I believe that push-pull was designed with the single purpose in mind: to get more efficiency from toobs and save electricity compromising sonic quality.
When a push-pull circuit is drawn correctly, the parallel between it and alternating current is obvious.
7N7
7N7
How about ECC85?
A friend of mine suggests that ECC85, when being applied as SE, will produce rich even harmonics, much richer than the odd ones. So if we need to compromise the excessive odd harmonics produced by differential pair while taking other advantages of the differential pair, the characteristics of ECC85 might be able to help.
BTW, why SRPP is good for driver but not ideal for pre-amp output? Push-pull?
Theoretically it has low distortion and low output impedance. Of course, the distribution of its odd and even harmonics are not even and will vary according to load impedance to have a theoretically optimal load under pre-defined working conditions. In other words, if the load impedance is constant and falls into the range of optimal load impedance. It is good. But if the load impedance is varying, it would be not longer ideal.
Is this your rationale?
A friend of mine suggests that ECC85, when being applied as SE, will produce rich even harmonics, much richer than the odd ones. So if we need to compromise the excessive odd harmonics produced by differential pair while taking other advantages of the differential pair, the characteristics of ECC85 might be able to help.
BTW, why SRPP is good for driver but not ideal for pre-amp output? Push-pull?
Theoretically it has low distortion and low output impedance. Of course, the distribution of its odd and even harmonics are not even and will vary according to load impedance to have a theoretically optimal load under pre-defined working conditions. In other words, if the load impedance is constant and falls into the range of optimal load impedance. It is good. But if the load impedance is varying, it would be not longer ideal.
Is this your rationale?
Class AB and B maybe. However, I believe class A push-pull to have many advantages over single ended, even better sonics in some applications. But to each his own, sonic quality is a matter of preference.
Agreed
Class B and AB are quite vulnerable to inductance leakage, particularly in case of triode PP, due to the instantaneous shut-off of one arm. Pentode with higher Ri is more suitable to Class AB application than triodes that usually have much smaller Ri.
Jeb-D. said:
Class B and AB are quite vulnerable to inductance leakage, particularly in case of triode PP, due to the instantaneous shut-off of one arm. Pentode with higher Ri is more suitable to Class AB application than triodes that usually have much smaller Ri.
Wavebourn said:
What nonsense!
I did two designs that use differential splitters. The one using a 6SL7, and the other cascoded 6BQ7s. In neither case, does this give anything like "Transistor Sound". (Then again, my solid state designs don't have that either. It's close, but not as good as hollow state.) That solid state, Big Box, amp sounds considerably worse.
If the differential doesn't sound right, don't blame the topology. Blame your just plain bad design, becuase that's where the fault lies.
Wavebourn said:
No, and perhaps yes.
And interest in hi-fi does after all have an external aesthetic element, hence fora such as this!
7N7
7N7 said:
Most audible errors happen when currents change directions and go through zero. Currents are only means that are needed to record and playback sounds.
As I have remarked elsewhere (and as is well known) we live in an imperfect world. Everything is a compromise.
You appear to be endorsing single-ended topology.
I cannot accept the inevitable shortcomings of the output transformer in such topology, and listening to numerous SE amplifiers has convinced me. They are easy to live with but I prefer to listen without "rose-coloured spectacles".
I cannot see the problem with Class A push-pull, as of course neither valve in the push-pull pair is ever switched off, and careful attention to operating points will ensure that this also applies to differential pairs in the input and driver stages too.
A SE OTL on the other hand could be interesting; I believe that Tubelab has been looking into this.
7N7
You appear to be endorsing single-ended topology.
I cannot accept the inevitable shortcomings of the output transformer in such topology, and listening to numerous SE amplifiers has convinced me. They are easy to live with but I prefer to listen without "rose-coloured spectacles".
I cannot see the problem with Class A push-pull, as of course neither valve in the push-pull pair is ever switched off, and careful attention to operating points will ensure that this also applies to differential pairs in the input and driver stages too.
A SE OTL on the other hand could be interesting; I believe that Tubelab has been looking into this.
7N7
7N7 said:
I appear to be endorsing what helps to achieve optimal results. At home I use hybrid amplifiers: class A single ended for mids+highs, class A+C push-pull for woofers and subwoofers, all of them have no output transformers. Some of my concert amps are push-pull with output tubes, and one model is hybrid A+C for PA in fields. I use different topologies for different optimums: I can save an energy on driving woofers and subwoofers, but I'm not willing to save it when driving mid-and high frequency line arrays; I can save battery energy in fields, but I won't save it on concerts in halls when all details of sounds of singers and acoustic instrument's players are desired.
Since I was taught to develop and manufacture military equipment where money is not a problem, we had extensive study and training of optimization skills, all designs had to be optimized for several criteria: mass, size, vibrations, pressures, temperatures, radiation, live creatures that consume dielectrics, electrical parameters, etc.... Now working with audio I use the same skills, always searching for optimal results regarding given criteria.
So, sometimes SE are preferable, sometimes PP are preferable, in terms of the main criteria that are sound quality and energy consumption. SE amps will always sound cleaner in terms of perception of clearness, when PP helps to save electricity compromising sound quality. It is an axiom for me.
Dear 7N7, one more question.
I find there are some tubes that are specially designed for the differential pair applications in early computer circuits, such as E90CC, E92CC, etc.
Are they suitable for audio applications like diffrerential phase splitter?
Thanks in antipication.
I find there are some tubes that are specially designed for the differential pair applications in early computer circuits, such as E90CC, E92CC, etc.
Are they suitable for audio applications like diffrerential phase splitter?
Thanks in antipication.
Re: Dear 7N7, one more question.
I have just at a look at the characteristics for these two. Linearity is not good with either of them, although it is true that the 2nd harmonics developed as a result of this would be cancelled if the diff pair was a good one, but the problem with this is that with valves like these (and 6J6) which have a common cathode, you have to test a load of them to find examples with matching anode characteristics, otherwise you would not have a proper diff pair.
7N7
Kenneth Zhu said:
I have just at a look at the characteristics for these two. Linearity is not good with either of them, although it is true that the 2nd harmonics developed as a result of this would be cancelled if the diff pair was a good one, but the problem with this is that with valves like these (and 6J6) which have a common cathode, you have to test a load of them to find examples with matching anode characteristics, otherwise you would not have a proper diff pair.
7N7
Re: Dear 7N7, one more question.
I've generally thought that many vacuum tubes created for early computer circuits were built with a different set of design criteria. Mostly, they were built to withstand being driven either to complete saturation or total cutoff, possibly for long periods of time. Linearity in between the extremes took a back seat.
As a result, many computer tubes make lousy audio tube.
Kenneth Zhu said:
I've generally thought that many vacuum tubes created for early computer circuits were built with a different set of design criteria. Mostly, they were built to withstand being driven either to complete saturation or total cutoff, possibly for long periods of time. Linearity in between the extremes took a back seat.
As a result, many computer tubes make lousy audio tube.
Re: Re: Dear 7N7, one more question.
Yep, like 6N6P, 6N30P...
They are bad, bad, bad!!!
Ty_Bower said:
Yep, like 6N6P, 6N30P...
They are bad, bad, bad!!!
After building a heap of different topologies and design over the years, well balance PP is the only way to fly for me. Most SE amps are too syrupy and sweet for want of a better expression.7N7 said:
Yeah. Got it.
The tubes devoted to computer circuit applications are designed to have good saturation and cut-off characteristics, just as their solid counterparts that are serving modern digital applications.
Linearity for them, is not a priority.
The tubes devoted to computer circuit applications are designed to have good saturation and cut-off characteristics, just as their solid counterparts that are serving modern digital applications.
Linearity for them, is not a priority.
- Status
- Not open for further replies.