Hi,
Check out http://users.auth.gr/~mixos/projects/nopcb/audio/005/
Any comments would be most welcome.
Cheers,
Vivek
Check out http://users.auth.gr/~mixos/projects/nopcb/audio/005/
Any comments would be most welcome.
Cheers,
Vivek
Did you copy it from there ?
http://space.tin.it/scienza/fladelle/
Whose design is it then?
Klaus
ps: just a personal opinion: I will never make any non-class-a amp with mosfets again... this sound always failed to satisfied me
http://space.tin.it/scienza/fladelle/
Whose design is it then?
Klaus
ps: just a personal opinion: I will never make any non-class-a amp with mosfets again... this sound always failed to satisfied me
Why does the design need emitter resistors for Q1 and Q2 Pierre?
I notice there is no miller cap across Q7 - this is often used to help stability and linearise Q7s Ccb.
This may be the reason why the designer has chosen a high frequency feedback path through R8 and C9 from the gate drive. As a rule I don't like multiple feedback paths, especially from the gate drive because the voltage signal here is highly distorted, especially with FET outputs.
The bias arrangement for the FETs is too crude for a single-ended VAS. The bias voltage will be asymmetrically modulated by the drive current, especially with C10 being a puny 1uF. This should be replaced with a BJT and a 50uF cap in parallel. As Pierre says you might get thermal runaway if the FETs don't have source resistors - depends upon the choice of FETs.
C3 and C5 should be 1000uF or more. The circuit will perform better if the LTP and VAS stages have regulated psus.
Otherwise I like it. It uses competent and independent CCS for both LTP and VAS outputs. It is about as simple as it can be. It uses a VAS output current of >20mA for good gate drive capability.
I notice there is no miller cap across Q7 - this is often used to help stability and linearise Q7s Ccb.
This may be the reason why the designer has chosen a high frequency feedback path through R8 and C9 from the gate drive. As a rule I don't like multiple feedback paths, especially from the gate drive because the voltage signal here is highly distorted, especially with FET outputs.
The bias arrangement for the FETs is too crude for a single-ended VAS. The bias voltage will be asymmetrically modulated by the drive current, especially with C10 being a puny 1uF. This should be replaced with a BJT and a 50uF cap in parallel. As Pierre says you might get thermal runaway if the FETs don't have source resistors - depends upon the choice of FETs.
C3 and C5 should be 1000uF or more. The circuit will perform better if the LTP and VAS stages have regulated psus.
Otherwise I like it. It uses competent and independent CCS for both LTP and VAS outputs. It is about as simple as it can be. It uses a VAS output current of >20mA for good gate drive capability.
25 watt mosfet amp
I recognize this little amp. One of my absolute favourites!
A very good quality NO NONSENSE design.
25 Watt is also a good choice. The need for more Watts is unusual.
Parts are cheap.
Why not build a biamp, active system
with a multiple number of this amplifier.
As traderbam points out, you can do modifications (mod),
if you like to do that.
The orignal is coming from "RED FREE CIRCUITS DESIGNS"
They have a lot of other useful and sound circuits.
Recommended.
RED FREE CIRCUITS DESIGNS - Audio
I recognize this little amp. One of my absolute favourites!
A very good quality NO NONSENSE design.
25 Watt is also a good choice. The need for more Watts is unusual.
Parts are cheap.
Why not build a biamp, active system
with a multiple number of this amplifier.
As traderbam points out, you can do modifications (mod),
if you like to do that.
The orignal is coming from "RED FREE CIRCUITS DESIGNS"
They have a lot of other useful and sound circuits.
Recommended.
RED FREE CIRCUITS DESIGNS - Audio
An externally hosted image should be here but it was not working when we last tested it.
technical data - not that bad
Technical data:
Output power: well in excess of 25Watt RMS @ 8 Ohm (1KHz sinewave)
Sensitivity: 200mV input for 25W output
Frequency response: 30Hz to 20KHz -1dB
Total harmonic distortion @ 1KHz: 0.1W 0.014% 1W 0.006% 10W 0.006% 20W 0.007% 25W 0.01%
Total harmonic distortion @10KHz: 0.1W 0.024% 1W 0.016% 10W 0.02% 20W 0.045% 25W 0.07%
Unconditionally stable on capacitive loads
---------------------------------------------------------------------
Here are some other Audio Circuits and Amplifiers
at the homepage Vivek first mentioned in this thread.
Audio Circuits with Amplifiers
Technical data:
Output power: well in excess of 25Watt RMS @ 8 Ohm (1KHz sinewave)
Sensitivity: 200mV input for 25W output
Frequency response: 30Hz to 20KHz -1dB
Total harmonic distortion @ 1KHz: 0.1W 0.014% 1W 0.006% 10W 0.006% 20W 0.007% 25W 0.01%
Total harmonic distortion @10KHz: 0.1W 0.024% 1W 0.016% 10W 0.02% 20W 0.045% 25W 0.07%
Unconditionally stable on capacitive loads
---------------------------------------------------------------------
Here are some other Audio Circuits and Amplifiers
at the homepage Vivek first mentioned in this thread.
Audio Circuits with Amplifiers Traderbam,
Resistors in emitters of each transistor in the differential pair are needed to avoid saturation of theese transistors by important transients at the input. This effect has been discovered in 1972 by Pr. Matti Ottala :
For stability, phase compensation is always implemented in this kind of design. Here is C9 and R8. This phase compensation causes a time-lag in signal propagation, and when a transient signal comes at the input, there is a very short time during which the overall feedback is inactive, causing a very inconvenient phenomenon called transient intermodulation distortion.
This phenomenon is totally unnoticeable by conventionnal distortion measurement in continuous sinewave, but addition of some resistors in the emitters of the input pair causes a very audible improvement in sound quality with musics as percussions, hapsichord, etc. (about 68 to 100 ohms).
Furthermore, another problem exist with C7. This electrolytical capacitor works with near zero volt dc across it. Under this conditions, electrolytical capacitors are unable to produce oxygen, wich is needed to maintain the integrity of the aluminium oxyde layer along the time. This causes distortion because of the asymetrical behaviour of the capacitor (which is not apparent when the capacitor is new, and the oxyde layer intact...) To avoid this effect, C7 must be made from 2 series connected capacitors, side reversed, and the common connexion must be biased at 3 to 5 vdc with a large resistor, say 100 kohms. But some distortion can remain at the very low frequencies (quite inaudible). Very high quality capacitors are recommended for C7.
Regards, Pierre Lacombe.
Resistors in emitters of each transistor in the differential pair are needed to avoid saturation of theese transistors by important transients at the input. This effect has been discovered in 1972 by Pr. Matti Ottala :
For stability, phase compensation is always implemented in this kind of design. Here is C9 and R8. This phase compensation causes a time-lag in signal propagation, and when a transient signal comes at the input, there is a very short time during which the overall feedback is inactive, causing a very inconvenient phenomenon called transient intermodulation distortion.
This phenomenon is totally unnoticeable by conventionnal distortion measurement in continuous sinewave, but addition of some resistors in the emitters of the input pair causes a very audible improvement in sound quality with musics as percussions, hapsichord, etc. (about 68 to 100 ohms).
Furthermore, another problem exist with C7. This electrolytical capacitor works with near zero volt dc across it. Under this conditions, electrolytical capacitors are unable to produce oxygen, wich is needed to maintain the integrity of the aluminium oxyde layer along the time. This causes distortion because of the asymetrical behaviour of the capacitor (which is not apparent when the capacitor is new, and the oxyde layer intact...) To avoid this effect, C7 must be made from 2 series connected capacitors, side reversed, and the common connexion must be biased at 3 to 5 vdc with a large resistor, say 100 kohms. But some distortion can remain at the very low frequencies (quite inaudible). Very high quality capacitors are recommended for C7.
Regards, Pierre Lacombe.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.