Oscilloscope to sound card buffer

[unclejed613]

you can find better software.... SpectrumLab is what i've been using... https://www.qsl.net/dl4yhf/spectra1.html

 

[()blivion]

That... software... is... BAD ASS!!!, thanks. +1

 

[()blivion]

OK, So... I played with the new software and figured some things out. I also made some modifications to the circuit to fix some problems I was having. Now signals from one probe are not coupled into the circuitry of the other when the op-amp is in "1x mode" (voltage follower). I also added a LED to indicate when the circuit was being powered. This is because I don't have a power button; relying instead on unplugging the wall adapter or barrel plug. And it's nice to know that it's off when I'm using my PC for other things. And that I'm not at risk of blowing my eardrums out from having experimental electronics hooked up to my PC when trying to watch shows.

Here is a picture of me probing the PWM signal of a motor controller I made for my cobbled together generic Dremel tool. Take note that the frequency's are well above the 10K limit the other software was having, the one spur being almost at the sample rate of the sound card. (If I had the source code for the PWM controller still I could tell you if the frequency read out was right or not )

View attachment 65792

Problem
There genuinely is some kind of noise being generated in the device it's self. Exactly where and why this is happening, I don't know yet. But you can see it as all the light color bands to the left in the picture up above. And below here is a "base line" picture that tells some of the story also. It's is a zoomed in view of all the noise. The top third of the graph is what is seen with absolutely nothing plugged into the line in port on the sound card. The middle is with the device fully plugged in, in 1x mode, and the probe grounded. The bottom third is with just the headphone-to-headphone cable plugged into the line in port, to act as an antenna of sorts. You will note that there is some noise present on the line even with just the cable plugged in. But it is small compared to when plugged into the device.

View attachment 65790

Anyway. I plan on probing different parts with various "shunts" and "bypasses" to ground and seeing if that eliminates the noise or makes it worse. Any other suggestions maybe?

 

[unclejed613]

first of all, realize this is not a linear display. you are seeing harmonics of 120hz that would be nearly invisible on an oscilloscope. it looks as if many of these harmonics are at least 60db less than the signal you are looking at, which is much smaller than the 1% or so resolution of an oscilloscope. there is an oscope display in SpectrumLab. the good thing about SL is that you can use the specialized functions of higher end sound hardware, including the higher sampling rates (192khz last time i looked) if your card has that capability. the other interesting feature of SL, that takes some playing with to begin to understand, is the "virtual audio lab" with oscillators filters, modulators, etc.. that make it extremely versatile.