IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Plugging an oscilloscope to a mixer, possible use to detect audio flaw
Tyrexionibus
post Nov 14 2013, 03:23
Post #1





Group: Members
Posts: 27
Joined: 14-October 13
Member No.: 110929



It's been more than 10 years since I heard about a DJ who tried to plug his mixer's RCA (master out) outputs to a 2-channel oscilloscope.
This is in fact completely possible since the oscilloscope has standard N-type/B-type coax connections BUT they can be converted to an RCA one using adapters.

The idea has become so stuck in my mind that now I wanna try it out.

What am I wanting to obtain?

The scope, in the vertical section, controls the Volts-per-Division (V) graph. In the horizontal one, the time (in seconds, s) plus an optional X-Y graph viewer.
Now: since the audio signal that quits the mixer is, to all intents and purposes, an electrical signal, by inputting it in the scope I would be able to view the audio wave flowing up and down during a live set.

The meaning of this is to detect flaws in the signal like excessive distortion, or bad equalization. (Referred to flaws as the signal going out from the mixer, not put out of the loudspeakers in the club!)

In fact, some time ago I saw a picture which showed that if distortion was absent, the waveform was sine-like as it should be (unless other schemes had been selected). Furthermore, if distortion was present, the waveform was sine-like only partially: the distortion caused in fact the WF to be displayed as flat (completely) in some points - this was a sign that something was not good. The more flat parts were present, the more distortion there was.

I need your help to clarify some things. First, does the oscilloscope show a fixed image that changes only if triggered (by pressing a button, for example) or a continuously flowing one?

Plus.. I have a quote from another source.

"Assuming the scope has at least two channels to display. If you feed the test signal to one channel, and the output to the other channel, then adjust the two scope channels for the same size traces. Now invert one channel, and set the scope to ADD. If the two signals are exactly the same, no distortion, then the inverted should exactly cancel the non-inverted, leaving a trace at zero. Any difference between the input and output would result in the two not exactly cancelling, and the result would be a display on screen of the remaining distortion itself.

(So the scope would be completely useless this way)

Another technique would be to put the scope into X-Y mode. Put the test signal on the vertical, and the amp output on horizontal. If the signals are exactly the same, you get a slanted straight line on the screen. Adjust the amplitude or the scope sensitivity to make the line a 45 degree slant. If the output is not identical, any distortion or phase difference will result in the slanting line spreading into various shapes.

With this method you can look at stereo signals. If you applied left and right to X and Y axes, and fed it mono, you got a straight slanted line. But when the channel signal differed, you got a more chaotic picture."

Is this.. right? lol

Any help is greatly appreciated.
Tyrexionibus
Go to the top of the page
+Quote Post
saratoga
post Nov 14 2013, 04:10
Post #2





Group: Members
Posts: 4718
Joined: 2-September 02
Member No.: 3264



If you want to learn how to use a scope, you can do this, but you won't be able to see anything all that useful in terms of audio quality. Scopes just show you the time domain waveform output by sound card with a very low dynamic range. Unless your sound card is extremely bad, the output will look essentially identical to what you see on your PC.

Regarding your questions:

QUOTE
The meaning of this is to detect flaws in the signal like excessive distortion, or bad equalization. (Referred to flaws as the signal going out from the mixer, not put out of the loudspeakers in the club!)


No you won't be able to do this aside from huge distortions like severe clipping (which you can much more easily recognize with your ears!). If you're interested in measuring effects like this, use RMAA and your sound card's line in.

QUOTE
Is this.. right?


Assuming your scope supports those features, you could try and make those measurements, however the dynamic range on a scope is very limited, so you will most likely not be able to measure all but the most extreme distortion. Scopes are made for measuring very large bandwidths (e.g. RF) at moderate to low dynamic range (a few tens of dB). If you want to measure audio signals (50-100dB at very low bandwidth), you should use a sound card. They're made for recording audio.

Go to the top of the page
+Quote Post
Glenn Gundlach
post Nov 14 2013, 08:54
Post #3





Group: Members
Posts: 359
Joined: 19-April 08
From: LA
Member No.: 52914



The scope will show you clipping and that's about all. It's not a spectrum analyzer so it won't give you insight on EQ or any other spectral parameter.

If you haven't done so, download Audacity (it''s free) and look at audio waveforms as that is what you would see (with LESS detail) on your scope.

G
Go to the top of the page
+Quote Post
dhromed
post Nov 14 2013, 16:41
Post #4





Group: Members
Posts: 1244
Joined: 16-February 08
From: NL
Member No.: 51347



If you're hearing excessive distortion, and you are seeing excessive distortion on the scope, at least you can sort of tell where in the amplification path the distortion is coming from. But beyond that, it's not much use. It would be like looking at a weather website to see if it's raining outside.

This post has been edited by dhromed: Nov 14 2013, 16:42
Go to the top of the page
+Quote Post
DVDdoug
post Nov 14 2013, 19:14
Post #5





Group: Members
Posts: 2441
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



Wikipedia has a nice picture of clipping on an oscilloscope. But, that's NOT audio (music or speech), it's a pure sine wave test tone (well, it was a sine wave before it was clipped).

A 'scope works something like a video or a movie with a series of still images, except they are not really still, and if you slow-down the horizontal sweep you can see the waveform being "drawn". The way you get an image like you see in the Wikipedia picture is to feed-in a repetitive waveform and trigger the sweep at the same point on the waveform every time, so that all of the repeated images are the same. Or, if you have a digital 'scope you can capture & freeze the image.

If you search YouTube, I would guess you can find some video of what audio looks like on a 'scope. It basically looks like a blurry mess. Or, with a digital 'scope you freeze the image so it looks more like what you see an audio editor.

An oscilloscope is handy for troubleshooting (I use one every day at work for non-audio stuff), but most of the time you are looking for rather gross errors, like signal or no-signal, or the wrong level on a pulse, etc... If you are working on an amplifier with distorted output (distorted badly enough to see on a 'scope), you could run a test-tone through it and check at various places on the circuit board to find where the distortion is being introduced.

I was just working on a non-working digital board, and I found some address lines going between zero and about 1V, instead of up to 5V. (It was a bad CPU chip.) If you have two digital data or address lines shorted together, you' usually get a random stair-step waveform, instead of switching cleanly between zero and 5V. Since the data lines are not repetitive, a digital 'scope that can hold the waveform is very handy for that kind of thing.

This post has been edited by DVDdoug: Nov 14 2013, 19:15
Go to the top of the page
+Quote Post
Tyrexionibus
post Nov 18 2013, 03:17
Post #6





Group: Members
Posts: 27
Joined: 14-October 13
Member No.: 110929



I had already seen the picture of a distorted sinewave on an oscilloscope, that's been the inspiration of this thread.

There's another, which shows very little clipping and flat zones - in fact it's to be zoomed: http://mwpersons.com/articles/2013/oscillo...%20clipping.jpg

I would like to try out a spectrum analyzer, but they are so expensive! Cost lies on bandwidth: the wider the bandwidth, the higher the cost - and in fact the Agilent-HewlettPackard analyzers which range from 9 kHz to 3 GHz are priced $9,000 + accessories.

For an audio mixer (to be used in a live set), I guess the requirements for bandwidth are determined by the sum of these rules:
- To range at least between the mininum and the maximum frequencies hearable by the human ear - 20 Hz to 20 kHz (the major rule of psychoacoustics)
- To range to the maximum sampling frequency used by the mixer

(Is this right?!)

The first frequency range is 20 Hz to 20 kHz and that's about all.
The second frequency range depends on the mixer: I have a DENON DN-X1700 which has a maximum of 96 kHz as sampling frequency (which I personally never use for live sets as it sucks lots of resources on the Mac running Ableton Live and it also heats up a lot the mixer). I use 44,1 kHz or 48 kHz sometimes.

Let's assume the maximum SF - I'd need a spectrum analyzer which has a 20 Hz - 96 kHz range. Do they even exist? 20 Hz is extremely low, and I'd guess only a very advanced spectrum analyzer used for special lab purposes would be capable of detecting it. Such instrument would have a very high upper bandwidth limit and would cost a lot.

Am I right, again? smile.gif

This post has been edited by Tyrexionibus: Nov 18 2013, 03:18
Go to the top of the page
+Quote Post
saratoga
post Nov 18 2013, 03:48
Post #7





Group: Members
Posts: 4718
Joined: 2-September 02
Member No.: 3264



QUOTE (Tyrexionibus @ Nov 17 2013, 22:17) *
Let's assume the maximum SF - I'd need a spectrum analyzer which has a 20 Hz - 96 kHz range. Do they even exist?


In this range dedicated spectrum analyzers are generally not used because a simple ADC will work much better. Do you have a sound card? That would be the ideal tool.

Spectrum analyzers are used for frequency ranges were ADCs are very expensive or impractical (e.g. GHz). At lower frequencies the annoyance of using an analyzer is unnecessary because ADCs exist.

Go to the top of the page
+Quote Post
DVDdoug
post Nov 18 2013, 23:14
Post #8





Group: Members
Posts: 2441
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



QUOTE
I had already seen the picture of a distorted sinewave on an oscilloscope, that's been the inspiration of this thread.
That's a (distorted) sine wave. Before you buy an oscilloscope, it would be a good idea to see what constantly-changing voice or music looks like.

A quick search turned-up Winscope, a FREE soundcard/software oscilloscope.

Apparently, it looks at the soundcard input, so in order to see a waveform for what I was playing, I had to go to the Windows Control Panel and select Stereo Mix as my recording input. Not all soundcards support Stereo Mix, but that's one way to record what's coming out of your computer speakers. If you have a desktop computer with line-in, you can connect your mixer and use Winscope.

It doesn't exactly look like a real oscilloscope because the triggering is a bit slow (I didn tplay-around too much with the adjustments), but it is something like you'd see on a digital oscilloscope. (And, quite a bit different from what you'd see on an analog 'scope... On an analog 'scope a non-repetitive waveform would never "freeze").

A software/soundcard oscilloscope isn't going to replace a real 'scope, but it should give you an idea of the limitations of looking at a real-time audio waveform.

QUOTE
For an audio mixer (to be used in a live set), I guess the requirements for bandwidth are determined by the sum of these rules:
- To range at least between the mininum and the maximum frequencies hearable by the human ear - 20 Hz to 20 kHz (the major rule of psychoacoustics)
- To range to the maximum sampling frequency used by the mixer

(Is this right?!)
For casual audio use 20-20kHz is fine.

The audio can only go up to half the sample rate, so with a 96kHz sample rate, your signal can't go above 48kHz. (To over-simplify it a bit, you need to sample the top-half of the waveform at least once and the bottom-half of the waveform at least once. So, that's at least twice per cycle.)

For more serious troubleshooting or analysis, you'd like to know if you have 100kHz (or 1MHz, etc.) noise or ringing or in your signal. It's also good to check for any DC (zero Hz) offset.

This post has been edited by DVDdoug: Nov 18 2013, 23:35
Go to the top of the page
+Quote Post
Tyrexionibus
post Nov 19 2013, 00:44
Post #9





Group: Members
Posts: 27
Joined: 14-October 13
Member No.: 110929



Today I went to my electrician and talked about the oscilloscope on an audio mixer. He said that the configuration I was wanting is impossible to do, because:

"Plugging the mixer's RCA output to the oscilloscope's (coax-to-)RCA input is impossible. I won't obtain any results. I would obtain if I plugged a frequency generator to the RCA input of the mixer, and then the RCA output to the 'scope. This would detect "what frequencies the mixer is unable to transmit well, and thus distorts". Such a configuration is impossible to do without a very advanced training"

I think he's misunderstood what I want. If I plug a frequency generator I would obtain many test tones - tones that the oscilloscope would display. OK, this reveals internal troubles in the mixer. But I don't want this; I want to view the audio going out from Ableton Live in the oscilloscope, and view if it's distorted. I don't need a test tone of a frequency generator.

Anyway testing it software-side is good. Winscope.. I think it's for Windows only, and I'm on OS X 10.7.5 Lion. Don't worry about the sound card: my mixer is a DENON DN-X1700 which has a built-in SC which supports stereo mixing at 96 KHz maximum with 8 channels (mono; or 4 stereo channels), via USB. It's excellent, as it even shrinks the resources sucked up by Ableton Live. In fact OS X, when detecting an external sound card, "commissions" it to all audio management and thus the CoreAudio built-in SC, in a situation like this, is used only for the signal to pass thru.
Go to the top of the page
+Quote Post
DVDdoug
post Nov 19 2013, 01:45
Post #10





Group: Members
Posts: 2441
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



QUOTE
Today I went to my electrician and talked about the oscilloscope on an audio mixer. He said that the configuration I was wanting is impossible to do, because:

"Plugging the mixer's RCA output to the oscilloscope's (coax-to-)RCA input is impossible. I won't obtain any results.
Your average electrician doesn't carry around an oscilloscope. It's more of a tool for an electronics technician or engineer than for someone who works with electrical power. (I use one every day and there is one in front of me on my workbench right now.)

I do agree that you may not get useful results.

I have some RCA - BNC Adapters and I've done it!!! You will see a signal/waveform! Even without an adapter, you can clip the 'scope probe to a male RCA connector. (Recently, I actually connected-up a soundcard-output, not a mixer.) I did have one small problem... Without a load (other than the high impedance oscilloscope input) there was DC-drift in my signal. IIRC the signal drifted-up until the signal was centered around +5VDC, instead of being centered around zero. This is caused by leakage through the soundcard's output capacitors, and a ~10K resistor across the soundcard outputs took care of it.

BTW - You can generate "test tone" WAV files with an audio editor and play them on your soundcard, in place of a signal generator.

This post has been edited by DVDdoug: Nov 19 2013, 01:47
Go to the top of the page
+Quote Post
saratoga
post Nov 19 2013, 03:03
Post #11





Group: Members
Posts: 4718
Joined: 2-September 02
Member No.: 3264



I really cannot make this more clear:

You do not want to use a scope.

The software tool you want to use for this is RMAA.

The test can be run from a Mac, but to view the results you will need a PC.
Go to the top of the page
+Quote Post
Tyrexionibus
post Nov 25 2013, 03:14
Post #12





Group: Members
Posts: 27
Joined: 14-October 13
Member No.: 110929



That was clear, a "physical" scope is not required for this purpose. The only use I meant for this was during a live set, because during such a performance the computer's display is already busy for.. err.. Ableton Live smile.gif

Anyway thanks for the info, I will check out RMAA smile.gif
Go to the top of the page
+Quote Post
RonaldDumsfeld
post Nov 25 2013, 15:24
Post #13





Group: Members
Posts: 310
Joined: 12-June 09
Member No.: 70617



What audio interface are you using?

Anything half decent ought to come with software able to display your signal in numerous informative ways. FFT analysist, oscilloscope, X-Y plot, phase analysis etc etc.

If you are short on display real estate it will be easier to carry around a second screen rather than an oscilloscope. You might even be able to convince the promoter to put your stream through the house projection system.



Go to the top of the page
+Quote Post
Gecko
post Nov 25 2013, 17:51
Post #14





Group: Members
Posts: 934
Joined: 15-December 01
From: Germany
Member No.: 662



What exactly is the fundamental problem? Do you suspect that some of your hardware is broken? Are people complaining about distorted sound?

Keep your meters in the green and watch the clip indicators of the amplifiers. To reiterate what others have said: watching audio on an oscilloscope is not very informative.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 19th April 2014 - 06:14