Author Topic: Measuring Audio Interface Outputs with Multimeter - Need help please  (Read 7144 times)

0 Members and 1 Guest are viewing this topic.

Offline themadhippy

  • Super Contributor
  • ***
  • Posts: 2569
  • Country: gb
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #25 on: November 24, 2022, 02:46:18 pm »
And if your going for totally accuracy don't forget to load the meter down so its the correct input impedance for your audio output interface

Quote
RMS Voltage (simple equation) = Peak Voltage

VRMS = Vpk  x  0.7071
Vpk  = VRMS x1.414
« Last Edit: November 24, 2022, 02:51:51 pm by themadhippy »
 
The following users thanked this post: Voltzs

Offline ledtester

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #26 on: November 24, 2022, 03:07:09 pm »
RMS Voltage (simple equation) = Peak Voltage  :)

RMS * sqrt(2) = peak voltage
 

Offline VoltzsTopic starter

  • Contributor
  • Posts: 19
  • Country: at
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #27 on: November 24, 2022, 03:23:33 pm »
And if your going for totally accuracy don't forget to load the meter down so its the correct input impedance for your audio output interface

Quote
RMS Voltage (simple equation) = Peak Voltage

VRMS = Vpk  x  0.7071
Vpk  = VRMS x1.414

Thanks for your reply - how do I load the meter down to match the impedance to my audio output interface? I guess there is an option on the Aneng 8008 for this?

I attached a screen shot of my interface's specs - though the monitor outs (which is what I am testing) may differ from the line outputs.
« Last Edit: November 24, 2022, 03:26:44 pm by Voltzs »
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7825
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #28 on: November 24, 2022, 03:26:07 pm »
I just ordered the Aneng 8008 - but as it only displays RMS voltage how would I calculate the Peak voltage which is what I need?

Why do you need peak voltage?  4dBu is 1.228Vrms across 600 ohms.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online wasedadoc

  • Super Contributor
  • ***
  • Posts: 1356
  • Country: gb
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #29 on: November 24, 2022, 04:00:56 pm »
Thanks for your reply - how do I load the meter down to match the impedance to my audio output interface? I guess there is an option on the Aneng 8008 for this?
No, there is no such option on any general purpose multimeter.  Such a multimeter is usually desired to place only the lightest load on whatever voltage source it is measuring.  Typically the multimeter puts a 10 MOhm (=10,000,000 Ohm) load.  The Aneng 8008 is no exception.  You need to provide your own 600 Ohm resistor.

If you get an "audio millivoltmeter" it will likely have an option to switch in a 600 Ohm termination.  But expect to pay many times the price of the Aneng.
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 7942
  • Country: us
  • Retired, now restoring antique test equipment
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #30 on: November 24, 2022, 05:39:15 pm »
"dBu"  is actually a voltage measurement, and does not imply the impedance at the node being measured.
"dB" is technically a power ratio, but can be used to express a voltage ratio (with the silent assumption that the impedances at each voltage measurement are equal).
A suffix after "dB" (e.g., dBm, dBu, dBV) indicates the denominator of the ratio (mW, etc.).
0 dBu is the voltage equal to the voltage that produces 1 mW (0 dBm) in a 600 ohm load, but may well be measured into an undefined high impedance.
In practice, if the load impedance seen by your source in use is not a very high resistance, you should add a suitable resistor and measure the voltage across it for an accurate value.
If your source has an output impedance of, say, 600 ohms and your normal load is 10,000 ohms (a "bridging" value), the difference between 10 k and infinite load is 0.5 dB.
 

Offline ledtester

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #31 on: November 24, 2022, 10:13:57 pm »
 

Offline DavidAlfa

  • Super Contributor
  • ***
  • Posts: 5896
  • Country: es
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #32 on: November 24, 2022, 10:34:26 pm »
You need to connect a 600ohm resistor to the multimeter tips.
Or measure with a receiver connected, starting at min. level and rising the level slowly to avoid overloading the input.
« Last Edit: November 25, 2022, 05:43:34 am by DavidAlfa »
Hantek DSO2x1x            Drive        FAQ          DON'T BUY HANTEK! (Aka HALF-MADE)
Stm32 Soldering FW      Forum      Github      Donate
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16607
  • Country: us
  • DavidH
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #33 on: November 25, 2022, 05:34:17 am »
Some modern multimeters, and many older ones, have the frequency response and even readout in dBu or dBm to support audio applications.
 

Online wasedadoc

  • Super Contributor
  • ***
  • Posts: 1356
  • Country: gb
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #34 on: November 25, 2022, 09:49:40 am »
Some modern multimeters, and many older ones, have the frequency response and even readout in dBu or dBm to support audio applications.
Any sub US$50 examples?
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14181
  • Country: de
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #35 on: November 25, 2022, 09:59:14 am »
Moden audio gear does not use impedance matching. So the output are low output resistance, like 10-100 ohm, audio input are more like 10-100 kOhm input impedance.
The 600 Ohm are mainly used to convert the power scale to a voltage scale. AFAIK the u in dBu is there to show it is about measuring the voltage and not actual power.
Some audio range gear still sometimes has 600 Ohm impedance, but this is more like a problem.

Many of the new DMM chip sets include RMS reading, not by an analog RMS -> DC converter, but by using a moderately fast SD ADC and than do digital RMS calculation. These DMMs (e.g. likely the mentioned Aneng 8008) are not that bad for AC readings, but a bit different from the old style analog RMS converters with difference weak points:
The digital ones are usually relatively limited bandwidth (e.g. 1-5 kHz) with a usually relatively sharp drop and the maximum crest factor near full scale may be limited.
The analog ones are somewhat sluggish to react, have problems with small signals (e.g. < 1-5% of FS) and often a bandwidth that depends on the amplitude. The response at low frequencies (e.g. < 30 Hz) is often not good (no longer true RMS) - they have to compromise with response speed.
So it depends which version is preferred. Just because it is cheap the build in RMS option is not bad - in some aspects it even beats much more expensive old style meters with analog RMS converters.
 
The following users thanked this post: RJSV, golden_labels

Offline RJSV

  • Super Contributor
  • ***
  • Posts: 2120
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #36 on: November 25, 2022, 10:12:28 am »
   There needs to be A WORD for this...STAMPEDE !!!

   Dear lord God...And sorry to you, after posting a decent question, for enduring this mess of unhinged responses...
   1.) First of all, looking at picture, I see a stereo jack, with the voltmeter across,...Left channel hot, measuring whatever voltage is, related to...right channel, HOT ???  That's going to be ??? Not even sure, as that relates to what program you've got playing (music), or likely it's 2 separate sine waves, left and right channel.  But you probably need to measure one channel, relative to the that stereo jack ground...that would be one or the other if the two 'tip' signals, on the stereo jack.
   Now, please note; my response isn't gonna go off into the weeds, with A CIRCUIT YOU NEED TO BUILD FIRST.
Yeah, sure, folks, that's like telling (this poor OP);
  Before you type in your response ...you NEED to build a...A Table, for your laptop, (get some wood...).

OK 2.) My Velleman DVM890F has 2VAC range, but likely not accurate for audio at 1 khz, as some responses have indicated here.  Between the Spec. and, assuming proper connected probe(s), there might be an acceptable de-rating, like "reading at 1 khz is low and multiply by 1.5 or whatever".

3.).  After all this 'stampede' of things that OP needs to build, or 'resolve'...by the time it gets to "YOU NEED 600 ohms, in series...NO, YOU NEED one 600 ohms on EACH probe tip...so that's 2 resistors, (and it's virtually an infinite impedance meter. OK, 1 million ohms, whatever.  By this time, though, I'm done...even if the 600 ohm thing is valid.

   Folks responding: I am very disappointed, as this poor fellow / gal is getting a stampede of stupid, instead of help.  Is it too much EGGNOG, out there tonight ?
   Hopefully, my sarcasm doesn't scare off a beginner with good questions, it's these 'responses' that are a ludicrous STAMPEDE.
   Maybe, it would help to ask some folks at recording studio: "What meter is good for reading db levels, at standard 1 khz, along with maybe a corrective factor for frequency response, at 1 khz ?".

   SOME replies here have that, if you can sort out this mess.
 
The following users thanked this post: Ground_Loop

Offline RJSV

  • Super Contributor
  • ***
  • Posts: 2120
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #37 on: November 25, 2022, 10:17:36 am »
...oh and I believe there might also be some confusion about 'line' or 'speaker' outputs from or to your 'monitor'.  A speaker output probably reads lowish voltage, vs line out, when measuring open circuit.
 

Offline The Soulman

  • Frequent Contributor
  • **
  • Posts: 949
  • Country: nl
  • The sky is the limit!
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #38 on: November 25, 2022, 10:33:34 am »
RJ,

I haven't replied to this thread because op's question isn't clear to me and the thread gained enough attention already.
But as a response to your post:
1) that's likely a balanced signal on that jack.

3) 600 Ohms resistor could be parallel to simulate the minimum load impedance, as the output impedance is not zero this will have some effect.
edit: it all depends on the entire setup/signal chain, in the good old (analog) days a 1KHz sine wave signal and a calibrated meter was used to line all different stages up. 

Everything recording/playback device (where absolute values matter) is digital these days, if you want fancy:
https://www.orban.com/freeorbanloudnessmeter

Personally I'd use what is in on my digital consoles together with a blackmagic smartscope duo 4k to show signal levels and x-y plot.

« Last Edit: November 25, 2022, 10:38:20 am by The Soulman »
 
The following users thanked this post: RJSV

Offline VoltzsTopic starter

  • Contributor
  • Posts: 19
  • Country: at
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #39 on: November 25, 2022, 11:00:14 am »
RJ,

I haven't replied to this thread because op's question isn't clear to me and the thread gained enough attention already.
But as a response to your post:
1) that's likely a balanced signal on that jack.

3) 600 Ohms resistor could be parallel to simulate the minimum load impedance, as the output impedance is not zero this will have some effect.
edit: it all depends on the entire setup/signal chain, in the good old (analog) days a 1KHz sine wave signal and a calibrated meter was used to line all different stages up. 

Everything recording/playback device (where absolute values matter) is digital these days, if you want fancy:
https://www.orban.com/freeorbanloudnessmeter

Personally I'd use what is in on my digital consoles together with a blackmagic smartscope duo 4k to show signal levels and x-y plot.


Hi there Soulman, I thought my question was worded clearly - and in your above reply you actually touched on a key part of what I have been trying to establish: matching the inputs and outputs of analog gear that is connected!

Most professional audio companies will include what calibration signal was used to establish 0VU (+4dBu) at the product's analog outputs, most typically either -20 or -18 dBFS will equal 0VU.

Let's say I do not know if my interface's outputs where calibrated for 0VU to mean -20 or -18 dBFS - what happens down the line when I calibrate my speakers to a specific dBSPL is I do not know if this dBSPL refers to one particular level or another.

Besides speaker calibration, there are countless other reasons to want to level match gear which, assuming from your reply, you understand.

This brings me to my search for a multimeter capable of reading the voltage level at my interface's outputs when playing either a -20 or -18 dBFS 1kHz signal.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16607
  • Country: us
  • DavidH
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #40 on: November 25, 2022, 04:53:35 pm »
Some modern multimeters, and many older ones, have the frequency response and even readout in dBu or dBm to support audio applications.

Any sub US$50 examples?

I did a search and the lowest cost option I found was the UNI-T UT61E which has a bandwidth of 10kHz and costs $72 on Amazon.  It does not read out in dB but that is not really necessary.  I assume that you want a new meter and not an older used one.
 

Online BeBuLamar

  • Super Contributor
  • ***
  • Posts: 1192
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #41 on: November 25, 2022, 07:02:26 pm »
dBu doesn't need the 600 ohms resistor. Only dBm needs it because it's specified in term of power. dBu is in term of voltage only. And yes it's RMS voltage not peak.
 
The following users thanked this post: RJSV

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 3345
  • Country: ua
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #42 on: November 25, 2022, 09:34:47 pm »
you're needs to add proper termination load on your source before measurement.
Usually audio equipment uses 600 Ω impedeance, but you're needs to check it for your audio equipment.
This is needed because unloaded source will have about twice higher Voltage.

When you place proper termination load, you're needs to switch your DMM to measure AC RMS Voltage.
Just FYI, on your screenshot DMM is switched to measure diode drop down, this is the reason why you see random value.

+4 dBu = 1.228 Vrms

dBu doesn't need the 600 ohms resistor. Only dBm needs it because it's specified in term of power. dBu is in term of voltage only. And yes it's RMS voltage not peak.

But since any signal transmitter has some internal impedance, it's Voltage output depends on receiver equipment impedance. This is because transmitter and receiver impedance is a kind of Voltage divider. If you try to measure unloaded source, you can get up to 2 times higher Voltage. So, if you want to measure real Voltage, proper termination load is required.

In some cases receiver equipment can use Hi-Z input, in such case termination load is not required for Voltage measurement. But if receiver has 600 Ω input and measurement is performed with disconnected receiver, then you're needs to put 600 Ω termination load on transmitter for proper Voltage measurement.
« Last Edit: November 25, 2022, 11:36:58 pm by radiolistener »
 
The following users thanked this post: RJSV

Online BeBuLamar

  • Super Contributor
  • ***
  • Posts: 1192
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #43 on: November 26, 2022, 12:43:02 pm »
I feel 1.28V at -20dB is too high because at 0dB it's 12.8V.
 

Online donlisms

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #44 on: November 26, 2022, 08:18:20 pm »
Wow.

The goal you have expressed is called "gain structure."  The idea is that within in an audio system (used for making live or recorded sound), there are many places along the way to adjust the signal level, and having it high "here" and low "there" will produce the same output level as having it low "here" and "high" there, so... which one is right?  :)

The answer is "it depends," but generally speaking, the goal is to get the input signal into a range that's good for processing and mixing, and keep that approximate signal level all the way through the processing, and then get it into a proper range for listening.  Listening is usually either headphones or a power amplifier feeding a speaker, sometimes with the power amplifier inside the speaker's box so you don't have to think about it very much.

Measuring the output level from one of your boxes should be a simple matter of using an AC voltmeter that can measure the appropriate range, which is typically (but not always) 2 volts down to a small number of millivolts.

Do not worry about 600 ohms; that came from a book, based on how audio was done a long time ago -- not how it's done today.  Measure the voltage with your meter, and that is that.  If you really need to know input and output impedances, first assume 100 ohms for an output impedance, and then assume 1000 ohms for a balanced mike input, and 10k ohms for a line input.  But to make use of those, now you have to learn about voltage dividers, and I don't think you're ready for that yet.  The bottom line here is that the output level will depend, to a small degree, on what you plug it into.  And my best advice is "ignore that."  Measure the voltage with the output unloaded, and assume that if the level drops, you will turn a knob or slide a fader to make it go up again.

So, yes, you want a meter that measure AC voltage below 2 volts, and hopefully, ideally, to 20kHz (or beyond).  While you can get by with this one particular task if it's limited to 1kHz, it will not be as good in the long run.  This has mostly to do with the fact that any audio signal other than a sine wave is going to have harmonics at 2kHz, 3kHz, 4kHz, and so on, all the way up, and the things that make the sound "what it is" are going to be above 1kHz.  If the meter cannot see them, any interesting signal (something beside a flute or a super-clean synth patch) will not be accurately represented.  For you, that low-level AC spec, with the voltage and frequency, is obviously the most important, but understanding the rest will help you.  (For example, understanding current might help you understand why you can't just keep adding more and more (unpowered) speakers to the same amplifier channel; why the amplifier maker says you can use it down to 4 ohms, or 2 ohms if it's a really good one.)

A balanced signal, which is what I think you're dealing with, has a positive version and a negative version.  On an XLR connector, these appear on pins 2 and 3; on a tip-ring-sleeve connector, they appear on the tip and the ring.  The signal level is the difference between those two points; that's what you want to measure.  Pin 1, or the sleeve, is ground, and it is not relevant to a balanced audio signal.  It's used for the shield, to help reduce noise.

I'm not sure why you introduced the word "peak."  There are two distinctly different kinds of "peak" in audio.  One of them is shared with electronics in general, i.e., that the peak level is sqrt(2) times the RMS level.  This kind of peak is extremely rare, one might even say "never!", to use such a level in audio.  It's always an RMS level, but... that level is changing all the time, because that what sound does: it gets louder and softer.  So that's the other kind of "peak": when it's loud loud loud.

(There is a third kind of "peak", used by companies that make cheap amplifiers and try to convince you they are powerful enough to blow up a house...  When they say "1000 watts peak power", they are saying "we are lying to you to get you to buy this thing.")

In other words, the idea of "peak" depends on the context.  If we were sitting together, I would ask you where that word came from; what did you see that made you start thinking about it?

As to the specific meter, I think the most helpful thing for you would be to learn to read the specs.  Look at each one, and try to learn what you need to learn to understand what they are saying.  First would be the ranges; look at the voltage ranges for both DC (which might be interesting, but not for your current problem), and for AC (which is very important).  So for AC, you might see "200mV", which is millivolts, the three digits after the decimal point.  Let nothing go unlearned!  Look for an answer for every specification, every measurement.  This sort of thing will serve you in the future as well.

As to your overall project, I think it's an interesting one and will give you lots of things to learn.  But at the same time, levels in audio are pretty much always subjective, especially because music and voice (and other) levels are changing all the time.  It's not "one level", it's an approximate "range of levels."  So getting it close is the best you can do.  You want it in a reasonable range so that your equipment can work properly, so you're not overloading the inputs (too high!) and you're not getting too much noise mixed in (too low!).  Then, you just see how you like it, and push it up or down as appropriate.

And it's always a matter of thinking about how to deal with it.  Suppose, for example, I'm mixing a recording of a band, and sometimes I listen to it on "these" speakers, which take 0dBu for full volume, and sometimes "those" speakers, which take -20dBu for full volume.  When I switch the speakers, I set the master (output) volume of the mixer higher or lower by 20dB, and everything is fine.  More likely what I would do is just slide the slider until it sounds like the level I want.  Generally speaking, that is NOT going to be the maximum volume the speakers are capable of producing, because my music has interesting peaks (there's that word!  This is the "loud loud loud" version) of at least 10dB over the normal, average level.  So now I need to make sure the average level is at least 10dB lower than the peaks, in order to keep the peaks from exceeding the speaker's maximum.  And that will still sound nice and loud, because acoustic levels don't usually have to be anywhere near as extreme as most marketing departments would like you to believe.

About your audio interface, I would bet that it's pretty darn close to what they say it is.   I have never measured any of mine; they just work.  And they work fine.  "More is more," and you turn the knob up, "less is less" and you turn the knob down.  Much more interesting is how much hissy noise comes out when you turn the knob up.

If you're not doing music or other "real" sounds, and instead you're actually doing something technical, then you've got a whole lot of considerations to think and learn about!  Much of the discussion changes.
 
The following users thanked this post: Voltzs

Offline EPAIII

  • Super Contributor
  • ***
  • Posts: 1057
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #45 on: November 28, 2022, 11:45:44 am »
Audiophiles are a real big PITA! They don't know what they are doing and they never let that stop them.

There are two ways of setting up an audio system. The first is for a professional system where audio must constantly be sent in different directions for different purposes. It can go to from item A to item B for one reason and then from item A to item C a few minutes later. In such a system ALL the equipment should be capable of sending and receiving a common audio level so that you are not constantly adjusting the levels as the work day progresses. This is where the system designer must establish a single level and implement it across the entire system. That single level is going to be a compromise because there is always going to be some items of equipment that are different. Pads and amplifiers can help a lot. Professional equipment, like mixers, will have ways to cut down or boost up each of the inputs. And in equipment with multiple outputs, each one is often individually adjustable.

A second type of system is mostly connected ONE way and left that way: no switching, no patching. In such a system the best level for connecting A to B can be determined by not only the standard levels that each of those items work with, but also by looking at the NOISE FLOOR and PEAK WHERE DISTORTION WILL OCCUR. The difference between those two levels is your operating range and, depending on the type of audio you are processing, an operating point (zero level) somewhere between them can be established. You want head room for the peak levels while not hearing the noise which should be at least 40 db down from the chosen operating point. Those operating levels may and probably WILL be different for every combination of two pieces of equipment. That is how such a system is optimized. And all this discussion about what meter to use and "true RMS", and other nit-picking specs is just a lot of hot air. Get a working, USED, NAME BRAND, ANALOG meter with a 0 to 1 Volt AC scale and take all your measurements with the circuits connected as they will be while operating: A output to B input. Forget the termination resistors too: they are for the other type of system I described above. Most such meters will respond well to a 400 or 1000 Hz test signal and give you a good indication of actual audio's relative levels. That is how such a system will operate (with permanent connections) and that is the only way that signal measurements in it make any sense. And the very last thing you need to spend money on is "true RMS".

As to how to determine the noise floor and distortion limit, the human ear is a great test instrument. Well, if it has not been exposed to too much extremely loud audio. And if your ears have been damaged by loud sounds, well you probably don't care anyway. So save your money for better beer or wine to sip while listening. Boosting the level further down the line, at the speakers, will help to hear the noise when no audio is present.

PS: Some NAME BRAND, analog meters include Simpson, Triplett, Hickok, Honeywell, RCA, Sencore, Siemons, Western Electric, and Weston. There are probably more but avoid anything made in the orient.
« Last Edit: November 28, 2022, 11:48:13 am by EPAIII »
Paul A.  -   SE Texas
And if you look REAL close at an analog signal,
You will find that it has discrete steps.
 
The following users thanked this post: Voltzs

Offline Brianf

  • Regular Contributor
  • *
  • Posts: 73
  • Country: gb
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #46 on: November 29, 2022, 09:42:13 am »
dBu doesn't need the 600 ohms resistor. Only dBm needs it because it's specified in term of power. dBu is in term of voltage only. And yes it's RMS voltage not peak.

^^^ That.

Audio equipment, except in a very few niche applications, hasn't needed 600R terminations since the last century. And even then it was only for a few special applications. Anything built in the last 40 years has been low impedance out, high impedance in, which renders the whole concept of termination meaningless.

Termination was all about the transfer of maximum POWER, today we are interested in the transfer of maximum VOLTAGE.
 
The following users thanked this post: Voltzs

Offline VoltzsTopic starter

  • Contributor
  • Posts: 19
  • Country: at
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #47 on: November 29, 2022, 11:34:13 am »
Wow.

The goal you have expressed is called "gain structure."  The idea is that within in an audio system (used for making live or recorded sound), there are many places along the way to adjust the signal level, and having it high "here" and low "there" will produce the same output level as having it low "here" and "high" there, so... which one is right?  :)

The answer is "it depends," but generally speaking, the goal is to get the input signal into a range that's good for processing and mixing, and keep that approximate signal level all the way through the processing, and then get it into a proper range for listening.  Listening is usually either headphones or a power amplifier feeding a speaker, sometimes with the power amplifier inside the speaker's box so you don't have to think about it very much.

Measuring the output level from one of your boxes should be a simple matter of using an AC voltmeter that can measure the appropriate range, which is typically (but not always) 2 volts down to a small number of millivolts.

Do not worry about 600 ohms; that came from a book, based on how audio was done a long time ago -- not how it's done today.  Measure the voltage with your meter, and that is that.  If you really need to know input and output impedances, first assume 100 ohms for an output impedance, and then assume 1000 ohms for a balanced mike input, and 10k ohms for a line input.  But to make use of those, now you have to learn about voltage dividers, and I don't think you're ready for that yet.  The bottom line here is that the output level will depend, to a small degree, on what you plug it into.  And my best advice is "ignore that."  Measure the voltage with the output unloaded, and assume that if the level drops, you will turn a knob or slide a fader to make it go up again.

So, yes, you want a meter that measure AC voltage below 2 volts, and hopefully, ideally, to 20kHz (or beyond).  While you can get by with this one particular task if it's limited to 1kHz, it will not be as good in the long run.  This has mostly to do with the fact that any audio signal other than a sine wave is going to have harmonics at 2kHz, 3kHz, 4kHz, and so on, all the way up, and the things that make the sound "what it is" are going to be above 1kHz.  If the meter cannot see them, any interesting signal (something beside a flute or a super-clean synth patch) will not be accurately represented.  For you, that low-level AC spec, with the voltage and frequency, is obviously the most important, but understanding the rest will help you.  (For example, understanding current might help you understand why you can't just keep adding more and more (unpowered) speakers to the same amplifier channel; why the amplifier maker says you can use it down to 4 ohms, or 2 ohms if it's a really good one.)

A balanced signal, which is what I think you're dealing with, has a positive version and a negative version.  On an XLR connector, these appear on pins 2 and 3; on a tip-ring-sleeve connector, they appear on the tip and the ring.  The signal level is the difference between those two points; that's what you want to measure.  Pin 1, or the sleeve, is ground, and it is not relevant to a balanced audio signal.  It's used for the shield, to help reduce noise.

I'm not sure why you introduced the word "peak."  There are two distinctly different kinds of "peak" in audio.  One of them is shared with electronics in general, i.e., that the peak level is sqrt(2) times the RMS level.  This kind of peak is extremely rare, one might even say "never!", to use such a level in audio.  It's always an RMS level, but... that level is changing all the time, because that what sound does: it gets louder and softer.  So that's the other kind of "peak": when it's loud loud loud.

(There is a third kind of "peak", used by companies that make cheap amplifiers and try to convince you they are powerful enough to blow up a house...  When they say "1000 watts peak power", they are saying "we are lying to you to get you to buy this thing.")

In other words, the idea of "peak" depends on the context.  If we were sitting together, I would ask you where that word came from; what did you see that made you start thinking about it?

As to the specific meter, I think the most helpful thing for you would be to learn to read the specs.  Look at each one, and try to learn what you need to learn to understand what they are saying.  First would be the ranges; look at the voltage ranges for both DC (which might be interesting, but not for your current problem), and for AC (which is very important).  So for AC, you might see "200mV", which is millivolts, the three digits after the decimal point.  Let nothing go unlearned!  Look for an answer for every specification, every measurement.  This sort of thing will serve you in the future as well.

As to your overall project, I think it's an interesting one and will give you lots of things to learn.  But at the same time, levels in audio are pretty much always subjective, especially because music and voice (and other) levels are changing all the time.  It's not "one level", it's an approximate "range of levels."  So getting it close is the best you can do.  You want it in a reasonable range so that your equipment can work properly, so you're not overloading the inputs (too high!) and you're not getting too much noise mixed in (too low!).  Then, you just see how you like it, and push it up or down as appropriate.

And it's always a matter of thinking about how to deal with it.  Suppose, for example, I'm mixing a recording of a band, and sometimes I listen to it on "these" speakers, which take 0dBu for full volume, and sometimes "those" speakers, which take -20dBu for full volume.  When I switch the speakers, I set the master (output) volume of the mixer higher or lower by 20dB, and everything is fine.  More likely what I would do is just slide the slider until it sounds like the level I want.  Generally speaking, that is NOT going to be the maximum volume the speakers are capable of producing, because my music has interesting peaks (there's that word!  This is the "loud loud loud" version) of at least 10dB over the normal, average level.  So now I need to make sure the average level is at least 10dB lower than the peaks, in order to keep the peaks from exceeding the speaker's maximum.  And that will still sound nice and loud, because acoustic levels don't usually have to be anywhere near as extreme as most marketing departments would like you to believe.

About your audio interface, I would bet that it's pretty darn close to what they say it is.   I have never measured any of mine; they just work.  And they work fine.  "More is more," and you turn the knob up, "less is less" and you turn the knob down.  Much more interesting is how much hissy noise comes out when you turn the knob up.

If you're not doing music or other "real" sounds, and instead you're actually doing something technical, then you've got a whole lot of considerations to think and learn about!  Much of the discussion changes.


Thanks for your brilliant reply - so informative and helpful! These days a musician/music producer has to have a solid understanding of their recording equipment if they want to get the best results. I will never be an electrical engineer and I am ok with that -- but I have been a student of audio engineering since buying my first microphone 20 years ago and while I am always learning something new, I want to also understand what some of these more technical terms actually mean in the real world - like input sensitivity, or dBu, etc and so on. I got the Aneng 8008 today and plan on testing my Interface's Outputs a bit later and will reply with my results!
 

Offline VoltzsTopic starter

  • Contributor
  • Posts: 19
  • Country: at
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #48 on: November 30, 2022, 04:37:18 pm »
Hi all! so I just did my first tests with the Aneng 8008 and with a 1kHz sine wave played at -20dBFS my interface's outputs read: 0.754

I know that 1dBu = 0.775 volts

Now I want to convert the 0.754 RMS signal to Peak.

I am absolutely terrible at math (at least I never paid attention in class  :palm:) - so with a calculator how would I convert this? Someone did already post an equation but I do not know how to input that into a calculator?

And one other small side note - the Aneng 8008 has a +/- 1-3% accuracy.

I thank you for your patience!  :-+
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7825
  • Country: us
Re: Measuring Audio Interface Outputs with Multimeter - Need help please
« Reply #49 on: November 30, 2022, 04:40:24 pm »
Now I want to convert the 0.754 RMS signal to Peak.

Press the following keys on your calculator:

0.754 x 1.414 =

But why do you want the peak value?  I don't think it has any meaning in your context.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 
The following users thanked this post: Voltzs


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf