It begins as an attempt to remove some reflections that don't exist anyway, and ends as a tribute to the well-built dust cap.
The dust cap. It prevents dust and dirt from getting into the flywheel gap, where there are only a few tenths of a millimetre between the flywheel coil and the magnet pole piece.
One of the engineers had an idea that the reason for the pronounced peak around 5 kHz on Scan Speaks 15WU, stemmed from some reflections that should occur inside the shape of the diaphragm itself. The cure for the reflections would be to remove the dust cap of the speaker and then insert a rod of damping felt. Scan Speak carried out the experiment and delivered the result.
Unmodified woofer. Frequency response on axis, as well as 30 and 60 degrees off axis. In addition: impedance measurement.
The varied job description of the dust cap
It's a familiar and distinctive part of the woofer's appearance, and a frequent victim of prying fingers that might find a way to press it in. The dust cap can come in many sizes and shapes, but what they all have in common is that they are designed to prevent dust from getting into the voice coil gap, where there are only a few tenths of a millimetre between the voice coil and the pole piece of the magnet. A crumb of bread in here can, in the worst case, destroy the speaker. Depending on the speaker's construction, the dust cap may also act as a brace for the diaphragm, and it may be designed to contribute constructively to the speaker's sound. We look at this in more detail here.
Dust cap removed. You will notice that a bolt has been fitted through the pole piece, as well as a sealing washer. The vent holes in the voice coil form hardly influence the outcome of the experiment.
The dust cap is cut off
The experiment starts with a measurement on a standard version of the 15WU, so that the upcoming modifications can be documented. The dust cap is then cut off and a through bolt is fitted to hold the felt rod in place.
Dust cap removed.
Clear dive and reflections occur
Without the dust cap, there is a marked dip in the speaker's frequency response between 2-3 kHz, with a depth of 8-10 dB and a very abrupt progression. Above 8-10 kHz, the frequency response shoots up from the starting point. Part of this measurement result is of course due to the absence of the dust cap, but sound reflections on the magnet pole piece also occur and show up here.
Felt rod fitted. The felt is not compressed, as this gives the best damping/absorption characteristics at high frequencies.
Filt rod mounted
The through bolt is now extended and a stack of felt washers with a hole in the middle is fixed in the centre of the membrane. The "well" formed by the voice coil is now filled with felt and the felt rod itself is long enough to reach the leading edge of the membrane.
Felt rod fitted. Frequency response smoother, but otherwise unchanged.
Experiment fails
The subsequent measurement shows that the felt rod does not attenuate reflections inside the membrane shape. The frequency response is essentially the same as when the felt rod is not present, although slightly smoother. In general, however, the rather large amount of felt does not actually affect the frequency response at all and the experiment demonstrates no occurrence of reflections in the membrane shape itself. One can learn something after all.
The triumph of the dust cap
The example shows that the dust cover of a 15WU has a function besides protecting against dust and dirt. In this woofer, a significant part of the sound that will cooperate with the tweeter is emitted from this dust cap. There can be little doubt that the material properties of the dust cap and its size and shape have a decisive influence on the sound of the entire loudspeaker. The tonal range where the ear is most sensitive is actually reproduced by the dust cap of a Scan Speak 15WU. The experiment with the reflections ended in a dead end, but the curious pilferers are welcome to find something else to deform.
