Hi all
I have recently started the jurney of making impedance measurements on drivers. I am using REW with Focusrite Scarlett 2i2 audio interface. I have Keysight lab calibrated DMM and a two digit precision scale.
Rsense is 99.32ohm. I am running the driver on a small amplifier.
I have made several measurements with success and can generate T/S parameters in REW. I am however after reading V. Dickasons Loudspeaker cookbook v8 interested in using Speakerbench and their advanced parameters method.
I succeeded in uploading the zma files and create a Json container. But the data seems to have an error - it says there is an issue with the quality. Can you guide me on what i am doing wrong?
Measurements in REW
Imported into Speakerbench - Quality issue :/
The wierd thing is that if i chose the Z measurements, they seems to be imported correct
I have recently started the jurney of making impedance measurements on drivers. I am using REW with Focusrite Scarlett 2i2 audio interface. I have Keysight lab calibrated DMM and a two digit precision scale.
Rsense is 99.32ohm. I am running the driver on a small amplifier.
I have made several measurements with success and can generate T/S parameters in REW. I am however after reading V. Dickasons Loudspeaker cookbook v8 interested in using Speakerbench and their advanced parameters method.
I succeeded in uploading the zma files and create a Json container. But the data seems to have an error - it says there is an issue with the quality. Can you guide me on what i am doing wrong?
Measurements in REW
Imported into Speakerbench - Quality issue :/
The wierd thing is that if i chose the Z measurements, they seems to be imported correct
Member
Joined 2003
I can't speak for Speaker bench, but the extra bumps in 300Hz range are an indication that the added mass is not well attached.
I have attached the ZMA files. The driver is a 5,25" 4ohm bas/midrange from MB Quart. It is a test device, and i have two. Both are performing with like 5% difference when comparing the T/S parameters i calculate in REW.
I use Tesa tack to attach some small coins to the membrane - good point about the 300hz resonance. I have tried two different benc vises, with same result, but they both clamped the basked and not the magnet as other seems to prefer.
I am on a learning curve 🙂 All your guidance is much appreciated
I use Tesa tack to attach some small coins to the membrane - good point about the 300hz resonance. I have tried two different benc vises, with same result, but they both clamped the basked and not the magnet as other seems to prefer.
I am on a learning curve 🙂 All your guidance is much appreciated
Dcibel has named the problem.
In my experience, coins and magnets are difficult to attach resonance-free. It is easier with soft modeling clay.
Furthermore the soft clamping on the basket could cause problems (see chapter 3.1.3).
Here you can find some information on measuring Thiele Small parameters: https://www.artalabs.hr/AppNotes/LIMP-HB-D2.53.pdf
Regards
Heinrich
In my experience, coins and magnets are difficult to attach resonance-free. It is easier with soft modeling clay.
Furthermore the soft clamping on the basket could cause problems (see chapter 3.1.3).
Here you can find some information on measuring Thiele Small parameters: https://www.artalabs.hr/AppNotes/LIMP-HB-D2.53.pdf
Regards
Heinrich
Ente and Dcibel. You were right. Used only tack and no coins. It seems like a result with less resonances. I will give it a try, and make three measurements and upload to Speakerlab
Curve (3) old measurement with coins and tack
Curve (6) new measurement with only tack
Just a short update - i uploadet it to Speakerbench and got the quality to "fair" 🙂 Thats better - but my inner OCD likes if the response was Perfect 🙂
How can i do my measurements better?
Curve (3) old measurement with coins and tack
Curve (6) new measurement with only tack
Just a short update - i uploadet it to Speakerbench and got the quality to "fair" 🙂 Thats better - but my inner OCD likes if the response was Perfect 🙂
How can i do my measurements better?
Last edited: