Many commercial speaker manufactures choose the low cost option for their designs.
They do not put Zobel, LR, or LC impedance networks in their loudspeakers.
Surprise!
They know that a typical solid state amp has a damping factor of 50 or more.
Why waste manufacturing parts and labor costs that will do nothing for the majority of their customers?
The same loudspeaker will sound different on a tube amplifier that has a damping factor of say from 2 to 4.
Of course, there are other causes of different sound from tube amps versus solid state amps.
When designing an amplifier, consider the loudspeaker you will use.
When designing a loudspeaker, consider the amp you will use.
I bet there is some discussion of that in the various Loudspeaker threads on this forum (or I hope so).
They do not put Zobel, LR, or LC impedance networks in their loudspeakers.
Surprise!
They know that a typical solid state amp has a damping factor of 50 or more.
Why waste manufacturing parts and labor costs that will do nothing for the majority of their customers?
The same loudspeaker will sound different on a tube amplifier that has a damping factor of say from 2 to 4.
Of course, there are other causes of different sound from tube amps versus solid state amps.
When designing an amplifier, consider the loudspeaker you will use.
When designing a loudspeaker, consider the amp you will use.
I bet there is some discussion of that in the various Loudspeaker threads on this forum (or I hope so).
Consider the loudspeaker
Yes, and a transconductance amplifier would be naturally most compatible with an aperiodic single-driver full-range speaker.
In my vocabulary aperiodic is synonymous with extremely damped bass resonance.
Yes, and a transconductance amplifier would be naturally most compatible with an aperiodic single-driver full-range speaker.
In my vocabulary aperiodic is synonymous with extremely damped bass resonance.
I'm not too sure what's the goal here. Anyway, I think that fig. c) in post #1 should be modified in a way that the current sensing resistor is in the load's return path, i.e. between the load and ground or between the OT's secondary and ground. This way the feedback signal is proportional to the output current exclusively. The circuit as shown is a mixture of current and voltage feedback, as the feedback signal contains the output voltage as well as the output current. With a strict separation of both you can adjust both separately to achieve whatever goal you want. Even zero or negative output impedance is possible with applying adjustable positive (!) current feedback.
Best regards!
Best regards!
I am wondering where this myth of high distortion of no-feedback pentode amplifiers is coming from. Compare, for example, 2A3 and 807 data sheets. PP AB1 2A3 puts out 15 W at 2.5% distortion. PP AB1 807 puts out 72 W at less than 2% distortion. Which one has less distortion?
No-feedback 807 at 800 V plate and regulated screen and bias supplies sounds very refined and immediate.
I am working on a sub system consisting of no- feedback PPP 807 driving a cluster of 24 12" drivers, each in a very small enclosure, to keep resonance out of working band. The prototype turned very promising. At this stage, it is more of architectural problem than anything else.
Single-ended pentode, no feedback: how much distortion?
EL95 12%
6V6 8%
6L6 11%
EL84 10%
EL34 10%
Single-ended each. I got those numbers by quickly glancing datasheets, so not very scientific approach.
EL95 12%
6V6 8%
6L6 11%
EL84 10%
EL34 10%
Single-ended each. I got those numbers by quickly glancing datasheets, so not very scientific approach.
Single-ended 45: 5%
Single-ended 2A3: 7%
Not that much of a difference with pentodes. Not earth and heaven.
What matters is not distortion at full rated power though, but distortion relative to power output at a given circuit topology. Since distortion decreases with power output in a roughly linear fashion, 6L6 beats 2A3 at 2.5 W.
Single-ended 2A3: 7%
Not that much of a difference with pentodes. Not earth and heaven.
What matters is not distortion at full rated power though, but distortion relative to power output at a given circuit topology. Since distortion decreases with power output in a roughly linear fashion, 6L6 beats 2A3 at 2.5 W.
Last edited:
I'm quite sure that the 807 amplifier distorts less at 15 W than the 2A3.
It appears that in the tube heydays the manufacturers did agree to relate a tube's maximum output power in SE to a distortion figure of 10 %. Remember that SE operation mainly was aiming at domestic radio and TV Service where 10 % THD might have been tolerable, high fidelity yet had to be invented then.
Best regards!
Utter nonsense. Measure distortion under real load and it's not just that. Nothing really works at all with zero fb pentodes, regardless if SE or PP. Un-listenable 99% of time. Pentodes without fbk simply cannot cope with most loudspeaker's load. Triodes can. Forget datasheets.
Pentodes with feedback, especially if it's cathode fb from output transformer, is another world. Difference is like night and day.....
Pentodes with feedback, especially if it's cathode fb from output transformer, is another world. Difference is like night and day.....
You probably never heard a good PP pentode amplifier without feedback. Yes, they won't work with boxed speaker's because of uncontrolled bass. At the time when no-feedback pentode was mainstream topology, they didn't use speaker enclosures, and drivers were low Qms, light weight cone, that didn't require electric damping.
High end consoles of the time (late 40s - early 50s) that cost a price of a car used no-feedback PP pentode topology.
BTW, disregarding data sheets is utter nonsense.
High end consoles of the time (late 40s - early 50s) that cost a price of a car used no-feedback PP pentode topology.
BTW, disregarding data sheets is utter nonsense.
Kay, it would be overly simplistic and condescending to regard pre-HiFi era as dark times when nobody really cared about quality of sound. Read publications of that era, and you will learn that that the quest for quality sound was always ongoing and evoked keen interest of enthusiasts and engineers alike. That was the time of WE amplifiers and field coil speakers, which are regarded as unsurpassed by connoisseurs of today.
IMO, I regard HiFi times as step back, not step forward. Boxed speakers and Williamson amplifiers took us to the sad (and still ongoing) era of bland-sounding gear, where specification is king, and how it sounds is not important.
IMO, I regard HiFi times as step back, not step forward. Boxed speakers and Williamson amplifiers took us to the sad (and still ongoing) era of bland-sounding gear, where specification is king, and how it sounds is not important.
No. Feedback by a voltage. It decreases output impedance. Current mode would be, a pentode output without any feedback.
I said that loudspeaker comes first. It's not just a matter of impedance but the real mechanical specs particularly regarding suspension (both linearity and magnitude of Cms). Then one needs to find a suitable power drive. Current mode means high Zout which you don't get with standard transformer coupling regardless of what you use. I mean hundreds of Kohm.....not 1 or 2K. The standard amp with a resistor in series is ok for experimenting but not really a solution as a standard power drive or voltage drive will be likely better performing in all aspects. All best current amps I have seen in the past were OTL, both tube and SS technology. However, from what I could see (they were commercial products) they were not standard SE or PP output stages. Zout in the 200K (valve) to 1 Meg (SS) range. Unfortunately that company is out of business.
An output transformer with a 5000 Ohm primary to an 8 Ohm secondary has an impedance ratio of 625:1.
If the primary is 40 Henry, then the secondary is 40/625 = 0.064 Henry (64 milli Henry).
Not exactly a high impedance current drive to the loudspeaker, unless there is some feedback in the amplifier to make it act like a current source.
And that feedback is exactly how it is done.
I believe that Wavebourn might know a lot about how to do that.
He does know how to use feedback to create negative resistance.
I vaguely remember using negative resistance on a project that I worked on in the mid 1980s at a major test and measurement equipment company.
If the primary is 40 Henry, then the secondary is 40/625 = 0.064 Henry (64 milli Henry).
Not exactly a high impedance current drive to the loudspeaker, unless there is some feedback in the amplifier to make it act like a current source.
And that feedback is exactly how it is done.
I believe that Wavebourn might know a lot about how to do that.
He does know how to use feedback to create negative resistance.
I vaguely remember using negative resistance on a project that I worked on in the mid 1980s at a major test and measurement equipment company.
Last edited: