Oscilloscope question

[fm101]

I have a new digital storage oscilloscope, utd2102cex which says 100MHz bandwidth and 1GS/s sampling rate. I am not sure whether it is working correctly or is malfunctioned in which case I will have to return it. The problem is I measured 10KHz square wave output generated from audio soundcard with the oscilloscope the waveform does not appear square wave which is shown below.


I tried functionality provided by the oscilloscope(https://www.manualslib.com/products/Uni-T-Utd2102cex-8824295.html) but can't get a square wave waveform. I tried using the ground spring(usually used for RF range) shown below and I guess 10KHz square wave should be easily drawn with/without ground spring.


But at frequencies below 2KHz the signals(sawtooth, triangle, square wave) appears as expected. Below is 1KHz square wave on DSO.


Does it look like the oscilloscope is malfunctioned because 10KHz is not that high frequency and should be easily sampled and plotted by the DSO. If it is I have to return it immediately. Any advice on this matter.

Thank you

 

[wwfeldman]

you need to calibrate the probe

there should be a connetion on the scope with a standard signal
connect the probe and ground
and adjust the probe

it should be described in the manual

 

[fm101]

I have already done that and it is good. The standard signal which is 3V, 1KHz from the oscilloscope probe com is ok and accurate.

 

[KlausST]

Hi,

The problem is I measured 10KHz square wave output generated from audio soundcard

An audio soundcard is for audio signals. Not square wave.
It maybe is limited to 20kHz maybe a bit higher.

So what is a 10kHz square wave?
It consists of;
* fundamental sine, frequency 10kHz
* plus 3 times fundamental frequency sine = 30kHz (reduced amplitude)
* plus 5 times fundamental frequency sine = 50kHz (reduced amplitude)
* plus 7 times ...
* plus 9 times ...
* plus 11 times ...
... and so on. infinitely

so if the audiocard is limited to something above 20kHz ... it will be hard to get 30kHz out of it.
and surely not 50kHz or higher.

--> So what you get is the 10kHz sine, maybe a bit "distorted" by the 30kHz.

Use a square wave generator when you want a square wave.

****
For sure a 100Hz square wave generated from an audio sound card may look similar to a square wave:
Because the overtones of 300Hz, 500Hz, 700Hz, 900Hz ... below 20kHz are quite good to reperoduce.

Klaus

 

[FvM]

I agree with KlausST. Rather unlikely that a soundcard can generate true 20 kHz square wave. The output is low pass filtered to audio band, e.g. 20 or 25 kHz cut off frequency. The apparently non-periodic waveform is probably caused by the sampling frequency, could be 96 or 192 kHz.

 

[danadakk]

The sound card output is AC coupled. Your first scope capture seems like it has a low freq modulation
on it. Your scope AC or DC coupled on vertical ?

PC SOUND-CARD SCOPE INTERFACE FACILITATES DC RESTORATION

PC SOUND-CARD SCOPE INTERFACE FACILITATES DC RESTORATION: To start with I must appreciate acknowledge and thank ‘Christian Zeitnitz’ for the fantastic Sound-card Oscilloscope & signal generator software available at his website on which this Instructable is based. While working with the PC sound…

www.instructables.com

Regards, Dana.

 

[fm101]

Hi,

An audio soundcard is for audio signals. Not square wave.
It maybe is limited to 20kHz maybe a bit higher.

So what is a 10kHz square wave?
It consists of;
* fundamental sine, frequency 10kHz
* plus 3 times fundamental frequency sine = 30kHz (reduced amplitude)
* plus 5 times fundamental frequency sine = 50kHz (reduced amplitude)
* plus 7 times ...
* plus 9 times ...
* plus 11 times ...
... and so on. infinitely

so if the audiocard is limited to something above 20kHz ... it will be hard to get 30kHz out of it.
and surely not 50kHz or higher.

--> So what you get is the 10kHz sine, maybe a bit "distorted" by the 30kHz.

Use a square wave generator when you want a square wave.

****
For sure a 100Hz square wave generated from an audio sound card may look similar to a square wave:
Because the overtones of 300Hz, 500Hz, 700Hz, 900Hz ... below 20kHz are quite good to reperoduce.

Klaus

hmm, I don't know whether this will lead to the solution here.
Audio signals are just signals with frequencies (below 20KHz approx.) which can be heard. The audio signal shape can be anything square, triangle etc. I think soundcard can generate square square without any harmonics. I have used soundcard oscilloscope software and generator frequently which accurately plotted the same 10KHz square wave from soundcard, square wave below 20KHz from op-amp square wave generator, from arduino, from mobile phone function generator app etc.

--- Updated Feb 14, 2022 ---

I agree with KlausST. Rather unlikely that a soundcard can generate true 20 kHz square wave. The output is low pass filtered to audio band, e.g. 20 or 25 kHz cut off frequency. The apparently non-periodic waveform is probably caused by the sampling frequency, could be 96 or 192 kHz.

I agree that soundcard cannot generate 20KHz square wave upto required precision a 10KHz it does.

I also think that somewhere @KlausST makes a point that there is something wrong with the soundcard generated 10KHz signal which I am looking into.

--- Updated Feb 14, 2022 ---

The sound card output is AC coupled. Your first scope capture seems like it has a low freq modulation
on it. Your scope AC or DC coupled on vertical ?

PC SOUND-CARD SCOPE INTERFACE FACILITATES DC RESTORATION

PC SOUND-CARD SCOPE INTERFACE FACILITATES DC RESTORATION: To start with I must appreciate acknowledge and thank ‘Christian Zeitnitz’ for the fantastic Sound-card Oscilloscope & signal generator software available at his website on which this Instructable is based. While working with the PC sound…

www.instructables.com

Regards, Dana.

I tried both AC and DC coupling, same result. BTW, I used the PC sound card scope(and others) provided in your instructable url link.

 

[KlausST]

HI,

The audio signal shape can be anything square, triangle etc.

True .. and not. While misic is a mixture of different sounds with different overtones. ..it´s not able to reproduce a square wave perfectly. It can be close to it. Your 1kHz shows this perfectly. It is close to a square wave, but not perfect.
look at this:

You look other videos on "square wave overtones" or similar...

I think soundcard can generate square square without any harmonics.

This is self-contradictory.
A signal without harmonics is a pure sine wave. No other waveform.
A square wave needs harmonics to become a square wave.

While you can build fast analog amplifiers that can reproduce almost perfect 10kHz square wave ... this is not possible with digital systems or digital-analog systems. The limitation is because of Nyquist.

I don´t know your sampling frequency, thus I just use the standard CD sampling frequency of 44100Hz.
So with fs of 44100Hz you never can get an unequal harmonic overtone of 10kHz.

The first overtone is 30000Hz but the theoretical limit is below 22050Hz. No! zero! 30kHz amplitude.

Klaus

 

[fm101]

ok, i checked and square wave at 10KHz from arduino is good on the DSO. @KlausST was right in that the soundcard produces square wave with harmonics. It becomes significant after certain frequencies of about above 2KHz. I was sure that it was square wave from soundcard and didn't even think to check with actual square wave generator. At 5KHz the square wave output looks like square wave but with overshoots(harmonics). But the sine wave from soundcard well above 10KHz appears sine wave.

thanks @KlausST, thank you all
and thanks that the new DSO is ok

 

[c_mitra]

I think soundcard can generate square square without any harmonics.

It is not possible to have a periodic waveform (other than a sine wave) without any harmonics.

Let us assume (just for the sake of arguments) that your output is clipped at 20kHz. Sharp cut-off.

you can program it to produce a triangle or square wave or a saw tooth, but all will look like a sine wave.

Because sine wave is the only basis that do not have any harmonics.

In reality, the cut-off at 20kHz is not sharp; say the slope is 20db per decade (frequency).

In such a case, higher harmonics will be passed but attenuated. Hence you will see a distorted sine wave.

Look about the output you are getting:

every third and fifth wave will be distorted. But every 15th will be different.

Using a square wave and a tuned bandpass filter, you can produce sine waves of any odd frequency of the original square wave.

 

[KlausST]

Hi,

In reality, the cut-off at 20kHz is not sharp; say the slope is 20db per decade (frequency).

I'd say in most cases thus is true,
But especialy audio soundcards are rather optimized with their oversampling technique and the combination of digital and analog filters.
A 44kHz sampling system may have a passband ripple of less than 0.1dB up to 20kHz but then 60dB of attenuation at 22kHz.
This is way more than 20dB/decade.
This is about impossible for analog only filters..

But as said, this is true only for the audio ADCs and DACs...


Klaus

 

[c_mitra]

This is about impossible for analog only filters..

But as said, this is true only for the audio ADCs and DACs...


Klaus

you are right - I am ignorant about the digital filters and the specific area (DSP) has grown very fast...