USB soundcard for measurements?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm looking for a new sound card to do 2 channel measurements with using a laptop. I previously had a laptop that had a PCMCIA slot which I bought a nice Echo Indigo IO card for. Unfortunately I no longer have that computer and the two laptops I currently have either don't have a PCMCIA slot or are internally crippled so the Indigo feeds back.

So now I am looking for a USB soundcard. I need something that has a line in and line out and is full duplex (can play and record at the same time). I also need something that I can disable internal feedback on. In other words, I need to prevent the input from being fed to the output. I'm looking at the Creative X-fi 5.1 for $60:

http://www.newegg.com/Product/Product.aspx?Item=N82E16829102020

However, I've never seen a set of soundcard specifications that lists these two qualities that are important to me. So that's why I'm asking here - does anyone have an X-Fi and have you used it to take 2 channel measurements with successfully? Or does anyone have any other recommendations around this price range?

Thanks,
John
 
Hi John,

I use an M-Audio Transit USB with my Praxis rig with good results (it is one of their recommended cards), though the Praxis "audpod" has it's own mic pre. The only negative is that their Vista driver support is spotty.

If you're looking for something with phantom power, you might consider the M-Audio Mobile Pre USB. I've use this with an EASEra setup with good results.

-Chris
 
An externally hosted image should be here but it was not working when we last tested it.
[/B][/QUOTE]

Helmuth,29 euro its very cheap but when it comes to thailand. its very expensive(around 85 euro) diy in thailand thiks a lot when theywill buy.
 

Attachments

  • soundcard.jpg
    soundcard.jpg
    34 KB · Views: 329
here is the best for the price

http://cgi.ebay.com/ESI-U24XL-24-bi...5|66:2|65:12|39:1|240:1318|301:1|293:1|294:50

I did a lot of research in finding out what would give the most accurate result and this one came out to be the best gives the best of both worlds inexpensive with truly flat accurate readings
I was going to use my creative Audigy nx, but I am waiting to get one of these before doing any testing
 
I've been using the Behringer more, and my conclusion is it sucks. The signal to noise ratio, specifically the 1/f noise, makes it almost unusable for measurements at low frequencies in my system. There is so much noise I can't measure impedance at low frequencies accurately - the noise is making the impedance rise at low frequencies. I know I could get around this in various ways, but basically I don't want to and don't think I should have to. I've never had to before with other soundcards. I was measuring a speaker outside today as well, and with my mic (Linkwitz modded Panasonic capsule) around 5m away and mic my preamp gain all the way down,I couldn't get above the noise floor at 20hz without clipping the input of the sound card. Once again, I could get around that by building more attenuation into my mic pre, but it should not be necessary. Besides, I was already cranking on the woofers and didn't really want to go any louder. I was afraid my neighbors were about to start complaining.

I must say in the Behringer's favor, it was truly plug and play on my laptop running WinXP and it was cheap, but that's about the end of its advantages.

So now I'll try something else that's been previously suggested in this thread - just not sure what yet.
 
I've been using the Behringer more, and my conclusion is it sucks. The signal to noise ratio, specifically the 1/f noise, makes it almost unusable for measurements at low frequencies in my system. There is so much noise I can't measure impedance at low frequencies accurately - the noise is making the impedance rise at low frequencies. I know I could get around this in various ways, but basically I don't want to and don't think I should have to. I've never had to before with other soundcards. I was measuring a speaker outside today as well, and with my mic (Linkwitz modded Panasonic capsule) around 5m away and mic my preamp gain all the way down,I couldn't get above the noise floor at 20hz without clipping the input of the sound card. Once again, I could get around that by building more attenuation into my mic pre, but it should not be necessary. Besides, I was already cranking on the woofers and didn't really want to go any louder. I was afraid my neighbors were about to start complaining.

I must say in the Behringer's favor, it was truly plug and play on my laptop running WinXP and it was cheap, but that's about the end of its advantages.

So now I'll try something else that's been previously suggested in this thread - just not sure what yet.

Hi John

1/f noise is not a typical characteristic of a sound card, but it is very typical of a microphone. All mics exhibite 1/f noise to various extents depending on how the capsule is loaded by the first stage of the preamp. The input impedance needs to be very very high to supress the 1/f characteristic. In hearing aid mics, where noise is critical, we used tera-ohm load resistors on the FET front end. The low cost Panasonic capsules are not going to have this kind of capability. So my guess is that it is your mic/mic preamp, not the Behringer sound card.
 
Hi Earl,
My mic definitely has some 1/f noise of it's own, but that has previously been useable with other systems (computers and soundcards). And it is still workable with this one, just not the best. The real problem I am having is that my impedance measurements are not working. I believe I've tracked this down to 1/f noise that shows up on the inputs of the soundcard. To verify I'm not crazy, I confirmed with some quick math that if I add equal 1/f noise to the voltage across the DUT and to the voltage across the current sense resistor in the test jig I'm using where the 1/f noise exceeds the measured signal level at low frequencies (which is what is happening), then the expected result is impedance increasing at low frequencies, which is what I see in my measurements. I've never seen this behavior before with any other soundcard I've used with this same impedance measurement jig.
 
I've got a UA-1ex and I am having odd results using speaker workshop.
Once I got some feedback and when using the "advanced" drivers it disables the standard volume controls where you can mute the line in. I am going to try a different DIP switch setting and failing that try the standard driver.
 
Hi Earl,
My mic definitely has some 1/f noise of it's own, but that has previously been useable with other systems (computers and soundcards). And it is still workable with this one, just not the best. The real problem I am having is that my impedance measurements are not working. I believe I've tracked this down to 1/f noise that shows up on the inputs of the soundcard. To verify I'm not crazy, I confirmed with some quick math that if I add equal 1/f noise to the voltage across the DUT and to the voltage across the current sense resistor in the test jig I'm using where the 1/f noise exceeds the measured signal level at low frequencies (which is what is happening), then the expected result is impedance increasing at low frequencies, which is what I see in my measurements. I've never seen this behavior before with any other soundcard I've used with this same impedance measurement jig.

Yes John that is odd. never seen that before myself either. In fact I've never seen 1/f noise in anything but a microphone. Thats because it implies a small capacitance input impedance which is very odd. Do you use any capacitors in this circuit of yours?
 
Hi Guys,
i recently bought a Behringer UCA202 for having measurements with Visual Analyser ,
i'm so rookie in these areas (both measurements and soundcards), and wanted to start with the cheapest possible hardware solution.. that's UCA202 :)

i experience some strange behaviour :
when turning ON analog outputs, i can see some related measurements in the input channels, also if inputs and outputs are disconnected from the UCA202 !
It happens (slightly attenuated) also when turning on only one output channel (you can see both the inputs measuring somthing related to the output, which is not externally connected to the inputs anyway..)

May the cause be related to the next phrase?
I
I also need something that I can disable internal feedback on. In other words, I need to prevent the input from being fed to the output.
Can i have some workaround to solve this problem?

Do you know some other good Oscilloscope / Spectrum Analyzer / Signal Generation software?

thank you all! :)
 
Last edited:
The Behringer doesn't seem to be as good as the M-audio Transit. I use the Transit as a simple external USB soundcard for loudspeaker measurements, and I am very impressed with it. It's only full-duplex up to 48kHz sampling, and I'm not sure if Windows 7 drivers are out yet for it though.
Regarding a pc-based audio spectrum analyser. The (free) "Visual Analyser" program is very good indeed.
 
Do you use any capacitors in this circuit of yours?

No, just some resistors. I imagine if I used an external amplifier to drive the speaker at a higher level I would not have this problem, but I've never had to do that with other sound cards. I will probably try that tonight while I wait for an M-Audio Transit to show up.

when turning ON analog outputs, i can see some related measurements in the input channels

In my experience it's typical to get some crosstalk on all sound cards - you will see a bit of the output on the input. With MLS systems, I see a small impulse right at the start of recording the input, while the real measured (larger) impulse follows later. You can adjust your windowing to not look at the initial impulse. It's feedback that you want to avoid (well, avoiding crosstalk would be nice as well). Feedback would be at about the same level as the output signal, and that would typically make it about the same level or higher than the signal you are trying to measure which would cause problems. Crosstalk is generally much lower in level.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.