Test setup check

[ch83575]

I just got my Soundeasy program to work (after a 2 month debocle with the software key) and I decided before I do any speaker testing I should test the testing rig. I found the results appaling! I have my sound blaster 24 Live! hooked up to a sony mini system amp, but I am not sure if this setup will be high enough quality to test anything!

Here is an oscilloscope reading of the soundcard output and the amp output when attached to a 8 ohm resistive load:


The soundcard output itself is not really a square wave and the amp output is far from it!


Here is a spectrum analysis of the amp output with and without the tone generator:



Here you can see peeks in the background noise that make reading anyting below 80dB down impossible.

I guess my question is wether all of the gurus out there think this setup will allow me to design a good speaker or is the test rig itself too flawed to do any serious testing with? If it is no good which would you replace first the card or the amp?

Thanks!
Chad

[pinkmouse]

Which waveform is which?

[ch83575]

In the oscilloscope picture the bottom is the soundcard output and the top is the amp output into 8 ohm resistor.

Chad

[ch83575]

Well I think the problem is the amp. Here is a spectrum analysis of the noise floor of the card output without the amp:


Just a little 60Hz noise that is easy to recognize and ignore (is it possible to add better filtration on the card?). All of the rest of the garbage is the amp! So I guess the question now becomes: do you think that this amp is a suitable test bed, or should I invest in something better (maybe like an old NAD or something). Ideas?

Thanks,
Chad

[Dryseals]

I'm going to follow this thread, because I want to see your results. I looked real hard into several software programs and did not feel comfortable with the claims. So I went the other route and purchased a freq counter that will do 0.1 HZ resolution, a 15MHZ dual trace scope and an audio signal generator that will produce the different sign waves I'll need. I'm old school and worked in a cal lab on test equipment for many years back in the 70s and did not feel comfortable with software being able to render my sound card into a high dollar peice of test gear.

Maybe some on out that has done some side by side testing with calibrated test equipment can chime in.

Don't get me wrong, I'm not an old school computer hater, I do programming on control systems and have been for years. I have 5 PCs in the house networked, my first PC was purchased in 85, a 4Mhz turbo 8 rocket fast machine. Before that I had a TI 4A and for about 10 years I had a PC business on the side. I just did not feel comfortable with the software claims.

[pinkmouse]

Originally posted by ch83575
or should I invest in something better (maybe like an old NAD or something). Ideas?

I run a mono LM1875 GC on a little 15-0-15 20VA traffo for just about all the testing I do.

[soongsc]

Originally posted by Dryseals
I'm going to follow this thread, because I want to see your results. I looked real hard into several software programs and did not feel comfortable with the claims. So I went the other route and purchased a freq counter that will do 0.1 HZ resolution, a 15MHZ dual trace scope and an audio signal generator that will produce the different sign waves I'll need. I'm old school and worked in a cal lab on test equipment for many years back in the 70s and did not feel comfortable with software being able to render my sound card into a high dollar peice of test gear.

Maybe some on out that has done some side by side testing with calibrated test equipment can chime in.

Don't get me wrong, I'm not an old school computer hater, I do programming on control systems and have been for years. I have 5 PCs in the house networked, my first PC was purchased in 85, a 4Mhz turbo 8 rocket fast machine. Before that I had a TI 4A and for about 10 years I had a PC business on the side. I just did not feel comfortable with the software claims.

PC based testing are quite limited due to the following limitations:
1. Sample rate limit.
2. Stability of sampling and software real time processing.
3. Voltage range limitations.

If we know the limitations, we know when to use it and how to evaluate the results.

I'm interested in Sound Easy too!

[soongsc]

Originally posted by ch83575
: do you think that this amp is a suitable test bed, or should I invest in something better (maybe like an old NAD or something). Ideas?

Thanks,
Chad

I would recommend somthing that has the power to drive the limits of the drivers you might use in the future, probably at least 50W~100W.

What kind of setup do you use? Do you use anything to prevent the PC sound card from being damaged?

[ch83575]

Just for laughs I tried letting the soundcard drive the 8 ohm load (hell the box says it can so why not!?). When using my dummy resistor the square wave looked great! So I tried a pair of speakers... not so good! Here is the screenshot, top is the 8 ohm resistor, bottom is the speaker:


Maybe I will drag my Linn amp upstairs to the office just to see if it holds up better to the 30Hz square wave from hell, I hope it would, it cost me a lot of hard earned money .

Dryseals: Is this equipment for testing amps, or speakers? If you want to test speakers it seems you would need to at least add a spl meter. Sounds like a pretty good test setup for electronics.

pinkmouse: I am just finishing up my Wallin mic preamp today (hopefully) so maybe my next project will be a small chip amp... Wow, I thought I was getting into speaker building

soongsc: I dont have anything to prevent damage to the computer just yet, so I am doing all of these tests at very low levels (output around 1.5v). This makes the amp I am using look even worse to me. What if i was asking it to put out a whole watt!!?

[soongsc]

ch83575:
Since you're using the Wallin Preamp, are you using the Wallin Jig as well?