A couple of things jump out at me. First, it appears that a flatter diaphragm has fewer resonances, which would be a big advantage. Kind of dispels the idea of a deeper draw on the diaphragm, until diaphragm modes become more problematic than phase plug modes.
Second, for the example shown above, the disturbances are not significant until about 15 kHz. Is that really worth worrying so much about? (its a 2.3 % difference at 8 kHz. or about -32 dB. This would be very hard to measure let alone hear.
I do agree that this kind of analysis is exactly what is needed if one is trying to squeeze out the last few details of a design, even though I highly doubt that these improvements would be audible. But they do make for good marketing!
Second, for the example shown above, the disturbances are not significant until about 15 kHz. Is that really worth worrying so much about? (its a 2.3 % difference at 8 kHz. or about -32 dB. This would be very hard to measure let alone hear.
I do agree that this kind of analysis is exactly what is needed if one is trying to squeeze out the last few details of a design, even though I highly doubt that these improvements would be audible. But they do make for good marketing!
It should be noted that the example above is for an ideal situation where the chamber is driven uniformly across its whole face. If the surround is incorporated, i.e the excursion is progressively reduced to zero at the egde, it get considerably worse (the first picture shown).
A already tried a deeper draw, it's not that worse in fact (see the image attached). I can imagine it might have an advantage in the end. Hard to tell, the Celestion data doesn't actually look that better in the datasheets.
A already tried a deeper draw, it's not that worse in fact (see the image attached). I can imagine it might have an advantage in the end. Hard to tell, the Celestion data doesn't actually look that better in the datasheets.
How did you get this number? To me it looks like there might be otherwise a difference in the input pressures of several dB for the two channels. So I probably still not get it.
Attachments
Last edited:
OK, but one wants to drive all the channels with the same pressure, no? If ignoring these modes (which are excited even without the channels just because of the cavity curvature - I wouldn't think of that a few days ago), severe disturbances might be introduced just by this. At least this is how I read the results. I might be wrong about the importance in the overall picture however.
I only have doubts about the summing of the individual exits of the slots. It is my impression at the moment that for this to work properly, they should share the same output area, overlap in fact.
I only have doubts about the summing of the individual exits of the slots. It is my impression at the moment that for this to work properly, they should share the same output area, overlap in fact.
Last edited:
This may be the funniest thing I have ever read on this entire site or any audio website. I have been following this thread since the beginning and don't understand 99% of it - I am here to look at the pretty pictures with their cool shapes and colors. I feel like my 3 yo reading with her older brothers and sisters.
But it is extremely interesting and glad its happening on this forum, where I can tag along and try to understand more of it. Thanks and I will go back to lurking and learning, your efforts are appreciated.
I think it might have been that a semicolon was missing as I don't get the error on import if I change the line to:
Infinite_Baffle
Subdomain=2; Position=z; offset=80mm
(added semi colon after Position=z) and when I go in the subdomain tab of AKABAK I see there there are two subdomains and one has the check box infinite baffle checked.
however things arn't working still as far as generating plausible simulation results and the akabak project files are binary so its a bit difficult to check what the differences are.
Sorry, I misread the units on the graph, It would be 6 dB difference in 260 dB, a 2.3% difference in dB. It's not clear to me that this would mean a 6 dB problem in the radiated sound - I would think not. The differences could average out yielding no net change in radiation. Or they could couple to a HOM making it worse if there are HOM propagation at 8 kHz.
I would think that the channels should all have the same volume velocity. If the channels all have the same area then they should have equal pressure, but if the areas differ, then so should the pressures.
Correction, chapter 4 & 5 deal with (Smith related) phase plugs for compression drivers, while chapter 7 covers the KEF tangerine phase plug.
The dissertation is a monumental work of acoustical engineering. However, after rereading large parts of it, the question remains whether some of the (overly) complicated analyses are necessary, or even useful if alternative solutions are obvious. This is especially true for chapter 7 and to a lesser extent chapter 8.
Apart from a 3dB increase in power response, the added advantage of the KEF tangerine phaseplug seems questionable.
But hey, it looks cool so it'll probably sell
Last edited: