With a 6N7, no, but there are some other fine class B designs around. I use a variant of Dave Berning's topology, first seen in the Audionics BA150, then later the EA230. The latter is a pretty easy amp to duplicate and it runs cool as can be (6JN6 outputs, if memory serves). The screen drive Class B method has penalties in requiring rather high drive voltages and currents, but the linearity is superb.
Thanks for the replies guys. Yes, Sy, I was thinking about the drive requirements for your average class AB2/B stage, and I was wondering: why couldn't you take two 6V6's (for example) and bias them down to say, -30V or so on the grid - until they're only drawing about 1mA each - and then feed them from a standard phase inverter? Your quiescent current would be only 2mA, and because the grids are always negative to a degree you would not need any grid "drive".
Is this a possibility, or a truly bad idea?
Joel
ps. Did you notice that the RCA tube manual calls the typical characteristics of the 6N7 "Class B Power amplifier", yet the tubes are never biased into cut-off in the example they give. They are shown drawing 17mA at zero volts Vg, and no-signal... what the? This is class AB2.
Is this a possibility, or a truly bad idea?
Joel
ps. Did you notice that the RCA tube manual calls the typical characteristics of the 6N7 "Class B Power amplifier", yet the tubes are never biased into cut-off in the example they give. They are shown drawing 17mA at zero volts Vg, and no-signal... what the?
This is class AB2.
Interesting. You mean that the DC plate voltage and bias voltage was fluctuating at idle?
I guess if this worked great we'd see TONS of amps out there running this way. But maybe not... because you'd still need a heck of a well regulated PSU in order to go from zero to 70 or 80mA of current demand in a millisecond, and not have the voltage crash hard.
I guess if this worked great we'd see TONS of amps out there running this way. But maybe not... because you'd still need a heck of a well regulated PSU in order to go from zero to 70 or 80mA of current demand in a millisecond, and not have the voltage crash hard.
The thing with tubes is they never completely cut off. At -1000Vg (as if) there'll still be a few nanoamperes. IIRC class B is classified as idle dissipation = 1/10th full-signal.
By -30Vg on a 6V6 you'd be in light class C, unless screen voltage is like 4 or 500V.
Correction, more like >300V.
Tim
By -30Vg on a 6V6 you'd be in light class C, unless screen voltage is like 4 or 500V.
Correction, more like >300V.
Tim
Ok. Well, what if they were biased slightly higher, and we used an ultrafast diode to clip the bottom of the waveform off so that the tube doesn't ever go down into that horribly non-linear region Sy is running into?
What is the fastest, distortion free solid state diode that conducts at millivolts?
What is the fastest, distortion free solid state diode that conducts at millivolts?
Class B usually doesn't get much respect, but the overturning of the audio myth that "chip-amps can't sound good" should have us rethinking the possibilities. Those little chip-amps are pretty much Class B.
The most aggregious issue with Class B is the distortion at the turn-over, how do you minimize the notch -- is it possible to eliminate?
And if it doesn't work out, i guess you could always dial in a bit more bias and you have the standard fix -- Class AB.
dave
The most aggregious issue with Class B is the distortion at the turn-over, how do you minimize the notch -- is it possible to eliminate?
And if it doesn't work out, i guess you could always dial in a bit more bias and you have the standard fix -- Class AB.
dave
Hi,
O.K. but let's not forget a few things here:
Those chips quite likely use fast-switching D-MOS power MOSFETs that you can even run in Class D for audio frequency.
When talking about classes of operation we have basically four different classes namely:
Class A, B, C and D.
If you think to "fix" the inherent problem of Class B for audio use by biassing the output devices for a given amount of Class A operation you effectively have entered a subclass of operation called Class AB.
Other subclasses exist but that's not the issue.
The point is simply: it's either Class B or it must be something else:
"Class B operation is the opposite of class A. Both output devices are never allowed to be on at the same time, or the bias is set so that current flow in a specific output device is zero when not stimulated with an input signal, i.e., the current in a specific output flows for one half cycle. Thus each output device is on for exactly one half of a complete sinusoidal signal cycle. Due to this operation, class B designs show high efficiency but poor linearity around the crossover region. This is due to the time it takes to turn one device off and the other device on, which translates into extreme crossover distortion."
Snippet taken from a glossary found on the web.
Back to class B, the only way I can think of using tubes in Class B and still have acceptable audio results is the use of sweep tubes, as SY pointed out, still the x-over distortion would be there and having an OPT won't help matters either.
Which leaves us with PP OTL amplifiers which in turn have their own set of problems due to the non-existence of complimentary tubes.
Solid state devices do have complimentary pairs which at least allows for a simpler solution to that problem.
Well, allow me to say that it's a very old problem inherent to this particular class of operation. One which led most audio designers to compromise on sound quality (Class A) and power efficiency (Class B) by using judicious amounts of both classes.
This is essentially class AB operation and it's deeper subclasses.
To this day no design using only class B as an operating Class has succesfully tackled this problem using tubes as the amplifying/output devices.
While there may well be a solution to this, I remain skeptical about its validity for audio use both from a sonic as well as an economic POV.
This is of course just my opinion on the matter, YMMV.
Cheers,
O.K. but let's not forget a few things here:
Those chips quite likely use fast-switching D-MOS power MOSFETs that you can even run in Class D for audio frequency.
When talking about classes of operation we have basically four different classes namely:
Class A, B, C and D.
If you think to "fix" the inherent problem of Class B for audio use by biassing the output devices for a given amount of Class A operation you effectively have entered a subclass of operation called Class AB.
Other subclasses exist but that's not the issue.
The point is simply: it's either Class B or it must be something else:
"Class B operation is the opposite of class A. Both output devices are never allowed to be on at the same time, or the bias is set so that current flow in a specific output device is zero when not stimulated with an input signal, i.e., the current in a specific output flows for one half cycle. Thus each output device is on for exactly one half of a complete sinusoidal signal cycle. Due to this operation, class B designs show high efficiency but poor linearity around the crossover region. This is due to the time it takes to turn one device off and the other device on, which translates into extreme crossover distortion."
Snippet taken from a glossary found on the web.
Back to class B, the only way I can think of using tubes in Class B and still have acceptable audio results is the use of sweep tubes, as SY pointed out, still the x-over distortion would be there and having an OPT won't help matters either.
Which leaves us with PP OTL amplifiers which in turn have their own set of problems due to the non-existence of complimentary tubes.
Solid state devices do have complimentary pairs which at least allows for a simpler solution to that problem.
Well, allow me to say that it's a very old problem inherent to this particular class of operation. One which led most audio designers to compromise on sound quality (Class A) and power efficiency (Class B) by using judicious amounts of both classes.
This is essentially class AB operation and it's deeper subclasses.
To this day no design using only class B as an operating Class has succesfully tackled this problem using tubes as the amplifying/output devices.
While there may well be a solution to this, I remain skeptical about its validity for audio use both from a sonic as well as an economic POV.
This is of course just my opinion on the matter, YMMV.
Cheers,
fdegrove said:
And i too, have the feeling Joel will end up with a Class AB amp after all is said and done.... but like when doing a good feedback amp, you build a really good no-feedback amp 1st, i think it is worth having a crack at getting Class B as good as possible before turning up the bias.
Joel is doing all the work... it won't hurt us to give him a hand. We can learn from his successes (& at least as much from the failures -- a good example of that over in the buffered gain clone thread
)dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.