TRue RMS meters

[iwire]

Gar, I try but I never understand your posts, it might as well be in another language.

What you are saying seems to fly in the face of a book I bought about using DMMs

As far as a Fluke 87 the specs for my 87 claim

+ or - .7% +4 @ 50 to 60 Hz

or

+ or - 1 % +4 @ 45 Hz to 5 kHz

or

+ or - 2% +4 @ 5 kHz to 20 KHz

 

[gar]

080616-2033 EST

iwire:

I will try to clarify any comments of mine that are not clear to you.

Without a specific question I will assume my comments on RMS are a problem.

It turns out that in the analysis of AC and DC circuits that the average power dissipated in a resistor is Pave = Vrms*Vrms/R. It is assumed that R is a constant. Vrms is a value that produces the same heating effect for both AC and DC voltages.

A true RMS meter that is capable of measuring DC will read exactly the same value as a DC meter even though the construction of the two is quite different.

A true RMS meter provides a voltage reading, no matter what the waveform is, such that when the measured voltage is used in the above Pave equation the heat dissipated in the resistor is the same as that produced by a DC voltage of the same value.

If we simply full wave rectify an AC voltage and measure the DC average value of the rectified waveform we will for most waveforms have a difference in this reading relative to the RMS value.

The easiest way to build an AC voltmeter is to full wave rectify the AC waveform and measure the average DC value. The meter movement does the averaging mechanically. If your were to apply a 1 Hz AC voltage to one of these meters, a Simpson 260 for example, you would see the needle swing back and forth. But at 120 Hz you see no meter needle movement.

If the waveform being measured is a sine wave, then we know that the RMS value is 0.707 times the peak value, and that the average value is 0.636 times the peak. Since many applications are concerned with sine waves the meter manufacture introduces a fudge fact when calibrating the meter. They adjust scaling so the RMS of a sine wave is what the meter reads. When the meter is really reading 0.636 they adjust it so the needle is at 0.707. This is done by just using a slightly different resistor in the meter AC vs DC.

If we measure an AC waveform (with a meter that effectively measures the average of the waveform) that is quite different from a sine wave, then generally the measurement is incorrect. For example a short +2 v pulse of 1 millisecond on and 9 milliseconds off has an average voltage of 0.2 V. The RMS value is 2 squared, averaged = 4/10, and the sq-root of the averaged value is 0.6325 , That is much different than 0.2 .

.

 

[resistance]

080615-2116 EST

What is the significance of an RMS reading?

Start with DC. Voltage times current is defined as power and this is the rate of use of energy. Energy dissipated in a load causes heating. If the load is a constant resistance and the voltage is constant, then Power = V * I, where V and I are average and constant.

The instantaneous power is p = v * i. The average power would be the integral of v*i over a time period of interest divide by that time interval for either or both v and i varying. This does not require that the waveforms are identical and of the same frequency and/or phase.

In an AC circuit the instantaneous power is varying thru a cycle. For a resistive load and sine wave excitation it is proportional to sin t * sin t = 1/2( 1 - cos 2*t ). Thus, it has a DC componet with a superimposed double frequency component. When averaged over an integral number of half cycles of the line frequency the result is a constant.

Root mean square means you square an instantaneous value, average this over N cycles and take the square root of that average.

The RMS value of a waveform ( current or voltage ) applied to a resistance produces the same heating effect as a DC value of the same numeric value.

When you do the intergation and other calculations on a sine wave you get the result that the RMS value = the peak of the sine wave divide by the square root of 2 or a multiplier of approximately 0.707. If you calculate the average value of the full wave rectified sine wave it is approximately 0.636 of the peak value.

A Simpson 260 or 270 meter measures the average value of the input AC voltage, but the meter is calibrated to read the RMS value of a sine wave. If the wave shape is different than a sine wave, then there may be an error in the reading relative to the actual RMS value of that waveform.

A true RMS meter gives a fairly good approximation to the actual RMS value for many waveforms if the waveform is not too peaked relative to the average value.

.

How is U of M!

I use to live in Michigan, and would visit Ann Arbor on a somewhat regular basis. Use to love going to the Mongolian bar in the heart of Ann Arbor.
That was the good old days! I was accepted to attend the Ann Arbor Fire Department Academy, but they took so long getting back to me on my test scores and status, that I had already entered the military when they did contact me.
Oh well!

 

[iwire]

080616-2033 EST

iwire:

I will try to clarify any comments of mine that are not clear to you.

I just don't think we are communicating.

I believe I may be saying the grass is green and you are saying, no the sky is blue.

Thanks for trying.

 

[gar]

080617-1204 EST

iwire:

I would like to solve our communication problem.

So I need to understand where is the difference.

Slightly more on RMS meters. I looked via Google for some picture of the construction of a two coil movement. Found nothing.

A reference that describes this type of meter is
"Basic Electrical Measurements", by Melville B. Stout, 1950, second printing Jan 1951, Prentice-Hall.
On page p419 is a schematic drawing of an electrodynamometer movement, and a not very illustrative photo on page 420. Stout goes into some of the theory, features, and error factors of this type of meter.

These meters have a square law scale. One meter of this type is the Simpson 880 and it has not been manufactured for some time. There may be many old Weston meters like this still in use. One useful feature of the square law scale is that on a 150 V range it provides an expanded scale in the 120 V region of interest.

My 880 which is about 50 years old has a major calibration error reading 134 V with 121 V input. Probably weakened springs. Do I use it? No, not in a long time. However, I need it fixed because it is a useful tool for some purposes.

.