Discussion board help and admin topics > Problems and Solutions for Accurate Measurements

Measuring AC signal with a digital multimeter

(1/2) > >>

ayeaye:
My multimeter is MAS830B, but if you have any other cheap multimeter that has 830 in its type, then it most likely has the same integrated circuit, which means that it functions the same. It is not a bad integrated circuit.

All that is written about measuring AC in the specifications, is that it can measure voltage in its AC ranges up to 400 Hz, and that it measures average. In the AC ranges it is the average of the absolute values, which is a root mean square.

But digital multimeter uses an analog to digital converter to measure values, after that it evidently calculates an average of the measured values. Thus it should also be capable of measuring some AC signals in its DC voltage ranges.

So i tested it. I measured a square wave signal with 50% duty cycle, generated by Arduino.

It appeared that in the DC range, from 10 Hz to 2.5 MHz, it measured 2.43 V, which is an exact average of the TTL signal generated by Arduino, with an amplitude of 4.86 V. And the reading was stable. With more than 2.5 MHz, the measurements became less than that. With the same frequencies it measured 4.8 V in its AC range.

I don't really believe the specification, which says that the precision is +/- 3 (of the displayed value, like with 2.43 it's 2.40 to 2.46), but it's true that the reading can be 1 less (of the displayed value) than the real value, because it does not do rounding.

I did this simple experiment just to have some idea of how it measures. And the speed with which its analog to digital converter works. Use DC ranges to measure AC signal only when your signal is always above zero, or below zero, otherwise the result will be inaccurate. Otherwise, the accuracy of the measurement depends on your signal. You may roughly estimate though, that when your signal rapidly changes back and forth during less than 1 us, then the measurement will not be accurate, as the digital to analog converter doesn't measure that fast.

Ok, don't believe me, test it yourself. I just told about one experiment i did, to have a bit more idea about it. Hope it was useful for someone for some purpose.

verpies:

--- Quote from: ayeaye on October 31, 2016, 12:58:52 AM ---it measures average. In the AC ranges it is the average of the absolute values, which is a root mean square.

--- End quote ---
Wrong

ayeaye:

--- Quote from: verpies on October 31, 2016, 02:15:57 AM ---Wrong
--- End quote ---
Right. But then i any more cannot well explain what it does in the AC range. It should then do something like, measures peak, and when the signal resembles sine, it divides it by sqrt(2), when it doesn't, shows the value of the peak. Which is weird. And wrong, because peak is not the root mean square of the positive square wave.

It is said in its instruction manual about AC range exactly the following:
Frequency range: 40 Hz to 400 Hz
Display: Average (effective value of sinusoid)

In the DC range it looks like that it clearly calculates average. But i have not tried with other signals than square.

verpies:
IMO the meter rectifies and then applies that signal to an RC integrator and that is applied to a voltage divider, that always applies the same correction quotient assuming that the input signal is a sine wave.

AlienGrey:

--- Quote from: verpies on October 31, 2016, 01:02:02 PM ---IMO the meter rectifies and then applies that signal to an RC integrator and that is applied to a voltage divider, that always applies the same correction quotient assuming that the input signal is a sine wave.

--- End quote ---
It's now a RMS chip in the device it's only accurate around the 50 / 60 hz range depending on the timing as you said anything else use a scope and HF in a candescent bulb will vibrate the filament and lite the bulb that way so you might not necessarily have more power at all. (Rout Means Sqr) my advice is look up the chips and see what they do and how they work.

Navigation

[0] Message Index

[#] Next page

Go to full version