1. Adding a linear damper to the spider.
Practically all spiders resonate on 150...500Hz frequencies, the internal damping is non-linear - looks like dry friction. If you add flexible glue and foam, the resonance will disappear.
2. Adding a linear damper to the cone.
Cones often exhibit non-linearly damped resonances, especially those made of aluminum. Adding a layer of the same foam (full or partial), and covering the foam with a thin layer of UV epoxy creates a sandwich that can be very effective in making the driver "picture-perfect". Here is SB12PAC before (red) and after (yellow). The dash-dotten lines are for frequency response, and solid lines for the 3rd harmonic.
The non-linear artifacts, which were polluting the upper tweeter range, are also diminished.
You can add the damper on the top of the cone, and under it, both with hardening ribs and without.
3. Minimize hysteresis distortions.
*** You need to disassemble the driver. ***
Most of the mid-range distortions are due to magnetic hysteresis in the pole piece. The same distortions are clearly visible on the current probe. The maximum influence is due to the material of the pole. Here are 2nd and 3rd harmonics if the pole is made out of electric steel, SB12PAC's, home-made (no pressure / heat) Quazy Soft Magnetic Composite out of 500um iron and epoxy, 500um and epoxy, and laminate plates from an EI transformer:
Unfortunately, making DIY SMC with a high saturation limit is non-trivial.
The distortions also depend on the geometry of the pole and voice coil. In the ideal case, you need to add a "reverse-current direction" voice coil directly on the top of the pole (inside or outside of the voice coil) to minimize the sum of the eddy currents, which is non-trivial for DIY. A much easier solution is to add lots of copper rings, both inside and outside the voice coil, ideally the same height and position as the voice coil itself.
Here are 2nd and 3rd distortions, different voice coil re pole locations, with more and more copper added:
While measuring current, keep in mind that most of the ready-made sub-Ohm resistors, even foil, are non-linear. I'd suggest making your own out of a copper wire.
You can experiment with many kinds of DIY poles - steel with sector cuts, steel with sector cuts filled with laminate, steel with sector cuts filled with SMC, SMC with laminate, etc. You are practically guaranteed to improve the driver a lot and exceed your wildest expectations.
Practically all spiders resonate on 150...500Hz frequencies, the internal damping is non-linear - looks like dry friction. If you add flexible glue and foam, the resonance will disappear.
2. Adding a linear damper to the cone.
Cones often exhibit non-linearly damped resonances, especially those made of aluminum. Adding a layer of the same foam (full or partial), and covering the foam with a thin layer of UV epoxy creates a sandwich that can be very effective in making the driver "picture-perfect". Here is SB12PAC before (red) and after (yellow). The dash-dotten lines are for frequency response, and solid lines for the 3rd harmonic.
The non-linear artifacts, which were polluting the upper tweeter range, are also diminished.
You can add the damper on the top of the cone, and under it, both with hardening ribs and without.
3. Minimize hysteresis distortions.
*** You need to disassemble the driver. ***
Most of the mid-range distortions are due to magnetic hysteresis in the pole piece. The same distortions are clearly visible on the current probe. The maximum influence is due to the material of the pole. Here are 2nd and 3rd harmonics if the pole is made out of electric steel, SB12PAC's, home-made (no pressure / heat) Quazy Soft Magnetic Composite out of 500um iron and epoxy, 500um and epoxy, and laminate plates from an EI transformer:
Unfortunately, making DIY SMC with a high saturation limit is non-trivial.
The distortions also depend on the geometry of the pole and voice coil. In the ideal case, you need to add a "reverse-current direction" voice coil directly on the top of the pole (inside or outside of the voice coil) to minimize the sum of the eddy currents, which is non-trivial for DIY. A much easier solution is to add lots of copper rings, both inside and outside the voice coil, ideally the same height and position as the voice coil itself.
Here are 2nd and 3rd distortions, different voice coil re pole locations, with more and more copper added:
While measuring current, keep in mind that most of the ready-made sub-Ohm resistors, even foil, are non-linear. I'd suggest making your own out of a copper wire.
You can experiment with many kinds of DIY poles - steel with sector cuts, steel with sector cuts filled with laminate, steel with sector cuts filled with SMC, SMC with laminate, etc. You are practically guaranteed to improve the driver a lot and exceed your wildest expectations.
