True RMS vs averaging DMM

[Tubetec]

I spotted a great deal on a local second hand website , a fluke 77IV a few years old but more or less unused (cal cert expired 2019) ,T-5 600 clampmeter (well used but functions perfectly )and Martindale proving unit(440V ac) and voltage indicator again in more or less brand new condition ,around 1100 euros worth new , I got the lot for 190 euros .
The 77IV closely matches my 187 on all DC and ohm ranges but showed an appreciable difference on AC volts which worried me initially until I found out the 77VI isnt an RMS meter .
Then I found Daves teardown of the 77IV over on EEV blog , turns out the meter is identical to the model 179 including the PCB , only difference is a Tant cap is left out on the 77IV ,so its easy to turn your 77 into a 179 true RMS meter , although a re-cal is required after the mod .

Im just wondering in what circumstances is an averaging meter a benefit over an RMS unit ?

 

[Brian Roth]

What frequency were you measuring? A 187 has a flat HF response out to 100 kHz as I recall. Inexpensive meters crap out at a much lower frequency since they are intended for electricians measuring at 50 or 60Hz. Most of these are usually accurate at 400 Hz which was used in aircraft.

Recheck your measurements using, say, 100 Hz and see how the two compare.

Bri

 

[JohnRoberts]

Im just wondering in what circumstances is an averaging meter a benefit over an RMS unit ?

An averaging meter is easier, cheaper.

I did some work on a microprocessor based console meter where I coded up an RMS version. With identical attack/release times fed from a complex music source I could not perceive any difference between RMS and simple average meter response.

JR

 

[warpie]

I spotted a great deal on a local second hand website , a fluke 77IV a few years old but more or less unused (cal cert expired 2019) ,T-5 600 clampmeter (well used but functions perfectly )and Martindale proving unit(440V ac) and voltage indicator again in more or less brand new condition ,around 1100 euros worth new , I got the lot for 190 euros .
The 77IV closely matches my 187 on all DC and ohm ranges but showed an appreciable difference on AC volts which worried me initially until I found out the 77VI isnt an RMS meter .
Then I found Daves teardown of the 77IV over on EEV blog , turns out the meter is identical to the model 179 including the PCB , only difference is a Tant cap is left out on the 77IV ,so its easy to turn your 77 into a 179 true RMS meter , although a re-cal is required after the mod .

Im just wondering in what circumstances is an averaging meter a benefit over an RMS unit ?

As long as you measure sinusoidal signals I don't think it makes any difference.

 

[Tubetec]

Heres a link to the EEV blog page about the 77VI ,
https://www.eevblog.com/forum/blog/eevblog-1448-convert-a-fluke-77-iv-to-true-rms-for-10-cents/In this case the 77vi is equipped with the exact same RMS capable chipset as the 170 series only a 33uf tant smd chip is left off the board .

 

[JohnRoberts]

In theory "true RMS" metering should be more representative of the energy content of a waveform (area under the curve).

Some have argued (like dbx did last century) that RMS detection was more accurate for processing (touted about their old tape NR systems). In practice I found other factors more significant for record/playback accuracy (like the side chain design). I killed too many brain cells over tape NR design back in the day.

JR

 

[abbey road d enfer]

The 77IV closely matches my 187 on all DC and ohm ranges but showed an appreciable difference

What is appreciable? 10%? 1%?

on AC volts which worried me initially until I found out the 77VI isnt an RMS meter .

What were you measuring?
On a mid-frequency sinewave, there should be no difference. On a signal with a crest factor different than 1.414 (case for a sinewave), some difference may show up. A non constant signal, such as audio program may exhibit differences because the integration time.

Im just wondering in what circumstances is an averaging meter a benefit over an RMS unit ?

Average is usually simpler. For example, a galvanometer is intrinsically an average meter.

 

[[email protected]]

What is appreciable? 10%? 1%?

What were you measuring?
On a mid-frequency sinewave, there should be no difference. On a signal with a crest factor different than 1.414 (case for a sinewave), some difference may show up. A non constant signal, such as audio program may exhibit differences because the integration time.

Average is usually simpler. For example, a galvanometer is intrinsically an average meter.

The 77-IV is specified as 2% accuracy 45Hz-1KHz, the 187 is within 1.5% up to 20KHz, and has TRMS, AC millivolt ranges and dB readings.
The specification chart in the user manuals is worth looking at as it is quite different in different sensitivity ranges. If you know the capabilities and limitations of your instruments they are less likely to lead you astray.

 

[Speedskater]

the Fluke 77IV has a 6 Volt AC range, so you could measure the 20-20,000 Hz output of a digital source to get a handle on the meter's frequency response.

 

[user 41968]

One recent instance where an RMS meter was advantageous was measuring the carrier level of a 19/20 kHz (1:1) IM tone. The average level measures over a dB different from the "true" RMS.

 

[JohnRoberts]

One recent instance where an RMS meter was advantageous was measuring the carrier level of a 19/20 kHz (1:1) IM tone. The average level measures over a dB different from the "true" RMS.

with 19 kHz and 20 kHz which one is the carrier?

[edit- back in the 70s when I converted my heathkit SMPTE IMD to work with 19/20kHz I used the simple average meter in the Heathkit. I guess this makes all my benchwork back then suspect? [/edit]

JR

 

[user 37518]

I spotted a great deal on a local second hand website , a fluke 77IV a few years old but more or less unused (cal cert expired 2019) ,T-5 600 clampmeter (well used but functions perfectly )and Martindale proving unit(440V ac) and voltage indicator again in more or less brand new condition ,around 1100 euros worth new , I got the lot for 190 euros .
The 77IV closely matches my 187 on all DC and ohm ranges but showed an appreciable difference on AC volts which worried me initially until I found out the 77VI isnt an RMS meter .
Then I found Daves teardown of the 77IV over on EEV blog , turns out the meter is identical to the model 179 including the PCB , only difference is a Tant cap is left out on the 77IV ,so its easy to turn your 77 into a 179 true RMS meter , although a re-cal is required after the mod .

Im just wondering in what circumstances is an averaging meter a benefit over an RMS unit ?

When measuring pure sinusoidals, both the RMS and averaging DMM should be pretty much give the same reading, however, if there are harmonics present the averaging DMM will err. One of the most extreme cases is when measuring square waves, I've seen averaging DMMs give a reading which differs by several volts from the True RMS DMM.

True RMS DMMs are particularly important for people dealing with mains voltages and stuff like measuring voltage when there are reactive or switching loads which add a lot of harmonics into the line, for example in industrial applications, in such case, the True RMS DMM becomes a necessity. For audio, most of the time you will be measuring pure sine waves, like when calibrating a tape machine, or measuring the output of an equipment, in that case the True RMS DMM will not prove to be that much better unless there is something seriously wrong with your gear, in which case, you should be using an oscilloscope to see what is going on and not a DMM.

However, in general the True RMS multimeters also are more accurate than the averaging DMMs, not only due to the fact that they measure the RMS voltage more accurately, but rather because the meter itself is better (and more expensive).

 

[abbey road d enfer]

I've seen averaging DMMs give a reading which differs by several volts from the True RMS DMM.

Relative to what?

 

[user 37518]

Relative to what?

To the real value or multiple other True RMS meters, yes, you could argue that how do we know those meters show the real value, but you get what I mean, I do not want to engage in a "there is no spoon" argument

 

[moamps]

Here is a Quora type question intended especially for those members who think that DIYers do not need oscilloscopes, but that a DMM is sufficient.
The linear power supply consists of a transformer, a full-bridge rectifier, a capacitor and a 7812 regulator. The regulator needs a minimum ratio of Vin-Vout=3V to work properly.
Average DMM in DC position measures 17V at the reg input. What is the maximum voltage that can be measured in the AC position at the reg input for the regulator to work correctly?

 

[abbey road d enfer]

To the real value or multiple other True RMS meters, yes, you could argue that how do we know those meters show the real value, but you get what I mean, I do not want to engage in a "there is no spoon" argument

OK, seems you didn't get my jab.
"averaging DMMs give a reading which differs by several volts from the True RMS DMM"
A big deal if you're measuring a few volts, a iota if you're measuring 380kV AC lines.
I was taught never to mention differences in absolute value, but rather in relative.

 

[user 37518]

OK, seems you didn't get my jab.
"averaging DMMs give a reading which differs by several volts from the True RMS DMM"
A big deal if you're measuring a few volts, a iota if you're measuring 380kV AC lines.
I was taught never to mention differences in absolute value, but rather in relative.

Yes, abbey, I am given as a fact that we are not measuring 380kV lines and, by the way, using the square wave example I gave and given the possibility to measure equally a 380kV square wave or a 5V square wave with a DMM, the difference in volts between the averaging and True RMS DMM should be proportionally the same, not a iota and still a big deal. However, I think we are now in imaginary land...
I was taught to put things into perspective.

 

[JohnRoberts]

As often happens internet threads veer off into the weeds. It is instructive to understand how your meter interprets different waveforms. This is why we like using scopes for sussing out difficult waveforms.

JR

 

[sodderboy]

Fluke 8060A, baby. I accept no substitutes. I used one on my second day in the studio (the first day I just go to see the control room on a tour) and have needed no other DMM since. They need to be maintained over the decades of use, but completely reliable, accurate, and the absolute best recording studio meter. $375 was my price of admission in 1987, but they have come down a bit since. Had to retire that one in the oughties. . . I moved the case to a newer meter for the mojo.
Mike

 

[user 37518]

Fluke 8060A, baby. I accept no substitutes. I used one on my second day in the studio (the first day I just go to see the control room on a tour) and have needed no other DMM since. They need to be maintained over the decades of use, but completely reliable, accurate, and the absolute best recording studio meter. $375 was my price of admission in 1987, but they have come down a bit since. Had to retire that one in the oughties. . . I moved the case to a newer meter for the mojo.
Mike

I like Fluke meters, I own the good old 87V, but over time I became more fond of Agilent/Keysight meters, my U1282A is practically indestructible (Dave Jones threw it several times from the top of a dam into solid concrete and it still worked after several drops) and the accuracy is superior to the 87V.

Exactly what makes the 8060A the best studio meter? or better put, what does the 8060A have (besides the now yellowish old plastic) that no other of the newer—and much better— meters have?