As is often the case...a "good catch" because I have made the same error before.
If you have the chance, post the far-field response to the Google drive folder.
The GOOD NEWS...as you can see in dcibel's post above, when you get the driver and port areas corrected AND add the Diffraction Response, that trough at 300Hz mostly goes away.
For diffraction: Select Tools\Diffraction in the ribbon menu.
1. Enter your baffle width, height, edge treatment, and Sd and press "New" [Red circles below]
2. You can click on the woofer circle or mic icon and drag them around when they turn blue. This is convenient if you want to drag a tweeter around to look at the diffraction effects of being offset or closer/further from the top edge. I just select the woofer and enter the position in mm in the "View" section [Green rectangle] and then select the mic icon and enter the same position.
3. Set the distance to 5-30 meters (5,000-30,000mm) and Export [Blue circles] for the "Low frequency part" (i.e., the top portion of the Merger screen), I have used 10,000mm. Change the distance to 1 meter (1,000 mm) and Export for the "High frequency part" (i.e., the middle portion of the Merger screen) that we haven't used yet but will use when we add the farfield measurement(s).
(1) I don't think you need to add any delay at all to the port
(2) Add "Diffraction response 1000mm" to the line below all of your farfield measurements
(3) Move your transition frequency to 400Hz or maybe 350Hz. You may not have to adjust your scaling that you currently have.
(4) Try blending over 1, 2 and 3 octaves.
(3) and (4) are just to smooth out that trough just a little bit more. It may or may not be there in reality, but you can't do anything about it anyways. But I would find it distracting when working on the xo, so better to just smooth it out a little and forget it.
@4thtry said:
How did you calculate the -58mm port delay? That's only about 2.2 inches.
Thanks Bill!
Man oh man, I swear that I am actually a college graduate, and (mostly) still in possession of my faculties.
So I punched in the conversion into an app. Hindsight....... sheesh.
The actual path offset is 20 inches.....
But Chahly - Stahkist don't want speakers that look good, Stahkist wants speakers that sound good!
@a4eaudio said:
(1) I don't think you need to add any delay at all to the port
(2) Add "Diffraction response 1000mm" to the line below all of your farfield measurements
(3) Move your transition frequency to 400Hz or maybe 350Hz. You may not have to adjust your scaling that you currently have.
(4) Try blending over 1, 2 and 3 octaves.
(3) and (4) are just to smooth out that trough just a little bit more. It may or may not be there in reality, but you can't do anything about it anyways. But I would find it distracting when working on the xo, so better to just smooth it out a little and forget it.
Excellent tips, thanks!
I really like "can't do anything about it, just distracting"
But Chahly - Stahkist don't want speakers that look good, Stahkist wants speakers that sound good!
Ugh...I will try to explain one more time, then I'll leave this thread alone, Don knows where to find me if he wants my help. I hope he realizes that calculator screenshot say centimetres on it at least. Delay value is physical difference in mic location for each nearfield measurement mic location, relative to typical listening distance of 3-4m. It doesn't need to be very precise. For a bookshelf speaker such as this, you can assume straight line path, just use cabinet depth of 12" or 300mm as delay value for port measurement to place it in physical space relative to the woofer nearfield measurement.
You can move merge point and blend all you want to "hide it" in the squiggly's but resonant mode still exists in reality. Real suggestion might be to add more stuffing or try changing location of stuffing in the cabinet at the expense of low end output.
I'm sure that we have all made a number of simple decimal point or +/- sign mistakes when rushing to make a thread post. Myself included, of course. Don't worry about it, Don. As long as you end up at the right decision in the end, it is all OK.
Thanks dcibel - I have challenged your patience a number of times!
But now, I have successfully constructed "the" jig and completed dual channel measurements.
Making progress....
But Chahly - Stahkist don't want speakers that look good, Stahkist wants speakers that sound good!
Comments
As is often the case...a "good catch" because I have made the same error before.
If you have the chance, post the far-field response to the Google drive folder.
The GOOD NEWS...as you can see in dcibel's post above, when you get the driver and port areas corrected AND add the Diffraction Response, that trough at 300Hz mostly goes away.
For diffraction: Select Tools\Diffraction in the ribbon menu.
1. Enter your baffle width, height, edge treatment, and Sd and press "New" [Red circles below]
2. You can click on the woofer circle or mic icon and drag them around when they turn blue. This is convenient if you want to drag a tweeter around to look at the diffraction effects of being offset or closer/further from the top edge. I just select the woofer and enter the position in mm in the "View" section [Green rectangle] and then select the mic icon and enter the same position.
3. Set the distance to 5-30 meters (5,000-30,000mm) and Export [Blue circles] for the "Low frequency part" (i.e., the top portion of the Merger screen), I have used 10,000mm. Change the distance to 1 meter (1,000 mm) and Export for the "High frequency part" (i.e., the middle portion of the Merger screen) that we haven't used yet but will use when we add the farfield measurement(s).
So, I still have a dip 275-400, but it looks better than it did. Not as good as dcibel's example though.
Added in the far field measurements and merged. Looks pretty good to me. Comments?
How did you calculate the -58mm port delay? That's only about 2.2 inches.
(1) I don't think you need to add any delay at all to the port
(2) Add "Diffraction response 1000mm" to the line below all of your farfield measurements
(3) Move your transition frequency to 400Hz or maybe 350Hz. You may not have to adjust your scaling that you currently have.
(4) Try blending over 1, 2 and 3 octaves.
(3) and (4) are just to smooth out that trough just a little bit more. It may or may not be there in reality, but you can't do anything about it anyways. But I would find it distracting when working on the xo, so better to just smooth it out a little and forget it.
Thanks Bill!
Man oh man, I swear that I am actually a college graduate, and (mostly) still in possession of my faculties.
So I punched in the conversion into an app. Hindsight....... sheesh.
The actual path offset is 20 inches.....
Excellent tips, thanks!
I really like "can't do anything about it, just distracting"
Ugh...I will try to explain one more time, then I'll leave this thread alone, Don knows where to find me if he wants my help. I hope he realizes that calculator screenshot say centimetres on it at least. Delay value is physical difference in mic location for each nearfield measurement mic location, relative to typical listening distance of 3-4m. It doesn't need to be very precise. For a bookshelf speaker such as this, you can assume straight line path, just use cabinet depth of 12" or 300mm as delay value for port measurement to place it in physical space relative to the woofer nearfield measurement.
You can move merge point and blend all you want to "hide it" in the squiggly's but resonant mode still exists in reality. Real suggestion might be to add more stuffing or try changing location of stuffing in the cabinet at the expense of low end output.
I'm sure that we have all made a number of simple decimal point or +/- sign mistakes when rushing to make a thread post. Myself included, of course. Don't worry about it, Don. As long as you end up at the right decision in the end, it is all OK.
Thanks dcibel - I have challenged your patience a number of times!
But now, I have successfully constructed "the" jig and completed dual channel measurements.
Making progress....