What?!?
I know this paper very well, have been constantly referencing to it over the years as it deals with a flaw of current drive (actually, behavior of nonlinear mechanical systems with very low damping), not voltage drive. Please stop trolling me and stop distributing fake news.
When measuring with high source impedance, approaching constant current drive, then obviously -- as by sheer definition -- measured driver current distortion goes down. Acoustic distortion at LF may go up, though.
Am I then correct in understanding that current drive would produce different low end SPL performance with drive, i.e. 2.83 V would be different from 5.66V and so on?
That depends.. would you like to equalise the two and then talk about the remaining differences?
What's missing is a control test using a dummy resistor instead of a speaker.
Then we can check what part -- if any -- the amplifier plays. Many integrated class-D amps have a 'U' shaped distortion curve with minimum distortion at an excessively high output level. E.g. minimum HD @ 1-10W for a 100W amplifier, so that loud bangs and explosions will be much clearer and less distorted than voices in the <50mW range. (Priorities, eh? 😆 ) Therefore, the additional 20 ohm can force the amplifier to operate at a more favourable level.
That said, I don't understand the claim that velocity based EMF somehow compensates for a drop in Bl at high cone excursion. Something has to give. If the Q of the bass resonance 'flutters' in real time, can the height of this variation be measured?
Then we can check what part -- if any -- the amplifier plays. Many integrated class-D amps have a 'U' shaped distortion curve with minimum distortion at an excessively high output level. E.g. minimum HD @ 1-10W for a 100W amplifier, so that loud bangs and explosions will be much clearer and less distorted than voices in the <50mW range. (Priorities, eh? 😆 ) Therefore, the additional 20 ohm can force the amplifier to operate at a more favourable level.
That said, I don't understand the claim that velocity based EMF somehow compensates for a drop in Bl at high cone excursion. Something has to give. If the Q of the bass resonance 'flutters' in real time, can the height of this variation be measured?
Just curious to see how the SPL curves differ with different drive levels. But the more I think about it the more complex that gets, because with current drive V-ouput is no longer constant. Maybe 200 Hz or so would be a reasonable frequency for referencing V-output.
My end goal would be mapping BL with excursion without resorting to Klippel equipment.
Were comparative simulations with VC amplifier versus CC amplifier carried out for the paper?
Were comparative measurements (on the Viva) carried out with VC amplifier versus CC amplifier for the paper?
For small signal analysis, the response for either is the same at one level as another. Doesn't matter I or V.
Simply plot the (nonlinear) transfer of cone position vs instantaneous applied Voltage.
I published this control test in my posting #9
It is with 8.25R dummy resistor in place of the Transducer, with and without the series Resistor (0R/20R) and the correspondingly adjusted levels.
Sorry, i lost the the fundamentals.
It was certainly as smooth as a ruler, otherwise I would have saved it? Sorry...
That is easier said than done with AC. DC would be ideal but then you burn the VC very, very quickly with anything over a few Volts.
For a steady voltage, if the current goes up, the motional impedance that is dynamically generated must go down.
However, for a drop in back EMF to somehow compensate for a drop in Bl, presumably there must be an optimal Qtc that allows the impedance to drop by just the right amount.
The goal was to investigate the effect of the suspension nonlinearity under lowest possible damping. With an VC amp, too much damping is present to trigger the effect.
The authors have been very clear:
"As discussed in the previous section, in order to observe the jump resonance phenomenon the level of damping in the system has to be low. Given that a current amplifier eliminates the damping effects of BEMF, resulting in a lightly damped system"
And further, to note:
"The model was also used to investigate the effect of other system nonlinearities on the driver jump resonance behavior. Although the jump resonance occurs solely due to the nonlinearities in the stiffness of the driver’s surround and spider, the nonlinearities in the BL factor affect the shape of the frequency response curve as well as the specific values of the jump frequencies"
More often than not acoustic distortion around resonance goes up with current drive (of course the frequency response must be dialed in to be exactly the same as under voltage drive).
I also found this behavior, unexpectedly, with some AMT tweeters. They love current drive in general but at the very bottom end of the passband distortion was usually a bit higher with CC.
With enough care applying DC offset for low-level BL measurement is feasible, for short term tests anyway.
Both the DC and the test signal must be current source.
The AC component of the driver voltage then reflects the BL change vs cone position, not entirely linear though. Acoustic measurement is better.
Point is we would like to know how much BL decreases with larger excursion, say from 1 mm one-way to e.g. 5 mm one-way. This requires higher drive voltages.