I would propose an alternative explanation: the increasing cross-sectional area of horns decreases total velocity, which creates an unfavorable pressure gradient that makes the flow along the walls more sensitive (high shear) to being tripped into turbulent flow, which has a higher associated acoustic frequency.
So if a horn diverges too quickly, at certain points in music, where the combination of frequency and loudness are "just right," you might get brief moments of soft hissing.
May well be like you describe @sjlee001 , as it was not a constant hiss but only at moments in the music.
Hello All,
The phase shield gizmo is a hard sharp obstruction in the throat of a tweeter. Is there a combination of frequency and loudness that generates some kind of harmonic noise?
I do not know, could be.
If you look at the data sheet of the SB Acoustics aluminum dome tweeter the coverage angle is wider than a tweeter without the phase shield gizmo. What I am wondering is there an audible/measurable trade off effect? Don't know! This is hobby DIY testing with real lab instruments.
The pair of SB Acoustics aluminum dome tweeter that I have here are in use. I have a second pair on the way from Madisound Speakers.
I suspect that if there is any added turbulence caused THD+N it will show up on axis.
The plan is to frequency sweep and level step while keeping an eye on the FFT output with the tweeter mounted to a flat baffle and a small assortment of waveguides and horns. The calibrated 1/4 inch GRAS microphone will be near field on axis.
Thanks DT
The phase shield gizmo is a hard sharp obstruction in the throat of a tweeter. Is there a combination of frequency and loudness that generates some kind of harmonic noise?
I do not know, could be.
If you look at the data sheet of the SB Acoustics aluminum dome tweeter the coverage angle is wider than a tweeter without the phase shield gizmo. What I am wondering is there an audible/measurable trade off effect? Don't know! This is hobby DIY testing with real lab instruments.
The pair of SB Acoustics aluminum dome tweeter that I have here are in use. I have a second pair on the way from Madisound Speakers.
I suspect that if there is any added turbulence caused THD+N it will show up on axis.
The plan is to frequency sweep and level step while keeping an eye on the FFT output with the tweeter mounted to a flat baffle and a small assortment of waveguides and horns. The calibrated 1/4 inch GRAS microphone will be near field on axis.
Thanks DT
Really nice. Is it a waveguide+cabinet simulation ? Have you got a picture of it ?
I wrote a blog post about time/phase alignment and the confusion surrounding these terms,
https://purifi-audio.com/blog/tech-notes-1/time-phase-alignment-acoustic-center-lobing-etc-14
https://purifi-audio.com/blog/tech-notes-1/time-phase-alignment-acoustic-center-lobing-etc-14
from above: "Vertical alignment of voice coils does not generally lead to even latency time nor a magic ‘time-aligned’ speaker. If you want linear phase (pure delay) you need a complex DSP solution or an extremely complex analog circuit. In all cases, measurements of the drivers mounted in their box are critical for a successful design."
I have intuitively always questioned physical alignment by voice coils. Perhaps it is a gross guide from PA speakers with compression drivers and horns. The designer needs to know acoustic centers, and only two-channel measurements will give that info, for simulation. With correct data you can make whatever xo topology you like.
https://kimmosaunisto.net/Software/VituixCAD/VituixCAD_Measurement_ARTA.pdf
I have intuitively always questioned physical alignment by voice coils. Perhaps it is a gross guide from PA speakers with compression drivers and horns. The designer needs to know acoustic centers, and only two-channel measurements will give that info, for simulation. With correct data you can make whatever xo topology you like.
https://kimmosaunisto.net/Software/VituixCAD/VituixCAD_Measurement_ARTA.pdf
Whilsth it is not so wrong to look at the voicecoil, the "time to wave" aka the time the movement energy travels from voicecoil to cone and starts the (sound) wave is not insignificant.
The approach as proposed by @kimmo with its Vituixcad is indeed quite usefull.
The 2 channel measurement with Arta gives also possibility to check .
I also used the approach to measure from a given mic position the mid and tweet separately and combined. Then to sum the separate measurement and add delay to match the combined measurement. Requires some calculation but doable. I used a special rig to get reproducable results.
The approach as proposed by @kimmo with its Vituixcad is indeed quite usefull.
The 2 channel measurement with Arta gives also possibility to check .
I also used the approach to measure from a given mic position the mid and tweet separately and combined. Then to sum the separate measurement and add delay to match the combined measurement. Requires some calculation but doable. I used a special rig to get reproducable results.
Are there any trends you saw, other than being shallow, in your optimisation study for your new tweeter that could be applied generally when hand designing? At the very least for CD in your design.
as always, it’s difficult to generalise. but i optimised a similar diameter WH for a plain non rounded box. The result was a deeper WG but also more shallow dispersion. probably the deeper WG hides the box edges better at the cost of more narrow dispersion. So heavy rounding makes quite a difference.
for good tweeter performance, I would start with a smooth transition without any major obstacles or profile changes from the tweeter to the waveguide throat. Attention should be paid to minimise any discontinuity from the waveguide to the cabinet. The shape of the profile of the waveguide determines the dispersion.
I mean this is probably no surprise to many. Some would say it is self evident.
I believe it is akin aerodynamics or fluid dynamics.