The below headphone amplifier was constructed using my current amp's two channels combined into a balanced class A amp.
This configuration simulated indicates odd harmonics to be dominant and monotonic and even harmonics to be much lower but constant in character.
The bandwidth is way into the MHz region and phase remains almost entirely at zero being unconditionally stable. Harmonics shown are for 100kHz fundamental just for interest.
This is my first experience of a class A amp that sound crisp as opposed to warm.
This configuration simulated indicates odd harmonics to be dominant and monotonic and even harmonics to be much lower but constant in character.
The bandwidth is way into the MHz region and phase remains almost entirely at zero being unconditionally stable. Harmonics shown are for 100kHz fundamental just for interest.
This is my first experience of a class A amp that sound crisp as opposed to warm.
Attachments
Last edited:
Still although 33Ω are matched, transistors may not be, or only in a small range of temperatures, so I adhere to the idea of DC offset difficulties.
Thanks for your comments. The unit that I am fooling around with has decoupling caps and the offset is of no concern.
The attached was the schematic straight from MicroCAP10 schematic capture and models are from a reputable source. (there is a photo amp that I am trying the concept on).
I am specifically interested in the power density distribution of the odd order harmonic content compared to the even order. The flatness of both frequency and phase response is interesting.
Would you care to comment on why there is virtually no even harmonics, maybe you have some mathematical explanation that you may want to share in this regard.
The attached was the schematic straight from MicroCAP10 schematic capture and models are from a reputable source. (there is a photo amp that I am trying the concept on).
I am specifically interested in the power density distribution of the odd order harmonic content compared to the even order. The flatness of both frequency and phase response is interesting.
Would you care to comment on why there is virtually no even harmonics, maybe you have some mathematical explanation that you may want to share in this regard.
Attachments
fT= 160MHz typical as per Philips, the original manufacturer of this transistor.
BD140-16 datasheet(3/8 Pages) PHILIPS | PNP power transistors
I have changed the schematic to a balanced complementary symmetry topology and found that the distortion takes on a completely different characteristic. Anyway guys, this was just sharing an idea with you nothing noteworthy really. C18 should not be there. It was an afterthought only and the sim is without it.
Attachments
Last edited:
fT= 160MHz typical as per Philips, the original manufacturer of this transistor.
Very much at odds with this datasheet which says 3MHz (but at 0.5A, not 50mA): http://download.maritex.com.pl/pdfs/sc/BD140.pdf
Perhaps the 0.5A rating is saturated switching, not linear region? Or perhaps the 160MHz figure is common-base, not common emitter?
Last edited:
Thanks for sharing, Nico and XRK, and for the attribution for a very good but simple amp!
I think that whilst SE delivers prominent even order - lop-sided asymmetric on the waveform - a balanced format creates identical even order on each half, giving a top and trough asymmetrical distortion which is odd. For that reason, I prefer to hear this amp in SE rather than balanced.
Rather than cancelling even orders, a balanced situation (which requires a unipolar supply BTW) tends to duplicate the same distortion top and bottom so turns it into odd order.
The big advantage is always simplicity and low cost. I would think offset would not be an issue with good gain setting resistors, 2x33R, but this is important because many HP do not take kindly to DC.
Hugh
I think that whilst SE delivers prominent even order - lop-sided asymmetric on the waveform - a balanced format creates identical even order on each half, giving a top and trough asymmetrical distortion which is odd. For that reason, I prefer to hear this amp in SE rather than balanced.
Rather than cancelling even orders, a balanced situation (which requires a unipolar supply BTW) tends to duplicate the same distortion top and bottom so turns it into odd order.
The big advantage is always simplicity and low cost. I would think offset would not be an issue with good gain setting resistors, 2x33R, but this is important because many HP do not take kindly to DC.
Hugh
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.