Speaker Impedance measurement ,circuit and sound card output

[icbcodc]

Hello,I'm new guy and sorry for my poor English first.

Something wrong and I can't get the Impedance curve when I'm trying to measure the impedance curve of my speakers(for cars) refering the help files of REW 5.19.

My sound car is Behringer UMC22 with input and ouput sockets like this:

I've confirmed with the seller that the OUTPUTS is line out,and the input(MIC/LINE 1) is line in. I use OUTPUTS 2(R) as the output and MIC/LINE 1 as input

My circuit is like this:

Actual connection is like this:

First part ends here for saving purpose.

 

[icbcodc]

I've checked the line I made many times,I'm sure the connection is correct.

I make a calibration for the Impedance Rig

I just short out the Rsense following the instruction in REW help, and check the new connection is correct like this:

Actual circuit is like this below, I 've checked the connection is correct.

output/input volume adjusted:

CHECK LEVEL IN REW:

AFTER CALIBRATION,is the result 100Ohm correct?

What does this mean : "100.00% of reference" ?

 

[icbcodc]

Short the test leads which should be connected to the pole of speaker later. Is this connection right?

The level for zero ohm calibration is too high,I don't know why.

I just turn down the main ouput level (to 8 clock) of sound card shown below.

check level again ,it seems O.K

There's nothing,is it right? Does this mean test leads resistance is zero?

nothing available.

Is there anything wrong?

 

[icbcodc]

Finally, I connect the leads to my speaker,and run the impedance measurement,there is nothing, the sound of sweep from speaker is very low.

What did I do incorrectly? Why I could not get the impedance curve?

 

[John Mulcahy]

Mic/Line 1 is a single, balanced input, not a stereo input. Inst 2 is the second input channel.

 

[icbcodc]

Mic/Line 1 is a single, balanced input, not a stereo input. Inst 2 is the second input channel.

Thank you so much John, It seems you are right.

In the UMC22 manual, "Inst 2" is "1 x ¼" TRS (Instrument)" .

I found TRS stands for Tip（signal）、Ring（signal）、Sleeve（ground）, so this is the stereo input, am I right?

In the manual, "MIC/LINE 1" is “XLR/TRS combo connector”,but this TRS is no stereo?

Maybe I should try it tomorrow.

 

[John Mulcahy]

I would say neither is stereo, they are both mono balanced inputs. Tip and ring are the + and - of the balanced connection.

 

[icbcodc]

I would say neither is stereo, they are both mono balanced inputs. Tip and ring are the + and - of the balanced connection.

My goodness,so I use the wrong soudcard.

I will buy a second hand USB soundcard (ASUS XONAR U3 PLUS) today.

Is this the right soundcard for measurement?

 

[John Mulcahy]

It is probably OK.

 

[icbcodc]

It is probably OK.

Thanks.

I've bought this soundcard,a 100 ohm sense resistor and two 3.5mm TRS connectors.

I'll report in a few days.

 

[icbcodc]

It is probably OK.

Hi John, I have a few question about the impedance measurement in latest REW version.

In the "Make a measurement" window, "Open circuit cal" means the following circuit?

"Short Circuit Cal" means the following circuit?

"Reference Cal" means the following circuit?

If my understanding is right, the value of the reference resistor should be 100ohn or lower?

Should I use a 10ohm resistor, when the uppper Rsense in the rig is 100ohm?

 

[John Mulcahy]

Yes, those are correct. The cabling up to the load connection should be left in place. The reference measurement resistor should be of the same order as the loads being measured, so for speaker measurements probably 50 ohm or lower.

 

[icbcodc]

Yes, those are correct. The cabling up to the load connection should be left in place. The reference measurement resistor should be of the same order as the loads being measured, so for speaker measurements probably 50 ohm or lower.

About reference resistor

My speaker (the load you mean) which will be measured is about 4 ohm,so in "3.reference cal",the reference resistor should be around 50ohm or lower?

About RSENSE resistor
My speaker which will be measured is about 4 ohm, but in REW help file, 100ohm Rsense resistor is recommanded, so I've bought a 100ohm Rsense resistor which is still not delivered.

But in the famous book of Joseph D'Appolito 《TESTING LOUDSPEAKERS》 1998, he used a 1000 ohm resistor. Is 1000 ohm RSENESE resistor still fine in the REW measuring circuit ? Because the 100 ohm RSENSE resistor is still not delivered,I have only a 1000 ohm resistor now.

Sorry,the version of my book is not English.

He manually calculated the value with many equations.

Second hand Asus Xonar U3 PLUS is received.

 

[John Mulcahy]

A higher sense resistor will increase noise levels, I don't recommend that. 50 ohm is fine for the reference resistor.

 

[icbcodc]

A higher sense resistor will increase noise levels, I don't recommend that. 50 ohm is fine for the reference resistor.

Thanks John.

I will buy a 50 ohm resistor to use.

And I made a test today,the U3 PLUS sound card was not passed "open circuit" calibration.

and in the check level window, the level of left and right channels are different.

I checked the circuit using a multimeter, it's right. So I have to get U3 PLUS refunded,and buy another USB soundcard.

 

[John Mulcahy]

And I made a test today,the U3 PLUS sound card was not passed "open circuit" calibration.

I'm not sure what you mean by "not passed", calibration is not a pass/fail. There is some noise at 50 Hz which looks like mains pickup but otherwise it looks OK. The difference between the channels isn't large enough to cause a problem, calibration compensates for it.

 

[icbcodc]

I'm not sure what you mean by "not passed", calibration is not a pass/fail. There is some noise at 50 Hz which looks like mains pickup but otherwise it looks OK. The difference between the channels isn't large enough to cause a problem, calibration compensates for it.

Oh,but why the software tells me the Calibration is not valid,I think it means "not passed".

 

[John Mulcahy]

Ah, I was looking at the wrong figure in your previous images. That is a larger difference than should be expected between the channels of a soundcard.

 

[icbcodc]

Ah, I was looking at the wrong figure in your previous images. That is a larger difference than should be expected between the channels of a soundcard.

Yes,I've taken this soundcard back to seller.

 

[icbcodc]

Ah, I was looking at the wrong figure in your previous images. That is a larger difference than should be expected between the channels of a soundcard.

Hi John

Unfortunatly,the test result is still bad with another soundcar (SB0271,soundblaster).

I've made sure that the left and right channel level in the software are same.

Should I adjust the balance of left and right channel to make the test result good or buy another soundcard?

 

[John Mulcahy]

What sense resistor value are you using?

 

[icbcodc]

What sense resistor value are you using?

My current RSENSE is 999 Ohm.

 

[jtalden]

I noticed in Post-15 REW 'Soundcard Preferences' picture that:
Measurement Channel; output = R, input = L
Reference Channel; output = L, input = R
I don't know if this cross channel setup could cause the channel balancing problem or not, but you may want to try changing it to the more conventional setup.

 

[John Mulcahy]

My current RSENSE is 999 Ohm.

The high sense resistor together with the input impedance of the soundcard is making a voltage divider that is reducing the input, hence the channel difference.

 

[icbcodc]

I noticed in Post-15 REW 'Soundcard Preferences' picture that:
Measurement Channel; output = R, input = L
Reference Channel; output = L, input = R
I don't know if this cross channel setup could cause the channel balancing problem or not, but you may want to try changing it to the more conventional setup.

I don't know how to do the conventional setup,sometimes I am also confused by the bad result. I just read the help file for impedance measurement and ask somebody who did that before and watch some Youtube videos.