What I mean, in other words, is that if the loudspeakers have technical/construction issues the amplifier is not going to cure them and one should fix the loudspeakers first.
I am simply going to say, the answer is mostly no.
Class A/B push pull has the ability to deliver more power , assuming your preamp can drive it harder. If not, just going to push pull does not buy you anything.
45,
You said in Post # 20:
"The more worrying non-linear behaviour of loudspeakers gets worse with low Zout amps."
Did you mean:
The more worrying non-linear behaviour of loudspeakers gets worse with low [Damping Factor] amps.
(Just the opposite).
You said in Post # 20:
"The more worrying non-linear behaviour of loudspeakers gets worse with low Zout amps."
Did you mean:
The more worrying non-linear behaviour of loudspeakers gets worse with low [Damping Factor] amps.
(Just the opposite).
No I meant just that. The non-linear behaviour of suspensions in particular becomes worse with high DF amps. They only kill the overshoot but the impulse response is worse: dirtier and longer with a number of (unwanted) inversions. An unwanted inversion is when the signal is rising, for example, during its damped oscillating motion and suddenly (because of unwanted 3rd, 4 th order terms in the function, and thus the equation, that describes the motion) the signal changes sign before reaching the peak and re-changes again immediately. Basically an inverse peak superimposed to the main one. This makes the impulse last longer...and of course it's dirty. Often the best damping factor is the self-damping of the speaker itself with "no help" (current drive). This is often overlooked and instead people care more about the (limited) overshoot that especially for the woofer has little to no consequences as the room acoustics dominates at low frequency....
Last edited:
45,
Thanks!
You gave another good point to consider.
I have previously talked about another speaker response that sometimes relates to damping factor, and some crossovers.
Suppose for example you have a 12dB/octave crossover to a driver, you have a resonant circuit, RLC.
The amplifier damping factor is one part of the R.
The series LC (low pass), or series CL (high pass) is loaded by the driver at the junction of L&C, or C&L.
The amplifier drives the top and bottom of the LC or CL crossover.
The other part of the R, is the speaker driver DCR at the junction of LC or CL, (but the driver load also includes any driver reactance which is dependent on frequency).
For the complete circuit of amplifier, crossover, and driver . . .
The Q of the RLC is according to the amplifier's damping factor, L's Q, C's losses (Q), driver DCR, and driver reactance versus frequency.
What does an RLC circuit do, when a transient signal comes and goes that is at the RLC's or CLR's resonant frequency? . . .
It Rings!
The higher the total circuit Q, the longer the Ring!
But nobody seems to care, or give an acknowledgement.
Just my opinion.
Thanks!
You gave another good point to consider.
I have previously talked about another speaker response that sometimes relates to damping factor, and some crossovers.
Suppose for example you have a 12dB/octave crossover to a driver, you have a resonant circuit, RLC.
The amplifier damping factor is one part of the R.
The series LC (low pass), or series CL (high pass) is loaded by the driver at the junction of L&C, or C&L.
The amplifier drives the top and bottom of the LC or CL crossover.
The other part of the R, is the speaker driver DCR at the junction of LC or CL, (but the driver load also includes any driver reactance which is dependent on frequency).
For the complete circuit of amplifier, crossover, and driver . . .
The Q of the RLC is according to the amplifier's damping factor, L's Q, C's losses (Q), driver DCR, and driver reactance versus frequency.
What does an RLC circuit do, when a transient signal comes and goes that is at the RLC's or CLR's resonant frequency? . . .
It Rings!
The higher the total circuit Q, the longer the Ring!
But nobody seems to care, or give an acknowledgement.
Just my opinion.
Last edited:
> nobody seems to care
If the Q is about unity, as it usually is, the ringing will be very mellow.
If you use complimentary high pass and low pass filters, reasonably similar, the mellow rings will cancel. (Yes, there are designs which don't; these tend to other flaws like impedance dips.)
If the Q is about unity, as it usually is, the ringing will be very mellow.
If you use complimentary high pass and low pass filters, reasonably similar, the mellow rings will cancel. (Yes, there are designs which don't; these tend to other flaws like impedance dips.)
PRR,
Correct.
Most are low Q.
Yes, a poorly loaded high Q series RL to ground, has a very low impedance dip.
Resonance essentially "begins" at a Q of about 3.
Consider a loudspeaker with a tweeter that is much higher efficiency than the woofer.
Sometimes there is no crossover for the woofer, it is left to roll off naturally, due to the cone mass, and the voice coil inductance.
Sometimes for that system, the tweeter is crossed over with a single series capacitor and series resistor, OK as long as the tweeter does not also require damping, and as long as the tweeter crossover frequency is high enough for a network that is only 6dB/octave.
But if instead there is a 12dB/octave network there, and the resistor is after the CL network, that is a problem.
The CL might not be properly loaded by the series resistor and tweeter.
A two resistor attenuator, R to ground, R to tweeter, is needed to load the Cl network.
Most Zobel networks are properly designed, but a mistake there could also create ringing, which is what the network is supposed to prevent the driver from doing.
Correct.
Most are low Q.
Yes, a poorly loaded high Q series RL to ground, has a very low impedance dip.
Resonance essentially "begins" at a Q of about 3.
Consider a loudspeaker with a tweeter that is much higher efficiency than the woofer.
Sometimes there is no crossover for the woofer, it is left to roll off naturally, due to the cone mass, and the voice coil inductance.
Sometimes for that system, the tweeter is crossed over with a single series capacitor and series resistor, OK as long as the tweeter does not also require damping, and as long as the tweeter crossover frequency is high enough for a network that is only 6dB/octave.
But if instead there is a 12dB/octave network there, and the resistor is after the CL network, that is a problem.
The CL might not be properly loaded by the series resistor and tweeter.
A two resistor attenuator, R to ground, R to tweeter, is needed to load the Cl network.
Most Zobel networks are properly designed, but a mistake there could also create ringing, which is what the network is supposed to prevent the driver from doing.
Last edited:
Here is an example. I am switching 4 - position switch from -7 to +7 Ohm output resistance. The speakers are JBL 990, 3-way monitors with complex crossovers.
Wavebourn - Integrated Edelweiss-3 with damping control switch | Facebook
Wavebourn - Integrated Edelweiss-3 with damping control switch | Facebook
Summer, there is no universal solution. So you either do like Wavebourn or will have to accept that an amplifier with fixed configuration can never be good for all seasons. What I wanted to point out is that normally low Zout amps are described as the best to drive loudspeakers and the typical SE amp the worst. That might probably be true overall because most commercial loudspeakers are (in principle...) designed for low Zout amps but this is not a weakness of the SE amp. It's just the way it is. So when you have to make a choice like the SE vs PP in this thread you need to think and act globally, including the loudspeakers you want.
45,
Agreed.
Now you have landed on it . . .
"All Generalizations Have Exceptions"
There are many loudspeakers that can work well with both SE and Push Pull amplifiers.
Tubes:
And putting all push pull amplifiers into a single category is not good either.
And putting all single ended amplifiers into a single category is not good either.
Putting all solid state amplifiers into a single category is not good either.
Just my opinions.
Agreed.
Now you have landed on it . . .
"All Generalizations Have Exceptions"
There are many loudspeakers that can work well with both SE and Push Pull amplifiers.
Tubes:
And putting all push pull amplifiers into a single category is not good either.
And putting all single ended amplifiers into a single category is not good either.
Putting all solid state amplifiers into a single category is not good either.
Just my opinions.
The best generalization is, "Any generalization may be wrong" 🙂
In my video was shown a 2 pentode stage SE amp.
Wavebourn,
You have some nice video links to your amplifiers, and other subjects.
The one that showed the sound versus the switch setting, and JBL speakers was very interesting.
Someday, if you have the time, it might be nice to make a video to illustrate the same amplifier into other speakers, ones that have larger impedance excursions, and different mechanical/acoustic damping too. Then set the amplifier 4 position switch to the extremes as you did with the JBLs.
You have some nice video links to your amplifiers, and other subjects.
The one that showed the sound versus the switch setting, and JBL speakers was very interesting.
Someday, if you have the time, it might be nice to make a video to illustrate the same amplifier into other speakers, ones that have larger impedance excursions, and different mechanical/acoustic damping too. Then set the amplifier 4 position switch to the extremes as you did with the JBLs.