Keithj01, keantoken, cone resonances and alike cannot be electrically damped, electrical damping works only at driver main resonance, please scroll down to heading "The assumed control of cone motion" for explanation:
https://www.edn.com/loudspeaker-operation-the-superiority-of-current-drive-over-voltage-drive/
https://www.edn.com/loudspeaker-operation-the-superiority-of-current-drive-over-voltage-drive/
That applies to the cone as viewed as a lumped mass (i.e. at low frequencies). At high frequencies, the cone acts as a transmission plane and looks resistive. Though the voice-coil inductance and resistance "get in the way" as far as damping is concerned.
George, I am working with a friend to produce an article for AudioXpress.
We haven't looked at a loadline yet, not sure how we would do that?
One other thing we see (expected) is that with increasing field power input, efficiency goes up and the T/S parameters also change.
Qts and Qes and Bl (of course) increase with field power.
Going from 10W to 20W changes efficiency from 1.8% to 2.5%.
Jan
Last edited:
Attachments
Tmuikku has a point. At Fs, the cone mass and suspension cancel leaving nothing in the way of the motional impedance of the motor. However the voicecoil and former are a very stiff mass which sits between the voicecoil and cone. This mass is what reduces coupling to the main cone and modes. You could, perhaps, cancel the mass of this cone with some twisted evil transformation of negative impedance, but having a flat response and low distortion after that would be a bear of a problem to solve.
Honestly, DSP is more practical. The modes don't radiate equally in all directions but across the listening axis the lowest modes can be reduced by use of DSP. It easy to do this improperly though.
@keantoken - What I'm saying is that the cone doesn't look like a mass, just as you would say a piece of coaxial cable doesn't look like a capacitor at high frequencies.
I'm sure the voice-coil itself adds some effects at high frequencies, but this isn't anything to do with what happens at fs (which is most certainly not a "high frequency).
I'm sure the voice-coil itself adds some effects at high frequencies, but this isn't anything to do with what happens at fs (which is most certainly not a "high frequency).
Yes, I'm fully aware the cone is a diaphragm with it's own characteristic impedance and reflection times. The electrical analogue is an inductor as the cone mass, feeding a transmission line with a number of series and parallel resonances (dips and peaks). When the inductor impedance is significant compared to the characteristic impedance of the transmission line, no driving resistance (motor damping) will be able to critically damp the resonance modes.
EDIT: You can look into Hartley loudspeakers for one approach to dealing with cone resonances.
EDIT: You can look into Hartley loudspeakers for one approach to dealing with cone resonances.
Last edited:
@keantoken - that's not right, at high frequencies, the cone doesn't present a mass (which anyway would be reflected as capacitance in the electrical domain), but is mostly resistive. Transverse waves ripple outwards across its surface from its centre.
I said the cone is like a transmission line, which can be resistive to the extent that it is well damped. If it's not well damped, then it has a series of resonances according to it's reflection time. At what level of detail will you stop pointing out things that I left out because I'm not writing a book?
Nice. I trust it will be a good article.
Like doing it with a magnetic circuit employing a given permanent magnet. Now except the gap and the shape of the magnetic path, you have the extra tool of variable H field (Ampere, turns)
1.8% is already quite high
Interesting stuff, looking forward to your findings.
I would think with a well-designed magnetic circuit and the field coil driven with constant current and the gap saturated even at high VC currents the motor distortion, notably BL(i), should be excellent.
But even with a non-saturated gap the magnetic circuit might be stiffer than with a pre-polarized magnet.
To check whether the gap is saturated you could plot mid-band SPL vs. FC current at constant VC voltage. When SPL stops to rise BL his maxing out, indicating saturation.
yellow: without series resistance at 2.8V (measured at 50hz)
green: with 20R in series. Level adjusted according to SPL. No EQ - just volume.
orange: D5 without series resistance
pink: D5 with 20R
magenta: D4 without series resistor
cyan: D4 with 20R
light gray: D3 without series resistor
red: D3 with 20R
yellow: D2 without series resistance
light blue: D2 with 20R
dark red: D2 - in the current - without series resistance
red: D2 - in the current - with 20R
Best regards
Bernd
Last edited: