[Chester]

Hi, I was just pondering the sound card calibration system after noticing that the 'ripples' on the loop-back measurement from my audigy 2 zs notebook look quite similar to some of the ripples in my frequency response measurement of my speaker and it hit me: line impedance... since the microphone (preamp) to sound card will have a different impedance setup than the sound card loop-back setup, will that not affect the determined frequency response? would there be a better way than a loop-back cable to measure the calibration file? many of my measurements appear to have the same ripple response in the midrange (but inverted since it is 'subtracted... could this mean that the ripple responses may be less than accurate; one would assume that if you are subtracting out something that should always be there, you would not have any 'trends' between the measurements that look like what you are subtracting out...

 

[brucek]

'ripples' on the loop-back measurement from my audigy 2 zs

Let's see the loopback graph.

since the microphone (preamp) to sound card will have a different impedance setup than the sound card loop-back setup, will that not affect the determined frequency response?

Not at all. This interface is the same as all line level interfaces, it's a high impedance connection, where the output impedance of the source is usually in the order of 50 ohms to 150 ohms and the input impedance of the load is in the order of 10Kohms to 50Kohms.

This is known as a voltage bridge connection where the preamp is acting as a voltage source and almost no current is drawn. Large variations in impedance have no appreciable effect on the response.

brucek

 

[JohnM]

Some ripple in the loopback measurement at very low frequency can occur if the soundcard has very extended low frequency response, those ripples can be removed by increasing the impulse response window width (the ripples are an artefact of the window truncating the response before it has fully decayed into the noise floor). Ripples in the midrange of the soundcard cal would suggest feedback from monitoring being active, that needs to be fixed to get a valid cal result.

 

[Chester]

the ripples are in the midrange...

The first image is there to show the soundcard cal, the second is trying to show how both measurements seem to have a lot of the same 'problem areas'... I am presuming it is a feedback issue; how would I go about fixing that?.. basically the input (measured) is some how getting routed to the output again on the card... or is it some other type of feedback?

 

[Chester]

I just remeasured the cal file and it appears to be much smoother now; I believe it is the exact same setup I had when measuring the first time but clearly something changed...

 

[brucek]

I am presuming it is a feedback issue; how would I go about fixing that?

It is for sure a feedback issue. It's because the line-in is enabled in the Windows Playback Mixer. The line-in should only be enabled (unmuted) in the Windows Recording Mixer, not in the Playback Mixer - mute it in Playback.

brucek

 

[Chester]

got it; thank you for your help Bruce and John

 

[JohnM]

For the Audigy 2ZS the advanced controls, accessed by clicking the red + in the REC section of the Creative Surround Mixer, have a check box for "Record without monitoring" that needs to be selected (under XP at least, may not be applicable under Vista).