Hello
I'm using Speaker Workshop to measure my drivers.
The results are good, both frequency and impedance.
But there is two things I dont really understand or getting right.
First; after measurements are done, I estimate S/T parameters, this works fine, and I get good results for FS, Vas, Qes, Qms, Qts and Re.
But Le is always very low, like 0,2mH, L1 says 0,3mH and Series Res is 13,7 ohm ? (this is measurments of a Seas CA18RNX)
I do the impedance measurements through a amp, because my soundcard has not enough output power.
Second; when measuring frequency, I get some strange (?) results of phase. Look at the attached image, this is my measurement of the CA18RNX, distance 1m, and 3mS window.
My measurements are in black, and measurements from Seas are in red. As you see the freq is good, but the phase is very different
Why?
I'm using Speaker Workshop to measure my drivers.
The results are good, both frequency and impedance.
But there is two things I dont really understand or getting right.
First; after measurements are done, I estimate S/T parameters, this works fine, and I get good results for FS, Vas, Qes, Qms, Qts and Re.
But Le is always very low, like 0,2mH, L1 says 0,3mH and Series Res is 13,7 ohm ? (this is measurments of a Seas CA18RNX)
I do the impedance measurements through a amp, because my soundcard has not enough output power.
Second; when measuring frequency, I get some strange (?) results of phase. Look at the attached image, this is my measurement of the CA18RNX, distance 1m, and 3mS window.
My measurements are in black, and measurements from Seas are in red. As you see the freq is good, but the phase is very different
Why?
Attachments
Regarding the acoustic measurements in your graph. The two measurements were probably measured at different distances. If you want to show the phase correctly, it is necessary to position the beginning of the window at the root of the beginning of the impulse. SW also has a feature to remove time of flight automatically. I am assuming you have gone through all the calibration procedures mentioned in the SW help file.
"Time of flight" is the time it takes for the wave to get from the source to the measuring mic.Norcad said:OK , that make sense, but I dont understand what "time of flight" means
And how can I get SW to remove this automatically?
I also wandering what "minimum phase" exactly means?
I can post some chart later, all my measurements are on my labtop.
Thanks!
To get SW to remove the time of flight:
With the frequency response plot highlighted, go to the menu Transform -> Delay... -> check the "remove access delay" with the delay time set to "0", and click okay. The result is not perfect all the time, and sometimes I go though this procedure two or 3 times. Regarding SW, you should be able to get more responses in the Audua forum.
Time of flight is simply the time it takes the sound to travel from the speaker to the microphone as it "flies" through the air. This is the additional delay everyone speaks of.
Can't speak for your software but the principle is to re-define t=0 or zero time somehow. I would guess you either enter a distance somewhere or enter a marker on an impulse response plot. Search your menus for some promising options.
Minimum phase - ahh. A tricky one. I think I might get flamed for going out on a limb here but in my mind.... minimum phase just means that the phase response is a function of the magnitude response (and vice versa) via the hilbert transform. It also has the lowest group delay for that particular magnitude response. I think people use minimum phase for everything that isn't linear phase (i.e. pure delay: phase is linear on a linear frequency axis)
Did you sort out your Le/Re problem? If you're not taking the amplifier into consideration in the calculations then that could introduce errors. Can you post an impedance plot?
Can't speak for your software but the principle is to re-define t=0 or zero time somehow. I would guess you either enter a distance somewhere or enter a marker on an impulse response plot. Search your menus for some promising options.
Minimum phase - ahh. A tricky one. I think I might get flamed for going out on a limb here but in my mind.... minimum phase just means that the phase response is a function of the magnitude response (and vice versa) via the hilbert transform. It also has the lowest group delay for that particular magnitude response. I think people use minimum phase for everything that isn't linear phase (i.e. pure delay: phase is linear on a linear frequency axis)
Did you sort out your Le/Re problem? If you're not taking the amplifier into consideration in the calculations then that could introduce errors. Can you post an impedance plot?
OK, so I had guessed right about "time of flight"
Its just the delay before the pulse.
The pulse starts at 4,7mS, and I set the markers at 4,5 and 7,5ms. At 7,6mS I got the first reflection.
I have done this in both Arta and SW, and the phase looks similar.
In Arta I have seen a choise of minimum phase, but I didnt know what it mean.
In SW I havent seen it, maybe I just have to look better
So the minimum phase, is what the phase would have been if the freq resp was flat? Or is it the other way around ?
I think you have right about the Le problem, I forgot to make the amp calibration when I started a new project. I must try that at see if its help, maybe that will do something with the phase also?
Most amps do have some phase issues in the upper freq right?
Its just the delay before the pulse.
The pulse starts at 4,7mS, and I set the markers at 4,5 and 7,5ms. At 7,6mS I got the first reflection.
I have done this in both Arta and SW, and the phase looks similar.
In Arta I have seen a choise of minimum phase, but I didnt know what it mean.
In SW I havent seen it, maybe I just have to look better
So the minimum phase, is what the phase would have been if the freq resp was flat? Or is it the other way around ?
I think you have right about the Le problem, I forgot to make the amp calibration when I started a new project. I must try that at see if its help, maybe that will do something with the phase also?
Most amps do have some phase issues in the upper freq right?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.