Author Topic: Monitoring audio oscillators on analog oscilloscopes  (Read 5656 times)

0 Members and 1 Guest are viewing this topic.

Offline mrs_audio_boy

  • Contributor
  • Posts: 5
  • Country: de
Monitoring audio oscillators on analog oscilloscopes
« on: September 03, 2016, 03:56:45 pm »
Dear regulars and pros,

I am a beginner and this is my very first post. I have some questions but I probably will confuse things and especially the technical terms, so bear with me.

I am an analog synthesizer enthusiast and in order to learn more about analog sound synthesis, I want to monitor audio signals to both hear and see them to understand them better. In order to "stay true" to the analog world, I bought an analog oscilloscope, the Grundig GO 15Z. It is a 15 Mhz twin-channel, technical details can be found here: http://www.radiomuseum.org/forumdata/upload/d_Grundig_GO15-Z_man.pdf. It was quite cheap but it is working quite well as far as I can tell. As with all analogue equipment, I like how it displays the signals, without those digital pixel "staircases". In order to monitor the audio signals of an Arturia Microbrute analog synthesizer, I hook up the Grundig to the Microbrute's headphones output using a BNC-to-3.5mm adapter cable. When playing back a note in the upper register on the Microbrute, I adjust Grundig's time base switch accordingly; this switch is labelled "s/cm", the values range from 0.1s/cm to 0.3 microseconds/cm. When, for example, I play the highest C, I adjust the switch to 0.1m and I can see very clearly the waveform, with several wavelengths on the Grundig's screen displayed. Usually, I like to display two, three or four wavelengths, to get a clear overview of the audio signal. The signal is displayed very sharply and the picture is very stable, so that I can "study" unhurriedly the different effects of applying the cutoff, resonance or pulse width to the audio signal using the Miscrobrute's knobs. So far so good.

When playing lower notes, the audio frequency goes down, naturally. Thus, to get a proper view of the signal, I have to adjust the time switch accordingly, of course. When playing the lowest C on the Microbrute, I have to set the switch to 3m or 10m to see at least one or three wavelengths, accordingly. But when dialing these values, the horizontal sweep becomes visible. When the time base switch is set from 0.3m to 1m, the picture starts to flicker slightly but visibly; but when switched to 3m, the picture flickers very heavily and strains my eyes. When switched to 10m, you can clearly see how the signal is quickly drawn from left to right. This leads to a very unstable picture and it is very wearisome to look at the Grundig's screen and work with the audio signal. What I actually want to achieve is a stable picture without much (or ideally without any) flickering, regardless of the audio oscillator frequency or magnification on the oscilloscope; i.e. regardless of how high or low the note is and how many wavelengths I want to display. In spite of my ignorance of the relationship of the different oscilloscope adjustment possibilities, I hope to have presented my initial position fairly clearly.

Now, my questions are: is the aforementioned flickering a normal behaviour of all analog oscilloscopes, or is it just a shortcoming of my particular model? Is there any way to get a stable picture, maybe somehow through higher horizontal sweep, independently of how many wavelengths are displayed? Is the aforementioned behaviour caused by the fact that my oscilloscope is "just" a 15 Mhz model? Is there a workaround on my ocsilloscope or do I need another model? If I get an oscilloscope with, let's say, 40 Mhz, will it show me all the audio signals without flickering? Do I need a digital storage oscilloscope, maybe? Also, a general question: What is your experience with visualising audio signals with analog oscilloscopes?


I hope my case and my questions make sense to you. I tried to find answers by watching numerous Youtube videos and reading some forum posts but unfortunately my ignorance couldn't be redeemed. Thus, I am very much looking forward to receive your helpful answers and feedback.


Greetings from Germany
MAB
 

Offline Tom45

  • Frequent Contributor
  • **
  • Posts: 356
  • Country: us
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #1 on: September 03, 2016, 04:23:10 pm »
Welcome to EEVBlog and to oscilloscopes.

At 10 msec per cm, it takes 0.1 seconds to sweep the beam across the 10 cm width of the screen.

The phosphor is lit by the beam and then fades away. The persistence of the trace is the time it takes to fade away. As you are seeing flickering, you know that the phosphor persistence is shorter than 0.1 seconds.

A storage scope, analog or digital, holds the display and in effect gives infinite persistence. Before digital scopes, storage was done in analog with special display tubes. With digital scopes it is done using capture to digital memory.

The other method in analog days was to use a camera to capture the display. Just set the camera's exposure time to something longer than the time to sweep the beam across the width of the display. 0.1 seconds in your example.

Tom
 
The following users thanked this post: slaterk93

Online JPortici

  • Super Contributor
  • ***
  • Posts: 2560
  • Country: it
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #2 on: September 03, 2016, 07:07:13 pm »
there is a tek 7623 on ebay, which is a variable persistence scope. it's not really analog storage, because the information is not acquired and stored into a memory and then displayed.. but it will keep the image formed during the last sweep for a long time (in theory indefinite or practically until you power off the instrument)
i had laid eyes on it but i don't have to feed my tek7000 addition :)
« Last Edit: September 03, 2016, 07:09:18 pm by JPortici »
 

Offline oldway

  • Super Contributor
  • ***
  • Banned!
  • Posts: 2174
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #3 on: September 03, 2016, 07:48:12 pm »
Quote
is the aforementioned flickering a normal behaviour of all analog oscilloscopes, or is it just a shortcoming of my particular model?
Yes, it is a normal behaviour. Most of analog oscilloscopes has P31 phosphor with short persistence. To reduce flickering, you can buy an oscilloscope with longer persistence phosphor P7.
Hameg oscilloscopes with P7 phosphor (long decay) have an orange screen filter instead of a green one for the P31 phosphor.
I don't recomend to buy an analog storage scope because storage CRT's had a very short live expectancy and storage function does not works generally no more.

To have a stable pattern at very low frequencies, you can't select automatic trigger.
You must use manual trigger and adjust the trigger level to have a stable pattern.

For your application, I think a digital oscilloscope would be better then an analog one.
With high sample rates of modern digital oscilloscopes, you will not have those digital pixel "staircases" anymore.
 

Offline mrs_audio_boy

  • Contributor
  • Posts: 5
  • Country: de
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #4 on: September 12, 2016, 11:52:08 am »
Thank you very much for you taking your time to read my lenghty entry and compose your answers. By the way, what I mean by "digital staircases" is that a digital oscilloscope shows the signal on its digital display that inherently is built such that you see pixels (unless the digital device has an expensive display with a very high resolution, I guess). Thus, you see an ugly picture where the sgnal appears pixelized, hence I referred to it as a "digital staircase". So using a digital oscilloscope, all the smoothness of the analog display is lost, also is lost the lovely glowing green (or, depending on the screen, sometimes another color like blue or orange). It is very unfortunate, that my case seems not to be solvable. Did indeed nobody want to monitor audio rate signals on an analog oscilloscope yet, and doing so without straining their eyes? So nobody has ever complained about that flaw ever, diring the the last like 70 years? Wow, I always wanted to be the first in something! Don't get me wrong, I am not trying to be pretentions or something, but it is very strange that one cannot display low-frequency audio oscillator signals in a comfortable manner on an analog oscilloscope. But maybe it's just me, who doesn't understand the relationship between the time switch and the horizontal sweep and that this whole situation is something inherently unchangeable because technically not solveable in any other way than present in my oscilloscope...
 

Z80

  • Guest
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #5 on: September 12, 2016, 12:28:28 pm »
Hi, as said the behaviour of your scope is normal for an old CRO.  If you want a big pretty hi res display, just hook up the headphone output to the line in on your PC and use one of the many free audio analyser programs.  The size and resolution of the trace will be limited only by your monitor.  You will likely also get an FFT spectrum analyser too which will give you much more information about subtle changes to your timbre than a scope trace.  The microbrute is a modern synth, no need to use antique retro tools :)
 

Offline oldway

  • Super Contributor
  • ***
  • Banned!
  • Posts: 2174
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #6 on: September 12, 2016, 12:44:43 pm »
I made some test to compare a crt with P31 phosphor and another one with p7 phosphor.
With P31 phosphor (my Hameg HM605), the minimal frequency without flickering is about 45 Hz (2ms/div).
With P7 phosphor (my Philips PM3233), I can look a signal with frequency as low as 5Hz without any discomfort (20 ms/div).
But for normal use, P7 phosphor is not recomended.
 

Offline mrs_audio_boy

  • Contributor
  • Posts: 5
  • Country: de
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #7 on: September 13, 2016, 10:48:01 am »
But... antique retro tools is exactly what I want, hence the whole post. Of course, I could have used a software oscilloscope; but then again, I also could have used a software synth instead of the Arturia Microbrute hardware. But I explicitely want to use hardware and analog equipment. Thank you for your answer nonetheless.

Thank you for the tests on your oscilloscopes, oldway. In the near future, since they seem to do the job, I will try to aqcuire the Hameg HM605 and/or the Philips PM3233 and compare them to my Grundig GO 15Z. Three last questions:
- Why is P7 phosphor is not recomended for normal use, is P7 phosphor toxic or something?
- Also, with "normal use" I presume you mean something that an electical or electronical technician would use to monitor high-frequency signals; so I am not a "normal user", since I want to display low-frequency audio signals, am I getting you right?
- What does actually the Mhz value of an oscilloscope mean? In other words, what will I gain (or maybe lose), when I would switch from a 15 Mhz model to a, let's say, 40 Mhz oscilloscope?
« Last Edit: September 13, 2016, 10:54:49 am by mrs_audio_boy »
 

Online JPortici

  • Super Contributor
  • ***
  • Posts: 2560
  • Country: it
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #8 on: September 13, 2016, 11:43:24 am »
More analog bandwidth: you will be able to see signals with a higher frequency content (the front end is inherently a filter . the -3db point in a 40 MHz scope will be at at least 40 MHz)

while i never used a scope with a P/ cathod tube i think that it would be more problematic for faster signals because of its persistence (jitter and noise buried in a thicker trace)
 

Offline oldway

  • Super Contributor
  • ***
  • Banned!
  • Posts: 2174
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #9 on: September 13, 2016, 07:58:11 pm »

We can distinguish Hameg's oscilloscopes with P7 phosphorus because they have an orange screen filter instead of a green one.
On Philips oscilloscopes, there is no way to distinguish them, I have not even found any inscription on the oscilloscope specifying that it has a P7 crt.

I only have discovered that my Philips PM3233 had a P7 crt after having bought.

The PM3233 has a 2 beams crt (when most of the tubes are single beam), which allows, combined with the P7 phosphorus, to visualize two simultaneous signals at very low time base speed.
So far I know, the PM3233 with P7 phosphorus are very rare.

Long decay P7 is not recommended for normal use because the afterglow is very troublesome in normal use.
When the signal changes, or that you change the settings, the previous image remains some time on the screen, it's very annoying.

I would say that the fact you want to display very low frequencies just puts you in a special situation.

My PM3233 was used for adjusting the stability of PID regulation where there are very large time constants, without having to use a storage oscilloscope.

It was also a special job.

As you measure only audio frequencies (20 to 20Khz), an oscilloscope of 10 or 20Mhz is more than enough.
To properly reproduce a square wave requires a bandwidth of at least 10x the frequency of the square wave.
So you need a bandwidth of at least 200Khz.
No need of a high frequency oscilloscope... :popcorn:

 
The following users thanked this post: mrs_audio_boy

Offline mrs_audio_boy

  • Contributor
  • Posts: 5
  • Country: de
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #10 on: September 14, 2016, 04:03:14 pm »
Thank you once again, oldway! Your comments were very helpful so far. I will refrain from acquiring the Philips PM3233, since the P7 phosphor seems to bring even more speciality to the table than my case. I will purchase the Hameg HM605, if I'll manage to get a fairly cheap deal on ebay and will report on that here, if this thread will still exist by then.

Thank you all and keep that electricity flowing! :D
 

Offline macboy

  • Super Contributor
  • ***
  • Posts: 1952
  • Country: ca
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #11 on: September 14, 2016, 04:55:25 pm »
mrs_audio_boy
Think about it this way. Say you are viewing a 20 Hz waveform. You said want to have 4 of them on screen at a time. A 20 Hz waveform only repeats 20 times per second by definition, and if you want 4 copies onscreen, then that only repeats 5 times per second. An analog scope literally directly draws the waveform to the screen as it happens, so the maximum rate at which the scope can re-draw the screen is in fact 5 Hz. A typical CRT computer monitor refreshes at >= 75 Hz for flicker-free viewing. So 5 Hz is very flickery.

You may not mind too much a digital scope which has some kind of "enhanced resolution" feature. This will allow the scope to sample at a much higher rate than necessary for a slow signal, and average each N extra samples into a single higher resolution sample. The result is much higher vertical resolution, sometimes 12 bit or higher, rather than 8 bit. This can result in a very nice smooth looking waveform for slow signals (i.e. any frequency audio signal).
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 1874
  • Country: us
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #12 on: September 15, 2016, 04:30:14 am »
The easiest person to fool is yourself. -- Richard Feynman
 

Offline oldway

  • Super Contributor
  • ***
  • Banned!
  • Posts: 2174
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #13 on: September 15, 2016, 08:23:53 am »
Even if HM605 is, in my opinion, one of the best Hameg analog oscilloscope (it is my workhorse for a very long time, i have buyed it in 1982 ... ), there are other very good oscilloscopes with lower bandwith, the HM203 or HM204, who could fulfill your needs.
 

Offline dentaku

  • Frequent Contributor
  • **
  • Posts: 831
  • Country: ca
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #14 on: September 15, 2016, 12:48:36 pm »
But... antique retro tools is exactly what I want, hence the whole post. Of course, I could have used a software oscilloscope; but then again, I also could have used a software synth instead of the Arturia Microbrute hardware. But I explicitely want to use hardware and analog equipment. Thank you for your answer nonetheless.

Thank you for the tests on your oscilloscopes, oldway. In the near future, since they seem to do the job, I will try to aqcuire the Hameg HM605 and/or the Philips PM3233 and compare them to my Grundig GO 15Z. Three last questions:
- Why is P7 phosphor is not recomended for normal use, is P7 phosphor toxic or something?
- Also, with "normal use" I presume you mean something that an electical or electronical technician would use to monitor high-frequency signals; so I am not a "normal user", since I want to display low-frequency audio signals, am I getting you right?
- What does actually the Mhz value of an oscilloscope mean? In other words, what will I gain (or maybe lose), when I would switch from a 15 Mhz model to a, let's say, 40 Mhz oscilloscope?

I use a Hameg HM605 and 2ms is the slowest I can set it before it flickers.
That's just the way analog scopes work. For example, I'm currently working with an analog kickdrum circuit and to get the entire sound with the whole decay tail I need to set it to 20ms/div so it's very flicker and there's no way around it.

The Microbrute looks like a fun synth by the way. The new Behringer DeepMind12 looks even nicer.
When it comes to scopes I'd still rather have a modern digital one.

Really old analog scopes didn't even have a trigger so it was very hard to get a waveform to even stand still on the screen.
W2AEW Alan Wolke has a nice video explaining analog scopes.
« Last Edit: September 15, 2016, 12:50:50 pm by dentaku »
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 4794
  • Country: au
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #15 on: September 15, 2016, 03:34:41 pm »
Dear regulars and pros,

I am a beginner and this is my very first post. I have some questions but I probably will confuse things and especially the technical terms, so bear with me.

I am an analog synthesizer enthusiast and in order to learn more about analog sound synthesis, I want to monitor audio signals to both hear and see them to understand them better. In order to "stay true" to the analog world, I bought an analog oscilloscope, the Grundig GO 15Z. It is a 15 Mhz twin-channel, technical details can be found here: http://www.radiomuseum.org/forumdata/upload/d_Grundig_GO15-Z_man.pdf. It was quite cheap but it is working quite well as far as I can tell. As with all analogue equipment, I like how it displays the signals, without those digital pixel "staircases". In order to monitor the audio signals of an Arturia Microbrute analog synthesizer, I hook up the Grundig to the Microbrute's headphones output using a BNC-to-3.5mm adapter cable. When playing back a note in the upper register on the Microbrute, I adjust Grundig's time base switch accordingly; this switch is labelled "s/cm", the values range from 0.1s/cm to 0.3 microseconds/cm.


When, for example, I play the highest C, I adjust the switch to 0.1m and I can see very clearly the waveform, with several wavelengths on the Grundig's screen displayed. Usually, I like to display two, three or four wavelengths, to get a clear overview of the audio signal.


""Wavelength" is the wrong terminology----what you are referring to are "cycles".
The duration of each cycle is that time it takes between a point on a waveform,& the equivalent point on the next cycle.This is normally expressed as between the first ,&  third zero crossings of a sinewave.
A "wavelength" is the distance a signal will travel in a given time.
For Electromagnetic waves this is specified  as in free space,but in air the differece is negligible.
Acoustic waves are different,with the wavelength varying with the media-it is different in air,compared to water,for instance
Quote
The signal is displayed very sharply and the picture is very stable, so that I can "study" unhurriedly the different effects of applying the cutoff, resonance or pulse width to the audio signal using the Miscrobrute's knobs. So far so good.

When playing lower notes, the audio frequency goes down, naturally. Thus, to get a proper view of the signal, I have to adjust the time switch accordingly, of course. When playing the lowest C on the Microbrute, I have to set the switch to 3m or 10m to see at least one or three wavelengths,cycles accordingly.
I assume this is C1 of the equal tempered scale,which is 32.70Hz.
Quote
But when dialing these values, the horizontal sweep becomes visible.
This is a very strange comment,as the horizontal scan is always visible---that is what you look at!
Is it possible that you have the 'scope set to "alternate sweep,& are seeing the other channel?
In AC coupled mode,this may appear in line with the centre line of your display.
"Alternate" will present one channel's output at a time,& will cause quite a lot of flicker"---that is why "chop" is used to display both channels at low frequency settings.
Quote

 When the time base switch is set from 0.3m to 1m, the picture starts to flicker slightly but visibly; but when switched to 3m, the picture flickers very heavily and strains my eyes.
At 1ms/div there should not be any visible flicker,& at 3ms/div,flicker should be barely discernible.
Quote

When switched to 10m, you can clearly see how the signal is quickly drawn from left to right.
Taken literally,this should not be evident until very much slower speeds are used.
Quote
This leads to a very unstable picture and it is very wearisome to look at the Grundig's screen and work with the audio signal. What I actually want to achieve is a stable picture without much (or ideally without any) flickering, regardless of the audio oscillator frequency or magnification on the oscilloscope; i.e. regardless of how high or low the note is and how many wavelengths cycles.



Now, my questions are: is the aforementioned flickering a normal behaviour of all analog oscilloscopes,
Yes,but not to anything near the extent you are experiencing.
Quote
or is it just a shortcoming of my particular model? 
Perhaps,or as suggested above,it may be an artefact of how you are using it.
Quote
Also, a general question: What is your experience with visualising audio signals with analog oscilloscopes?


Greetings from Germany
MAB

Analog Oscilloscopes were,& are, used widely to look at low audio frequencies,as well as others using the same time/div ranges.
In Analog TV,it was common to look at 2 fields of the video signal,requiring for 625 line systems,a total of 40ms scan time..
With most 'scopes  the most convenient setting is 5ms/div,with a total scan time of 50ms.

I have done this many times without experiencing any discomfort due to flickering.

So as to not rely upon memory,I set up the following test gear:

Tektronix 7613 Oscilloscope with 7A18 vert & 7B53A horizontal plugins.
As the 7613 is a storage 'scope (although I would only be using it in "non-storage" mode),I also included a
Jaycar/Digitech 10MHz single channel 'scope to "keep the Tek honest".

Signals were supplied by a Levell TG200D RC Oscillator & its frequency verified using a
Victor VC3651 counter.

Testing the 7613 with a 32.70 Hz signal, at 2ms/div,5ms/div,& 10ms/div, a stable display was obtained ,with perhaps a small
amount of flicker (it looked more like amplitude "jitter")

Stable triggering was obtained at frequencies down to the point where the display was just a moving dot.
At no time was the display objectionable enough to hurt my eyes.

The test was repeated with the Digitech with similar results.
 
The following users thanked this post: mrs_audio_boy

Offline mrs_audio_boy

  • Contributor
  • Posts: 5
  • Country: de
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #16 on: September 26, 2016, 05:56:17 am »
Thanks again to you all for your helpful comments! I will experiment more with oscilloscopes in the near future to gain experience in this topic.
 

Offline w2aew

  • Super Contributor
  • ***
  • Posts: 1729
  • Country: us
  • I usTa cuDnt speL enjinere, noW I aR wuN
    • My YouTube Channel
Re: Monitoring audio oscillators on analog oscilloscopes
« Reply #17 on: September 26, 2016, 04:12:12 pm »
But... antique retro tools is exactly what I want, hence the whole post. Of course, I could have used a software oscilloscope; but then again, I also could have used a software synth instead of the Arturia Microbrute hardware. But I explicitely want to use hardware and analog equipment. Thank you for your answer nonetheless.

Thank you for the tests on your oscilloscopes, oldway. In the near future, since they seem to do the job, I will try to aqcuire the Hameg HM605 and/or the Philips PM3233 and compare them to my Grundig GO 15Z. Three last questions:
- Why is P7 phosphor is not recomended for normal use, is P7 phosphor toxic or something?
- Also, with "normal use" I presume you mean something that an electical or electronical technician would use to monitor high-frequency signals; so I am not a "normal user", since I want to display low-frequency audio signals, am I getting you right?
- What does actually the Mhz value of an oscilloscope mean? In other words, what will I gain (or maybe lose), when I would switch from a 15 Mhz model to a, let's say, 40 Mhz oscilloscope?

I use a Hameg HM605 and 2ms is the slowest I can set it before it flickers.
That's just the way analog scopes work. For example, I'm currently working with an analog kickdrum circuit and to get the entire sound with the whole decay tail I need to set it to 20ms/div so it's very flicker and there's no way around it.

The Microbrute looks like a fun synth by the way. The new Behringer DeepMind12 looks even nicer.
When it comes to scopes I'd still rather have a modern digital one.

Really old analog scopes didn't even have a trigger so it was very hard to get a waveform to even stand still on the screen.
W2AEW Alan Wolke has a nice video explaining analog scopes.


Here is a slightly shorter version of the same basic talk that I presented a few months ago:
======================================
YouTube channel: https://www.youtube.com/w2aew
FAE for Tektronix
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf