PC sound card based distortion analyzers.

G

George S.

Veteran and General Yakker
Joined
Feb 18, 2020
Messages
5,062
So was reading up about the differences in oscilloscope crt vs digital displays and identifying distortion in a sine wave. Consensus seams to be the crt is better for seeing the "fuzziness" of distortion because the digital display "corrects" it out.
Several comments were made on different sites that pc based sound cards are the way to go when measuring distortion. Lee brought this up in a prior thread for me to check this out.
I can see using it on a preamp, the software and sound card feeds the preamp over standard mini jack to RCA cables, the preamp processes what it's been given, and it's back into the sound card microphone jack where it's processed and results shown.
Anybody here tried this and what are your thoughts. If so, what software worked best for you? Going to try it this weekend as I have several old Creative X-Fi cards and Dell desktops not being used.
 
Pretty tough to "see" distortion of less than 5% (even with a trained eye) on an o-scope whether it be digital or analog, so for hi fi work, useless. Even noise riding on the signal can be tough to see unless it's really gross. That's why you can pay big bucks for signal analyzers - they can see and measure it down in the weeds.

But I'm interested in your main question too. What experience does anyone have?
 
I bought one of the Rigol 4 channel digital storage scopes last year for $350. It has a lot of nice features and the trace SUCKS. I sent the first one back thinking it was broken. My 30 year old 50mhz Hitachi and 60 year old Tek 503 beat the pants off of it from a trace perspective. The 503 is razor thin. I can see oscillations and noise riding on a waveform with the analog scopes, that the digital ones won't show.

At my day job we have $25,000 Waverunners to look at 60hz inverters and occasional 10khz PWM. You can see everything and you should for $25,000. The waverunners have a learning curve with their touch screen menus. I tend to to point at the screen and unintentionally change settings and there is no "back" button.

I thought I must be the only person who cared about what the trace looked like until Ed mentiioned it out of the blue. Really, I'm not that crazy?
 
No, you're not Don, my little Leader 308S shows oscillations that I can't (or haven't yet) on my digital..
 
Building the probes/interface is going to take some research. Rest looks easy.
 
Well, I can see I really need to research the interface/probes and the input impedance of the Creative X-Fi pci cards I have. I also have a Asus U7 usb sound card, but use that for Software Defined Radio listening to short wave on the big 3 monitor Lenovo workstation.
Think I'll get a X-Fi working on a old Dell. Take some measurements for a baseline. Then repeat on the U7 on the Lenovo.
This will give me some idea if it's accurate or varies with hardware.
Will most probably just buy a quality prebuilt probe once I research them. Seems cheap enough.
 
The downside of the sound card & probe is that it is DC coupled only. Suppose you want to look at the ripple on a power supply. You use AC coupling on a conventional scope. Also there is no way to calibrate the probe to accurately look at rise times/square waves. Maybe there is no interest in that so it doesn't matter. 90% of the people who use a scope don't know they can and should calibrate the probe.

What I think is really important is people who are trying to learn something and this is an inexpensive entry point.
 
Another alternative is USB based scopes. I bought one about 15 years ago to use during training presentations to look at line voltages. Worked good until some idiot wired the line and neutral backwards on an outlet and blew the snot out of the scope. Bought another scope and isolation probes.
 
Been reading. Not going to work as well as a cheap used crt oscilloscope. So much info on the web takes me down that fork in the road. Now to dig into audio analysis.
 
Too much rigging and massaging needed to make a sound card work well. Something to play with, but doesn't seem to meet the end goals of easy to use with precision and accuracy. Going a different direction yet to be determined.
 
A question : why do you want to test THD + N ?
Are you a designer or do you want to do for amusmement ?
If you want to test THD + N you need an audio analyser .
The difference is that with a scope you are in the time domain
and with an analyzer you are in the ferquency domain .
Scope together with and audio oscillator is O. K .
for reparing : ripple , caps , bad contacts and so on .
( in other words you can check the levels )
You can also use and audio test cd with a cassette recorder
as AC voltmeter for this purpose .
To test noise and THD you need an audio analyser .
You can test noise and 3rd and 5th harmonics .
O.K . for designers ( Joe ) and commissioning ( Lee )
Ciao
Marco
 
Addendum :
To move from the time domain to the frequency domain
you need an instrument that implements the Fourier Transform .
( spectrum analyzer or some scopes with this function inside )

Ciao
Marco
 
To learn. With all this old "new" tech laying around not being used there have to be alternative ways of quantifying distortion rather than the old way of using expensive dedicated equipment. Next thing I'm going to look at is REW, supposedly it can also measure distortion. It's a program I'm somewhat familiar with. Very easy to use so far for integrating subs and mains.
 
Last edited:
wattsabundant said:
There is a nice looking HP334 on ebay right now that would be hard to beat. HP334 Add an HP 204 for another $100 or so and you have a decent test set.
Click to expand...
I'll definitely look when I get the chance. Now just a little north of you if your home just outside 270 in Sunbury, working. Thanks Don.
 
I have a HP 1740a scope that I have used for years. Works pretty much for what I need it for with the Phase Linear and also with my tube amps, also for setting up phone cartridges on turntables
 
You must log in or register to reply here.
Back
Top