The Auras i have are some 10 years old… one would expect teh drivers to evolve so i am not surprised.
dave
I wonder how the mic inside the port might cause a confounder? I guess I wouldn't assume the frequency response would be exactly the same at any point in the port, especially compared to just outside the exit of the port? But I don't have the expertise to know for sure.
If you try more ideas, I would include using butyl-backed aluminum sheets to function as the CLD, since it is easily implemented by others.
If you try more ideas, I would include using butyl-backed aluminum sheets to function as the CLD, since it is easily implemented by others.
It is not aerodynamic/chuffing noise from turbulence because the shape does not follow that required by the physics of turbulence. We can see this spectrum in the first plot in #349 of stv's thread (as well one or two other things creating sound). The sound in your plot is primarily coming from something else which, as you suggest, may be the flexing of the port walls. It may be combining with the acoustics within the cabinet. It may be leakage from the rear of the driver. Possibly one or two other things. Determining what is going on from a small number of measurements can often be difficult and sometimes there simply isn't enough information to know.
The first pair of graphs (which I unwisely called a plot) show aerodyamic sound from the turbulent breakdown of the vortex streets being shed by the Helmholtz resonance. In the first the coherent vortices have barely started to form turbulent vortices in other orientations before the energy gets dissipated by viscosity into heat. In the second as the speed is raised the turbulent energy cascade gets going with the generating and dissipating scales moving further apart and the dissipating scales moving to higher frequencies. At the highest speed and highest frequency there seems to be another sound generating mechanism present apart from the Helmholtz mass source, the turbulent breakdown of the vortex streets and the acoustic port modes. The last graph for the straight port has no identifiable aerodynamic sound from turbulence just the coherent mass source, acoustic port modes and something else.
So I'm not entirely sure either but I know how to find out though it is taking rather longer than I had initially expected. There's a surprise.
So I'm not entirely sure either but I know how to find out though it is taking rather longer than I had initially expected. There's a surprise.
I tried flipping the baffle 180 degrees on the top ported version and the ~600 Hz mode came back
There goes my hypothesis about a standing mode forming at the bend! (The commercial example is still very interesting )On the positive side, this opens another means to optimise the response, because with the port exit on the top I have a little space to play with woofer position on the front baffle.
Looks like, despite my early concussions, this design study is not quite finished!
I suppose, the fact the port material construction affects that 600 Hz mode means the port walls are not only acting to absorb energy inside the port but also inside the cabinet. It could even be the primary means of operation.
Thanks for keeping my on the 'truth path!'.
Making an approximate mesh of the speaker internal shape I'm trying to use AKABAK to simulate the internal cabinet modes. I think @andy19191 has said this is not ideal but the results pass a sniff test.
I would expect a length-wise standing mode around 950Hz and that is what we see. I set up 3 planes to help see modes that are not in the X, Y plane but most of it does seem to be there.
Any tips on getting good results is appreciated I've included the AKABAK file.
For example at about 700 Hz we see a mode that is propagating through the 3 planes. We also see higher pressure near the top of the cabinet than the bottom.
I would expect a length-wise standing mode around 950Hz and that is what we see. I set up 3 planes to help see modes that are not in the X, Y plane but most of it does seem to be there.
Any tips on getting good results is appreciated I've included the AKABAK file.
For example at about 700 Hz we see a mode that is propagating through the 3 planes. We also see higher pressure near the top of the cabinet than the bottom.
A BEM simulation won't include the mass sound source from the Helmholtz resonance or the aerodynamic sound from turbulence when present. Both sources of sound occur both outside the cabinet and within it. BEM also won't be including damping from stuffing and other sources of friction (likely small). What comes out the port will be undamped "leakage" from within the cabinet. The vibration of the port and cabinet won't be included either which we know to be significant given the effect of the CLD. The acoustic resonances in the port, cabinet and their interaction will be included although the magnitudes of the modes won't be correct because they will only be driven by sound from the cone and not the Helmholtz resonance or turbulence. However, these should be fairly small at the higher frequencies where the acoustic resonances are occurring. So it stands a fair chance of helping our understanding of what, for example, is causing a null. This assumes the port is included in the geometry obviously. A wall across the mouth of the port is likely to create an acoustic model that is a long way from that of the cabinet plus port that is of interest.
BEM solvers can have numerical issues with long thin shapes like the port. There are things that can be done to alleviate this but what is included in AKABAK I don't know but hopefully it is in the manual and/or users of AKABAK may know. I am assuming you haven't bought the software at full price and so cannot ask the author for support?
You can use BEM to find modes and mode shapes but whether AKABAK support this I don't know. Anyone? What you seem to be plotting is the forced response by the driver at a couple of frequencies. This will include all the modes (one for each element if you are using zeroth order elements) but only a few will be contributing significantly.
With luck AKABAK may be able to contribute significantly to understanding the acoustic response at the higher frequencies. Good stuff.
BEM solvers can have numerical issues with long thin shapes like the port. There are things that can be done to alleviate this but what is included in AKABAK I don't know but hopefully it is in the manual and/or users of AKABAK may know. I am assuming you haven't bought the software at full price and so cannot ask the author for support?
You can use BEM to find modes and mode shapes but whether AKABAK support this I don't know. Anyone? What you seem to be plotting is the forced response by the driver at a couple of frequencies. This will include all the modes (one for each element if you are using zeroth order elements) but only a few will be contributing significantly.
With luck AKABAK may be able to contribute significantly to understanding the acoustic response at the higher frequencies. Good stuff.
Hello Everyone!
The design study has been significantly revised to include more information and correct some erroneous conclusions. I added a study of the effect of woofer position on the baffle.
3D Printed Loudspeaker Port Design Study V2
The design study has been significantly revised to include more information and correct some erroneous conclusions. I added a study of the effect of woofer position on the baffle.
3D Printed Loudspeaker Port Design Study V2
That is useful. I hadn't realised that Dayton now provide information on the nonlinearity of their drivers. Have any other manufacturer's started doing this as well?
It makes putting together a nonlinear model rather more useful. Doesn't appear to be anything to directly support a thermal model but indirectly one could probably do something with the variation in linear parameters under hot and cold conditions. Good stuff.
It makes putting together a nonlinear model rather more useful. Doesn't appear to be anything to directly support a thermal model but indirectly one could probably do something with the variation in linear parameters under hot and cold conditions. Good stuff.