Free oscilloscope,spectrum analyser....

[ashok]

I am hijacking the original thread as I felt that many DIY guys with shoestring budgets may miss this very important site.
Kay put up this URL at http://www.diyaudio.com/forums/showt...215#post142215

It provides you with a spectrum analyser , an oscilloscope,a voltmeter and a signal generator. All for FREE. You have everything else you need - a sound card and your computer.

http://www.dazyweblabs.com/shannonsoft/page3.html

Thanks Kay. If these programs work well , they solve the problems of many DIY'ers.
Feedback to dazyweblabs will surely be welcomed, I think.
Have fun.
Cheers.

[Kay]

New Thread, good idea

If the programs are not enough, there will be no mistake to look at this linklist

http://members.chello.se/jpo/measurements.html

best regards
Kay

[hugobross]

just wondering wether the program, using your soundcard, can detect DC, or square wave signals
I think there will be an input cap, don't you?

EDIT: and what about the maximum input level?? does anybody know something about this?? I remember some soundcards uses 2 diodes anti-parallel (limiter) to avoid too high inputs.

[Bricolo]

and what's the maximum input frequency?

[ashok]

I expected these questions. First of all the standard seems to be Creative's Sound Blaster cards. The better ( more expensive) cards are similar or better in specs. So most people would be safe using these specs if they cannot get any from their card manufacturer.
The Sound Blaster AWE64 spec is available at http://support.ap.dell.com/docs/acc/4110d/specs.htm

Input impedance ( Line and CD input ) is about 47K ohms and input level is 0.707 Volts rms ( max ).
Other cards accept 1 volt or slightly more.
The mic input can handle about 0.070Volts rms.

DC input is not possible and the response is usually from 20 to 20Khz . The +/- 1db response usually does not happen in cheaper cards. With a card capable of recording at 48khz or 96khz you can get 24Khz or 48khz high end response, respectively. With 44.1Khz it is about 22Khz which is usually not achieved.

I have not checked out the software myself. I will do that this week and get back. Usually you can measure the response of the card and save it . Then you can use this to correct the response of anything else you measure. Thus your measurement capability would be a flat line.

Note that you need to work at as high a voltage as possible to get the best signal to noise ratio. So the signal generator should work at a fairly output , say 0.6 to 0.7 volts. The signal to your test device should be controlled by an external pot. The input should also be via a pot so that if you are measuring higher voltages , you can cut it down to a safe level. Always start with the pot set at zero output. In the spectrum analyzer mode you can easily see the onset of overload because you will get a very rapid increase in harmonic content , both amplitude and frequencies. Back off by about -3 to -5 db from there.

The first few hours is never easy. Once you get the hang of it , it becomes child's play. Don't get discouraged.
Cheers.

[Christer]

I downloaded the scope/spectrum analyzer from dazyweblabs
yesterday and have played around with it a little bit in loopback
to test my soundcard, a Soundblaster Live. I haven't read the
manual yet (is there one?) so I haven't figured everything out yet,
but I think the program can be useful. I found out today
that I could use 48k sampling freq. with my card, which improved
the frequency response, which is now 20Hz-20kHz +0/-1dB.
Actually, it is much better than -1dB if checking the actual levels,
but I haven't yet figured out how to plot with better resolution
than 1dB, so a small error can show up as 1dB in the plot.

When used as a oscilloscope, I haven't yet found out if there
is a trigger, so the scope function is of limited use right now.

Maybe the spectrum analyzer is the most useful function. It
redraws the plot every 32768 samples, I think, but it is
íncremental, so the precision gets better and better. If waiting
for some 30 - 60 s. it is possible to get the distorsion down to
-90dB for a large part of the spectrum. A word of warning here,
don't trust the built-in THD meter, especially in the sweep mode
where it plots THD vs. frequency. The figures are way off for
low frequencies. My guess is that it sums overtones over the
whole spectrum, but for low frequencies you will get artifacts
from the FFT analysis that show up as very-high-order overtones.
To get a correct THD figure, one would have to disregard these
artifcats. Don't trust the THD figures, but use the spectrum
analyser and look at the spectrum yourself, and your'e fine.

[JBL]

Well witha computer sound card their is always garbage at the end of the spectrum whatever you use so I hope that this was taken in consideration in the thd-meter.

[Christer]

Originally posted by JBL
Well witha computer sound card their is always garbage at the end of the spectrum whatever you use so I hope that this was taken in consideration in the thd-meter.

Well, that's true too, but I think it stops summing at 20kHz so
this is probably not a great problem with 48kHz sampling
frequency, at leat with my card.

The problem is that FFT analysis always seem to give these
very-high-order overtones. They can be clearly seen when
using the spectrum analyser. Exactly the same thing happens
when using the FFT function in SPICE, where no real hardware
is involved, so I assume this is due to the FFT algortihm itself.
Although having once read through the proofs for the foundations
of FFT and convinced myself that it works, I haven't understood
FFT deep enough to figure out quite why or when these
artifacts appear. I think it is related to the ratio between
sampling frequency and fundamental frequency, but that is
just a guess from SPICE simulations.

[Bricolo]

For those who tryed the program:


on my computer (intel 1.5Ghz, 320Mb RAM) it's sooo slow
the display is slow, like 10fps
my cpu is at 100%
the scope isn't able to sync with the signal

[vrbalthezr]

I checked my site stats and there was this incredible 360 hits (normal is 40) today which I traced back to this one message (thanks for the plug).
Thanks, everyone for checking out my site. Any input on bugs or feature additions will most certainly be welcome! The SG-2102 just got warble tones as of last week (for you speaker designers) mainly due to a request from my boss who needed them for test tones..

Vic
DazyWeb Laboratories