The driving force that moves the diaphragm is current. Lorentz force: F = I*L*B. L and B are constant, so only the current remains. THD grows with the current and can be a few percent because the glabal feedback is missing. Specifically for this current +/- 0.25A, FFT sees the following:
In the MAKE article, the circuit delivers THD < 1% up to full scale.
In the MAKE article, the circuit delivers THD < 1% up to full scale.
IMHO: thought a higher 2nd harmonic vs 3rd harmonic is the wanted harmonic in tube amps?
The Lorentz force is independent from frequency. Of course a Speaker coil is not a fixed wire if we look at the lorentz force. Indeed a moving coil induces a back emf. Thats what increases the speaker impedance at the resonant freq. From low to high frequency we have a increasing impedance due to the inductivity. Inbetween is a relatively wide "linear" range. This is the reason why we have mostly 2 or 3 way speaker systems. A current amp would need very high output voltages to overcome the impedance peak at resonant frequency which would make no sense to correct this, because at the resonant frequency the speaker would become too loud and maybe destroyed.
Your circuit idea is of course nice. If you could design it to be more independent of involved BJT hFe, a bit more power and local feedback ...
Have fun, Toni
The Lorentz force is independent from frequency. Of course a Speaker coil is not a fixed wire if we look at the lorentz force. Indeed a moving coil induces a back emf. Thats what increases the speaker impedance at the resonant freq. From low to high frequency we have a increasing impedance due to the inductivity. Inbetween is a relatively wide "linear" range. This is the reason why we have mostly 2 or 3 way speaker systems. A current amp would need very high output voltages to overcome the impedance peak at resonant frequency which would make no sense to correct this, because at the resonant frequency the speaker would become too loud and maybe destroyed.
Your circuit idea is of course nice. If you could design it to be more independent of involved BJT hFe, a bit more power and local feedback ...
Have fun, Toni
Aus meiner Sicht ist die Sache mit 2. und 3. Harmonischen frei erfunden. Ich merke keine Änderung in der Klangqualität, ob ich 2. oder 3. Harmonische hervorhebe.
None of my speakers have broken yet. And that is actually impossible because the output current cannot be greater than the quiescent current.
Thanks
I didn't want to complicate the circuit so that it would be easy to solder. The adjustment is also easy. But above all, I wanted to showcase the true nature of the bipolar transistor. The prejudices of tube lovers are not at all fair. BJT can do everything better
https://www.heise.de/select/make/2023/3/2311110054476202800
Its complicated - I believe most power FETs do have a negative tempco, once you get to very high currents (10's of amps?), but at the sort of currents you want to bias at its positive. Laterals start their negative tempco at 100's of mA so are not subject to runaway in the same way.
The sound that a loudspeaker emits, is not in function of the current in all frequencies. The current provokes acceleration, but the speed of the coil generates voltage. If the amp is driving a horn loaded driver, the sound will be only in function of the voltage applied. With unloaded diaphragm speakers, around the resonance, the sound is in function of the speed- voltage, above it becomes in function of acceleration-current until the wavelength of the frequency reaches that of the diaphragm at which it becomes loaded and in function of speed-voltage. This is the case of piston moving diaphragm. To get intelligible voice, the diaphragm must break_up at 2fr. to decrease the moving mass, this makes the voice coil generate vibration in function of voltage-speed to propagate through the diaphragm.
The speaker manufacturers design with voltage driven to get the expected response and what amplifiers should respect.
Actually, this circuit is a final stage. For the operating voltage > 4.5V you would need a preamplifier anyway. Therefore the high input resistance would not be so important. But of course the circuit can be modified as you say. However, efficiency and linearity are reduced in this case because usually Vgs > Vbe.
The speed is the consequence of the acceleration: V = a*t. And the acceleration is caused by the current.
This happens because of the adaptation to voltage amplifiers and not because of the laws of electrodynamics.
So you think you are the first person who understood the laws of electrodynamics. All those designing amps and speakers since a century, are just carpenters.
The current doesn't generate acceleration but a force F=IBL. The force is opposed by the spring of the diaphragm and the mass which is not constant at all frequencies. The speed is not S=at with your basic knowledge it is the integral function of acceleration or else you get the impedance 1/wf.
The current doesn't generate acceleration but a force F=IBL. The force is opposed by the spring of the diaphragm and the mass which is not constant at all frequencies. The speed is not S=at with your basic knowledge it is the integral function of acceleration or else you get the impedance 1/wf.
When a bipolar power transistor is current controlled, its temperature coefficient is reduced. Let's consider the case when Q3 is MJ15025 and the circuit is supplied with +19V. The quiescent current is then 1.2A. According to the data sheet, MJ15025 has a gain of 70 at +25°C (see figure below, position 1). If its temperature rises to +100°C, the gain is already 120 (position 2). But this would correspond to the quiescent current of 1.2A * (120/70) = 2A (position 3). But at 2A the gain is only 90. Based on the quiescent current at +25°C, this results in 1.2A * (90/70) = 1.6A. The approximation steps can be continued. This way we end up somewhere close to 1.5A instead of 2A.
In simple terms, the magnetic force generated by a coil is proportionate to current and number of turns (Also called ampere turns). With speaker coil having number of turns fixed, the magnetic force generated is then proportionate to the current which is the only variable. This generated magnetic force either attracts or repels (in proportion to the current) with the permanent magnet. The only thing that changes here is the current. All other mechanical dynamics bla bla do not change. No need to be a NASA scientist to explain this. Regards.
