Software audio analyzer to mic pre... impedance issues?

Has anyone used a software application like Rightmark Audio Analyzer, Visual Analyzer or Mac the Scope to test a microphone preamp?

The application sends test signals from the computer soundcard, but is the impedance mismatch an issue, particularly with transformer balanced inputs?

These applications look like a great alternative to test equipment for someone like myself with no money or space for such equipment. Just want to make sure they would be truly useful for my current need.

Source and load impedances are important in testing any audio gear. But to test a mic preamp with a GOOD soundcard and the sort of software you mention isn't too hard with the help of a simple level and impedance-matching jig. I'll be back later with more details.

First things first: verify that your soundcard is up to the task. If you're using RMAA, do a loopback test. If the soundcard has lousy frequency response, high distortion and noise and so on then it's pointless to try to use it for measurements. Also be sure to try a loopback test from the HEADPHONE out since for some tests, the headphone output's lower source impedance and higher level can come in handy.

What Dave said.

If you have a real audio oscillator, check it's distortion against the soundcard generator's. Use the lower one.

I built pads to make my -10 soundcard inputs +4 friendly. And for critical stuff, I use a real sig gen.
You could put some balancing stages or xfmrs inline to balance the soundcard inputs.

The onboard generator seems to put out about .04% thd. ok but not perfect.

I'm using one of these. Headphone +4 out and balanced in. $120.

I'm kind of in the same situation. I'm setting up my first test jig primarily for frequency response and THD measurements. I'll be using an Audiophile 9624 front-ended with an Apogee Rosetta. I'm hoping to get some good results.

If skipwave doesn't mind, I'll post back in this thread with questions or results... just since the topics are similar (I don't want to hijack).

I don't mind at all, I welcome any input on the subject. It would be cool if lots of people posted their various methods and experiences with this type of in the box testing.

[quote author="Larrchild"]What Dave said.

If you have a real audio oscillator, check it's distortion against the soundcard generator's. Use the lower one.[/quote]
Interesting this comes up here, I'm interested in what you're all getting. We recently had some about this as well:
http://www.groupdiy.com/index.php?topic=8559&postdays=0&postorder=asc&start=15

I compared a sinewave from DAW and got decent results (good enough for most stuff) but the hardware box had a cleaner sinewave:

I made a 24-bit 1kHz FS sine with CoolEdit & sent it out from my DAW
to a hardware distortion analyzer.
THD wasn't better than -83dB / 0.007% (@80kHz LPF), but will be usable for various stuff.
The residue was (unsurprisingly) very noisy.

The hardware box itself has a signal source as well (HP8903B) and it's a bit cleaner:
-93 dB / below 0.0025 % @ at a few volts of level.
Again, it won't matter much for most applications, but if you want to see
what's going on the hardware box has the advantage.

But as also said in that other thread, this hardware box still can't verify
the depths of a '5532

I just have to re-emphasize, it's not worth attempting unless you have a very accurate soundcard. Download RMAA and do a loopback test to see what's up. That Audiophile USB Larry pictured kicks butt in the loopback test. I bought one to use at work specifically for audio testing.

Here's a setup for ya...

I learned something interesting about computer sound cards. Because there is no hardware volume control in them, minimum mixer level in the PC isn't usually good S/N ratio. Crank the soundcard up and manually (resistively) attenuate the signal to the mic pre for best S/N. :thumb:

So far the best PCI soundcard under $500 has been the ESI Juli@ for me. It has balanced IO and the loopback test shows 0.0003% THD at 1KHz in RMAA. Very low noise floor. It costs a whooping $150.
For my laptop I use the Terratec Phase 24 with Firewire. It is not as good as the Juli@, but it gets down to 0.001% on most days. That one has balanced IO as well.
I have several balanced pads for input that terminate in 150 ohms and for outputs that terminate in 600 ohms.
You don't want to mess with adjusting levels on the soundcard. Set it for 0dB output and use pads to adjust the level. You need a few pads like 20dB, 40dB , 50dB. Then use you amp under measurement to adjust the gain back to 0dB on the output.

0.0003% THD :shock:

Impressive, dude.

Tamas, could you post an HTML report on that card (using the "report" feature in RMAA)? I'm interested because I'm considering upgrading the soundcard in my home computer if there's anything left of my tax refund once the bills have been paid.

Here's the report I ran on the Audiophile USB:
Link

I was limited to 48kHz by the processor speed of the old laptop.

You have that bump above 10k in your noise floor too, Dave. (albeit at -120)
I spent an hour moving computer monitors and grounding stuff before I determined it was inside the Maudio. Ha.

Thanks for the illustration Dave! That's exactly what I was looking for in terms of interfacing with the preamp and reflecting the correct impedances to the inputs.

Right, I was just about to post something along this same line. I was looking through the "meta" I noticed that Jakob had previously posted a link for a software version of some test equipment. I went to the site, it was written in dutch I think. Anyhow, I'm running a Presonus Firebox. The specs are as such:
Preamp bandwidth: 10hz-50khz
Pre-input impedance: 1.3k ohms
Instrument input impedance: 1M ohms
Pre-THD: <0.005%
Pre-EIN:-125db
Pre gain: 45db
Line in Z:10k
TRS out Z: 51 ohms
TRS main out Z: 51 ohms
ADC dynamic range: 107db
DAC dynamic range: 110db
Firstly (not to hi-jack the thread) I THOUGHT this box would sound pretty decent, however I'm no preamp guru. I havent really been able to realize the sonic quality I had hoped for. Are these specs indicative of something that should sound at least decent. I got my Champ working I threw a 58 in front of it, sounded horrible. Didn't seem to matter where I moved it. Any feedback regarding the published specs? Also balanced in/out (6in-8out) software mixer, blah, blah, blah. Would this at least work for the listed apps earlier in this thread? Sorry for babbling, thank, Ben.

You're all getting quite a bit better results than I got with my Terratec EWS88MT/D (THD -83dB / 0.007%).
I was lazy, I looped back out-to-in by means of routing on my mixer, which is of course a &*( thing to do for such a test.

I'll run the test again with a direct connection.
What I saw here http://learjeff.com/RMAA/results.html looks promising, 0.001% seems obtainable - and that would be better than my hardware box !

Very cool thread this - I was about to buy me some data logging hardware to be able to plot from my hardware THD-meter since it seemed to be the cleanest generator. Might turn out that the DAW is cleaner after all.

Regards

[quote author="NewYorkDave"]Here's a setup for ya...[/quote]

Hey thanks Dave, just tested - works perfect! :thumb: I'll need to build a permanent one on one of these crap boxes I have..

Here are my Juli@ results (symmetric I/O): http://www.digital-synthologie.de/diverses/[email protected]

Here's the Terratec EWS88MT again, but now directly
linked from unbalanced out to unbalanced in:
Terratec EWS88MT & EWS88D+ADA8000
I now get 0.0006% THD.

I also tried some other loops like the ADAT I/O of its sister-card (EWS88D) using the Beh. ADA8000 for DA & DA. This was not as good.

Finally I tried a direct loop of the EWS88D S/PDIF I/O but got some strange results, probably due to sync-settings.

But OK, the DAW now wins over the hardware signal generator.

And when some Beh.-stuff gets in the loop everybody suffers ! :wink:

I did the RMAA loop back test on my PC laptop, and the results were too hideous to mention. That means I gotta do it with my Mac.

Anybody know of a freeware application for Mac? I can't shell out the cash for MacTheScope.

Bump!

Has anyone got the picture NewYork Dave posted on the first page? It seems that it's gone