Meant we are having a Debate at work... The debate is on analong and ammeters that read either avergae ,peak to peak ,maxium voltage. I say Average voltage on an analog meter

 

An analog meter with no electronics typically reads the average voltage with the readout calibrated to read the value in RMS for a pure sine-wave. Some old RF analog meters detect the peak voltage with a rectifier, also calibrated to read the RMS value for a pure sine-wave.

 

Meant we are having a Debate at work... The debate is on analong and ammeters that read either avergae ,peak to peak ,maxium voltage. I say Average voltage on an analog meter

what they read depends on the waveform being measured.

 

There is such a thing as a hot wire ammeter that measures true RMS value independent of the shape of the waveform.

 

I was told its called effective which is about 70% of the peak

 

What was the debate?

 

I was told its called effective which is about 70% of the peak

Like I said, it depends on the waveform. RMS (sometimes called effective) is .707 of peak only for a sine wave.

 

The RMS value of a waveform is equal to .707107 times the peak value for a very large number of waveforms, not just sine waves.

Here are two easy to find example waveforms:

 

Meant we are having a Debate at work... The debate is on analong and ammeters that read either avergae ,peak to peak ,maxium voltage. I say Average voltage on an analog meter

The D'Arsonval type movement responds to average value. This is the type of movement that the traditional so-called VOM uses.

The moving iron (iron vane), hot wire, thermocouple, electrodynamometer and electrostatic movements respond to the RMS value.

 

The RMS value of a waveform is equal to .707107 times the peak value for a very large number of waveforms, not just sine waves.

Here are two easy to find example waveforms:

True. But you have to cherry-pick the waveform for that to be valid.

 

Like I said, it depends on the waveform. RMS (sometimes called effective) is .707 of peak only for a sine wave.

To be a bit picky, it is not ONLY for a sine wave; it is easy to construct other waveforms for which it is also true. But the sentiment is valid in the sense that if the waveform you are measuring with an average responding meter calibrated for RMS is not a sine wave, then it is all but guaranteed that the measurement will be wrong.

The RMS value of a waveform is equal to .707107 times the peak value for a very large number of waveforms, not just sine waves.

Pretty much only in the sense that there are a very large number of rational values between 0.5 and 0.6. You can construct an infinite number of either.

Off the top of my head, though, I can't think of any other "simple" waveforms for which it is true -- though there is something nagging the back of my brain whispering that there IS one. Note that part of what I mean by "simple" is where the only parameter that describes the waveform, other than the name, is the amplitude or at least where the RMS value is Vpk/sqrt(2) regardless of the other parameters. Also note that fixing the waveform according to a magic parameter, such as defining the waveform to be a triangle wave with truncated at a specific fraction of the normal peak value, is cheating.

I just remembered the "simple" waveform for which the RMS value is Vpk/sqrt(2). A square wave that varies between 0V and Vpk.

 

True. But you have to cherry-pick the waveform for that to be valid.

There are an infinite number of such waveforms. It's the sine wave that has been cherry-picked.

 

To be a bit picky, it is not ONLY for a sine wave; it is easy to construct other waveforms for which it is also true. But the sentiment is valid in the sense that if the waveform you are measuring with an average responding meter calibrated for RMS is not a sine wave, then it is all but guaranteed that the measurement will be wrong.

One of the most measured waveforms is the grid voltage. It is no longer very close to a sine due to all the capacitor input power supplies in the world. If one wants an accurate RMS value, using an "average responding" meter is the wrong way to go; to do so all but guarantees a wrong measurement. Fortunately, so-called "True RMS" meters are commonly available.

Pretty much only in the sense that there are a very large number of rational values between 0.5 and 0.6. You can construct an infinite number of either.

What do you mean, only? However you define "infinite", denumerably infinite is still a big number.

Off the top of my head, though, I can't think of any other "simple" waveforms for which it is true -- though there is something nagging the back of my brain whispering that there IS one. Note that part of what I mean by "simple" is where the only parameter that describes the waveform, other than the name, is the amplitude or at least where the RMS value is Vpk/sqrt(2) regardless of the other parameters. Also note that fixing the waveform according to a magic parameter, such as defining the waveform to be a triangle wave with truncated at a specific fraction of the normal peak value, is cheating.

You seem to be trying to find a way to rule out anything but a sine. The only reason the sine function is "simple" is that it has been defined by us humans with a single word. It is defined as an infinite series, is differentiable to all orders, retains the same "shape" after repeated differentiation; it is "magic" in a number of ways. Is there not a "magic parameter" involved in the infinite series definition of a sine, the manner in which the terms of the infinite series decrease?

To give examples of waveforms with the property under discussion necessitates combinations of various mathematical functions, which gives a result not previously defined with a name consisting of a single word. I guess that makes these waveforms not "simple".

I can find an infinite series to represent the clipped triangle wave; then there won't be an apparent "magic parameter".

I would claim that even your example:

I just remembered the "simple" waveform for which the RMS value is Vpk/sqrt(2). A square wave that varies between 0V and Vpk.

has a "magic parameter", namely the requirement of a 50% duty cycle.

Maybe what bothers you about my previous examples is the straight line truncation (clipping) of the underlying waveform, leading to a discontinuous first derivative. Here's a "smooth" waveform with the requisite RMS = Vpk/Sqrt(2) property:

Alternate half waves can be flipped in polarity, giving a wavefom with a DC component like your square wave example:

 

Meant we are having a Debate at work... The debate is on analong and ammeters that read either avergae ,peak to peak ,maxium voltage. I say Average voltage on an analog meter

Do you really mean ammeters?

I ask because most current responding meters are atually rather more sensitive and need (internal) shunts to indicate amps.

 

There are an infinite number of such waveforms. It's the sine wave that has been cherry-picked.

By cherry pick I mean you are picking waveforms that happen to have the same average to RMS value as a sine-wave. In general, that's not true, and unless you know the exact wave-shape when you measure the voltage/current you don't know whether the meter is indicating true RMS. So having a large (or infinite) number of waveforms that have the same average as RMS value is generally useless information in real-world measurements with an average responding meter.

 

I believe we are all saying the same thing.
A meter is either showing true RMS value or it is not true RMS value.

Any conversion from average or peak to RMS is a crock.

 

I believe we are all saying the same thing.
A meter is either showing true RMS value or it is not true RMS value.

Any conversion from average or peak to RMS is a crock.

So you are saying we should never use an average responding meter that's calibrated to read the RMS of a sine-wave?

 

By cherry pick I mean you are picking waveforms that happen to have the same average to RMS value as a sine-wave. In general, that's not true, and unless you know the exact wave-shape when you measure the voltage/current you don't know whether the meter is indicating true RMS. So having a large (or infinite) number of waveforms that have the same average as RMS value is generally useless information in real-world measurements with an average responding meter.

The waveforms I'm describing don't have the same average as RMS; they have the same crest factor, the peak to RMS value.

The described abundance was cited only as a counterexample to the statement that only the sine wave has that crest factor. It wasn't intended to be an aid to measurement. It means that encountering a waveform with that crest factor (or close to it) that isn't a sine shouldn't come as a total surprise.

But absolutely pure sine waves are a rarity. In particular, the grid voltage isn't a very good sine wave these days. If a person wants to measure the RMS value of a waveform, use a "True RMS" meter; there's some helpful information for you.

Any conversion from average or peak to RMS is a crock.

Anybody who has a look at their local grid waveform with a scope will see how it has become flattened on top due to the large number of capacitor input power supplies on line.

True RMS meters are becoming the rule rather than the exception. Even low cost DMMs can be had with True RMS capability.

There is even a DMM that can switch modes, and measure waveforms either with an average responding algorithm or True RMS:

http://www.tequipment.net/YokogawaTY720.html

So, if you want to measure the RMS value of a waveform, get a True RMS meter.

 

The waveforms I'm describing don't have the same average as RMS; they have the same crest factor, the peak to RMS value.

The described abundance was cited only as a counterexample to the statement that only the sine wave has that crest factor. It wasn't intended to be an aid to measurement. It means that encountering a waveform with that crest factor (or close to it) that isn't a sine shouldn't come as a total surprise.

..........................

Average responding meters respond to the average of the waveform voltage. Why are you talking about the peak to RMS crest factor which is only pertinent if you have a peak responding meter (which are rare), not an average responding one?

 

What we are saying is get a True RMS reading meter and there is nothing to argue.