply "RMS" as defined by the acronym Root-Mean-Square.  Most simple or inexpensive meters that are supposed to measure AC voltages and current assume a sinusoidal waveform.   

Or to put it another way, the test meter is designed/calibrated on the assumption that the waveform is sinusoidal whereas the motor load current will definitely not be sinusoidal due to the continuously changing inductance of the motor windings as the rotor bars pass the pole pieces.  The transducer is responding to these non-linearities, as it is designed to do.

Many years ago when working on large generator synchronisation problems between remote oil platforms we could clearly see the rotor bar passing frequency superimposed onto the line frequency.  This was sufficient to cause major headaches before we resolved the issue.

Go Up