User manual of 5.25 pro tells that:
So it's just multiplier for Vb and conversion from fill-% to Qa and vice versa. One problem is that effect could be actually opposite with dense materials. I've seen smarter approaches in some applications. Wouldn't be impossible to add some complex transfer function for Qa and Vb, but equivalent circuit should remain the same to avoid massive amount of really nonproductive work.
So it's just multiplier for Vb and conversion from fill-% to Qa and vice versa. One problem is that effect could be actually opposite with dense materials. I've seen smarter approaches in some applications. Wouldn't be impossible to add some complex transfer function for Qa and Vb, but equivalent circuit should remain the same to avoid massive amount of really nonproductive work.
^Not so much step-by-step instructions available. Recommended design procedure for conventional boxed speakers includes Diffraction simulation.
I'm not sure what you include in room gain, but room acoustic simulation including DC mode is not available. Just in-room response simulation either by CTA-2034-A or with 1...4 earliest room reflections specified and activated in Room tab.
You can also create estimated transfer function for DC mode (and the lowest mode) with generic active filters or load response file created with some external tool into transfer function block. "Room gain" block is located in the beginning of XO circuit and shorted when not needed.
No plans to make complete room simulation. For example I don't design by features of some individual room because that policy is not universal enough.
I'm not sure what you include in room gain, but room acoustic simulation including DC mode is not available. Just in-room response simulation either by CTA-2034-A or with 1...4 earliest room reflections specified and activated in Room tab.
You can also create estimated transfer function for DC mode (and the lowest mode) with generic active filters or load response file created with some external tool into transfer function block. "Room gain" block is located in the beginning of XO circuit and shorted when not needed.
No plans to make complete room simulation. For example I don't design by features of some individual room because that policy is not universal enough.
Dear Kimmo! And yet ... It seems to me that it would be very useful to introduce into your program the display of the graph for the step response of the finished system. I consider this a very important parameter for control during the construction of the crossover.
These quotes are from an interview with John Dunlavy Stereophile magazine.
"It may come as a surprise—this is giving away a trade secret—but when I design a loudspeaker, I first design it by looking at the step response. I find that by playing around with the crossover network while observing the step response in real time, any change I make is immediately available. When I get the step response right, everything else goes along. It's implicit. It goes along for the ride."
I consider this a very important parameter! What do you think?
Thanks!
These quotes are from an interview with John Dunlavy Stereophile magazine.
"It may come as a surprise—this is giving away a trade secret—but when I design a loudspeaker, I first design it by looking at the step response. I find that by playing around with the crossover network while observing the step response in real time, any change I make is immediately available. When I get the step response right, everything else goes along. It's implicit. It goes along for the ride."
I consider this a very important parameter! What do you think?
Thanks!
***I saw this! Thanks! However, it is not convenient to control in real time by opening a new window! It would be more convenient to see this next to other main charts!!
^Phase response and excess group delay are at least as valid primary design parameters as step response which is actually result of IFFT calculation when source data is in frequency domain.
So if your design target is close to ideal step response, you can design by adjusting normal phase response to minimum phase response and monitoring that excess group delay is close to zero. First two curves are visible in SPL chart and the last one in GD & Phase -> additional step response chart is not needed while XO design.
Function of IR window is to provide export for FIR applications.
So if your design target is close to ideal step response, you can design by adjusting normal phase response to minimum phase response and monitoring that excess group delay is close to zero. First two curves are visible in SPL chart and the last one in GD & Phase -> additional step response chart is not needed while XO design.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Function of IR window is to provide export for FIR applications.
The latest build few minutes ago.
Rev. 2.0.47.3 (2020-04-11)
Main
* Export -> FIR transfer function for Driver changed to use axial response of driver including location, rotation and tilt i.e. X,Y,Z,R,T parameters. Improves accuracy with large constructions and side woofers.
Driver to be exported is selected with combo box in Optimizer window. No need to select also from crossover.
* Transfer function block G(f) selected in crossover is shorted before calculation of exported data. Exported transfer function file is loaded to selected Transfer function block G(f) and block is non-shorted.
Rev. 2.0.47.3 (2020-04-11)
Main
* Export -> FIR transfer function for Driver changed to use axial response of driver including location, rotation and tilt i.e. X,Y,Z,R,T parameters. Improves accuracy with large constructions and side woofers.
Driver to be exported is selected with combo box in Optimizer window. No need to select also from crossover.
* Transfer function block G(f) selected in crossover is shorted before calculation of exported data. Exported transfer function file is loaded to selected Transfer function block G(f) and block is non-shorted.
Both demo projects Kontiainen and Epe-3W include XO variant with exported FIR transfer function for Drivers loaded into G(f) blocks:
An externally hosted image should be here but it was not working when we last tested it.
^At least it doesn't look impossible because current is usually calculated just one program line before power. Amplifier and buffer outputs are a bit different, but doable. C is not yet included in the list of resistances.
Most challenging is to add new feature so that it's understandable for others
Most challenging is to add new feature so that it's understandable for others
Rev. 2.0.48.0 (2020-04-13)
Optimizer
* Added Tilt dB/oct text box for Axial response of Driver target curve, located below SPL text box. Tilt limits -1...+1 dB/oct.
* Searching for Tilt dB/oct from total SPL target line added to binocular button.
* Added 14 deg/dB/oct offset to phase response target if HP or LP slope is not linear-phase.
* Ratio of HP and LP frequencies limited to avoid errors in tilt handling.
Power dissipation
* Added Show Current checkbox to calculate current spectrum instead of power dissipation.
* Added Inductors checkbox to Show curves group. Separated from resistors.
* Added Capacitors checkbox to Show curves group, primarily for current.
Rev. 2.0.48.1 (2020-04-14)
Main
* Added SPL max checkbox and text box for scale adjustment of SPL, Power&DI and unnormalized Directivity graphs. Manual value is saved to project file.
* Microphone offset group moved from Drivers tab to Room tab. More logical close to other listening/mic point parameters.
* Microphone offset group disabled when in-room response simulation is activated by reflection checkboxes. Geometry is easier to manage without multiple overlapping location and angle parameters.
Main
* Added SPL max checkbox and text box for scale adjustment of SPL, Power&DI and unnormalized Directivity graphs. Manual value is saved to project file.
* Microphone offset group moved from Drivers tab to Room tab. More logical close to other listening/mic point parameters.
* Microphone offset group disabled when in-room response simulation is activated by reflection checkboxes. Geometry is easier to manage without multiple overlapping location and angle parameters.
I have a question and an observation:
Question: Is it possible to change the reference angle? I am able to type in new angle (e.g., 10 or -10) into the box but after entering the number it remains on 0 deg hor.
Observation: When I try to change the value of an inductor, the series resistance also changes (direct numeric input or mouse scroll wheel). I can get the series resistance back to the original value or close by changing wire size.
Question: Is it possible to change the reference angle? I am able to type in new angle (e.g., 10 or -10) into the box but after entering the number it remains on 0 deg hor.
Observation: When I try to change the value of an inductor, the series resistance also changes (direct numeric input or mouse scroll wheel). I can get the series resistance back to the original value or close by changing wire size.
Could it be because I have loaded drivers into another crossover variant in my project that only have 0 deg data?
Edit: file name is DJ_tweet.frd
Last edited: