I tried something similar on a PA130-8.
https://www.diyaudio.com/community/threads/the-daytonator-pa130-8x.279043/
Slicing should be done in odd number like 5 to avoid reflection symmetry. The cones of TC9FD and SS 10F are pentagonal shaped even (glued on a circular surround) and this helps flatten the breakup peaks. Revelator has 5 slices. Use of flexible glue like RTV will probably dramatically change things.
A phase plug and its shape seems to have a bigger impact. In the end I added a coaxial tweeter to where the dustcap was.
https://www.diyaudio.com/community/threads/the-daytonator-pa130-8x.279043/
Slicing should be done in odd number like 5 to avoid reflection symmetry. The cones of TC9FD and SS 10F are pentagonal shaped even (glued on a circular surround) and this helps flatten the breakup peaks. Revelator has 5 slices. Use of flexible glue like RTV will probably dramatically change things.
A phase plug and its shape seems to have a bigger impact. In the end I added a coaxial tweeter to where the dustcap was.
Depends on the varnish. Spar is softer than hard. Poly softer than Spar.
When removing dust caps, one must consider the pole piece venting and if you are creating a place to whistle.
It might have been a better approach to find a couple of cheap speakers to use as guinea pigs if you want to experiment with doping or slicing cones. Years ago I tried doping cones of my speakers which had Phillips 20 cm paper cones - this after seeing a kit using these with a different crossover network and reportedly using carpet glue as the dopant. I did not like the result however in hindsight I should have applied the glue diluted with water in multiple coats.
Since then I have tried ladies nail varnish on paper cones by that process. That did no harm because I did not push the boundaries. Talking of boundaries by slicing the cone you could have tried narrow strips of removable painters tape on the back of the cone.
It looks a little flatter than original between 1.5 and just over 3kHz range where the ear is sensitive. This might work better depending where the crossover to a tweeter is placed. Cosmetically you could have done a tidier job, possibly with some suitable material glued or adhesive tapes across the cuts at the back of the cone. Maybe you should buy some cheap speakers to use as guinea pigs. I recently used removable painters masking tape to stabilise the disc drive of a CD player. I had bought this at an op shop and worked well for a while but started skipping. Opening it up I found that a cover to clamp the disc was not there. The movement of the motor did not affect the effectiveness of the tape.
This has worked well in the past. Of course the people telling us how well this works, would never tell us what they actually use for glue. Here's an example of a very good result:
https://techtalk.parts-express.com/forum/tech-talk-forum/66754-how-to-train-your-pm180-8
This rabbit hole led me to the archived pages of Mark McKenzie. He even used to post on this forum, but I can hardly find a single image of anything he ever created.
http://web.archive.org/web/20200129121454/http://madspeaker.com/page2.html
Older archives have a crafts section on his site, but it's quite useless since like all such "archived" sites the images are all gone.
Does anything of Mark remain?
http://web.archive.org/web/20200129121454/http://madspeaker.com/page2.html
Older archives have a crafts section on his site, but it's quite useless since like all such "archived" sites the images are all gone.
Does anything of Mark remain?
Just my 2C on some points mentioned here.
Electrical filtering. Its possible to entirely 'notch' out resonances on-axis from a driver but the tendency to 'ring' is still there in the driver assembly. Resonances will remain apparent in the harmonic distortion behaviour of the driver. I might notch out a peak at 5khz, but the cone will still show a higher 2nd harmonic at 2.5khz (2.5khz *2).
Doping or driver modifications can damp breakup and the associated harmonic excitation, but we will usually do this at expense of top end response tha ks to increase in weight or reduction in material stiffness. A loss in SPL will also be associated with increased mass. Its easy just to shift problems around in frequency and 'Q' but not actually fix anything.
Phase plugs can do a lot to improve both on and off-axis response. However they do not alter the total energy output at a specific frequency. They do redistribute the energy over space (dispersion angle). Like electrical filters, they do not change structural resonances or their tendancy for harmonic excitation. Generally phase plugs are used to equalise path lengths, or inhibit interference between, more than one sound source such as two sections of a cone that are moving out of phase due to resonant modes or mass imbalance.
I'd estimate most resonant breakup in cones is largely down to reflection of the sound wave as it travels from the center of the cone to the outer rim and encounters a sudden change of material density (I.e. the cone does not continue forever, it stops at the outer edge) this causes the wave to reflect back towards the center of the cone and interference with the next part of the sound wave. Messy behaviour ensues. The rubber / foam / paper surround job is to damp that high frequency energy when it reaches the edge of the cone, while also not limiting the general linear motion of the cone. There are also circular resonance modes that set up around the rim.
The combination of all these factors gets mighty complex, so a bit of practicle experimentation with a specific case is certainly worth trying and doesn't necessarily apply to every other case.
Electrical filtering. Its possible to entirely 'notch' out resonances on-axis from a driver but the tendency to 'ring' is still there in the driver assembly. Resonances will remain apparent in the harmonic distortion behaviour of the driver. I might notch out a peak at 5khz, but the cone will still show a higher 2nd harmonic at 2.5khz (2.5khz *2).
Doping or driver modifications can damp breakup and the associated harmonic excitation, but we will usually do this at expense of top end response tha ks to increase in weight or reduction in material stiffness. A loss in SPL will also be associated with increased mass. Its easy just to shift problems around in frequency and 'Q' but not actually fix anything.
Phase plugs can do a lot to improve both on and off-axis response. However they do not alter the total energy output at a specific frequency. They do redistribute the energy over space (dispersion angle). Like electrical filters, they do not change structural resonances or their tendancy for harmonic excitation. Generally phase plugs are used to equalise path lengths, or inhibit interference between, more than one sound source such as two sections of a cone that are moving out of phase due to resonant modes or mass imbalance.
I'd estimate most resonant breakup in cones is largely down to reflection of the sound wave as it travels from the center of the cone to the outer rim and encounters a sudden change of material density (I.e. the cone does not continue forever, it stops at the outer edge) this causes the wave to reflect back towards the center of the cone and interference with the next part of the sound wave. Messy behaviour ensues. The rubber / foam / paper surround job is to damp that high frequency energy when it reaches the edge of the cone, while also not limiting the general linear motion of the cone. There are also circular resonance modes that set up around the rim.
The combination of all these factors gets mighty complex, so a bit of practicle experimentation with a specific case is certainly worth trying and doesn't necessarily apply to every other case.
Mark McKenzie together with a certain Spangler wrote an AudioXpress article (around 2002 or so) about drastic improvements of a 3" TangBand polyconed FR driver. The surgery involved replacement of the -ringing- dustcap by a carefully tailored and shaped phase plug/diffuser, stabilization of the inside of the rubber surround by the well known "ring of glue" technique, plus -probably most important- two rings of dimples in the cone. If the somewhat small pictures of the measurements may be believed, this modification should yield one of the very best full ranges, where the drastic removal of cone (and surround interference) break-up has lead to increased bandwidth beyond 20kHz within a few dB, thus a ruler flat SPL.
Most important: these mods were not random hobbyist trials, but every of the three mods was supposed to be the result of cone-dustcap-surround vibration analysis. The article is unfortunately not specfic about the software used, but we may safely assume McK used FEM analysis or/and had access to laser interferometric equipment.
@keantoken : do you have acces to the work with the 10" drivers McK. did?
Most important: these mods were not random hobbyist trials, but every of the three mods was supposed to be the result of cone-dustcap-surround vibration analysis. The article is unfortunately not specfic about the software used, but we may safely assume McK used FEM analysis or/and had access to laser interferometric equipment.
@keantoken : do you have acces to the work with the 10" drivers McK. did?
A good video of FineCone software to design cones, dustcaps and phase plugs, by Peter Larsen.
Slicing is not in the demo. Slices and added glue rims etc. cone edge treatments will add Mms naturally, but reduction of efficiency in usable bandwidth is minimal.
SB Acoustic uses added mass on the backside of the cone edge on some models "ring of glue"
https://www.diyaudio.com/community/threads/sb-acoustics-satori-monitor.220330/page-45#post-4847329
Slicing is not in the demo. Slices and added glue rims etc. cone edge treatments will add Mms naturally, but reduction of efficiency in usable bandwidth is minimal.
SB Acoustic uses added mass on the backside of the cone edge on some models "ring of glue"
https://www.diyaudio.com/community/threads/sb-acoustics-satori-monitor.220330/page-45#post-4847329
Found the McK patent referred to:
https://patentimages.storage.googleapis.com/e6/79/69/7fd36a8aa85545/US20070092101A1.pdf
https://patentimages.storage.googleapis.com/e6/79/69/7fd36a8aa85545/US20070092101A1.pdf
His Tangband thread on this forum also shows a slit cone method, he says both will result in a very good driver.
https://www.diyaudio.com/community/threads/markmcks-tang-band-mods.28355/
Mark specifically asserted that simulation was not good enough to do this work. He mentioned laser measurements and sound pressure measurements so I speculate his method was entirely measurement based maybe with a sprinkle of math.
No, it has been difficult to find a way to access his material. I made a thread where we can collect stuff:
https://www.diyaudio.com/community/...imple-cones-and-madsound.381060/#post-6890582
BTW: Tenson I see you act all innocent but there you are in the old Tangband thread talking it up with McKenzie himself!
Cone break up would not be helped by uneven pressure inside the box which arises in the corners and where space is tight. With small diameter speaker drivers a 19mm depth of MDF puts the basket and cone in a tight spot if the hole in the baffle is a straight cut.
Sculpting material away from the inside edges of the hole or rounding/contouring the edges of the hole to increase air flow and reduce pressure are well known strategies but it is not clear if such have been applied in this case.
I don't know if boxes made from MDF can make the most of Scanspeak slit cone drivers.
Sculpting material away from the inside edges of the hole or rounding/contouring the edges of the hole to increase air flow and reduce pressure are well known strategies but it is not clear if such have been applied in this case.
I don't know if boxes made from MDF can make the most of Scanspeak slit cone drivers.