Double Blind tests *did* show amplifiers to sound different

[SIY]

One issue with auto bias, which was recognised even in the 1950s, is that a class AB1 output valve can be biased either for continuous sine wave or speech and music duties. Distortion will be very different under the two conditions. The reason is that with speech and music (assuming reasonable dynamic range) the average current drawn by the valve is low, close to its quiescent bias level. With continuous tone, the current draw is much higher, and so the bias voltage will rise due to the bias being derived by the current through the cathode resistor. An amplifier designed for sine-wave duty will have the wrong bias (too low) on speech and music, and one designed for speech and music will have too much bias on tones. That makes it difficult to measure distortion accurately, although manufacturers who understood this at the time, would create a fixed bias so the amp could be measured. The cathode bypass capacitor created a long(ish) time constant that smoothed out the bias variations on speech and music, but that wouldn't work for tone which by its nature is continuous.

High power valve amps typically had fixed bias using a separate supply that didn't have the issue with bias shifting with signal, but this added considerably to cost so wasn't used on most lower powered amps.

S.

Some of the later generations of autobias overcame that issue. Take a look at the designs from Morgan Jones and Menno Vanderveen. I’ve used Morgan’s circuit, and it does what it says on the tin.

Of course, one of the classics was the early digital autobias in the Audionics BA150 from the late 1970s.

 

[Vasr]

One issue with auto bias, which was recognised even in the 1950s, is that a class AB1 output valve can be biased either for continuous sine wave or speech and music duties.

Did you mean to say one issue with fixed bias? As the rest of the write-up seems to make a case for auto bias that wouldn't care if it is pure sine wave or music. So that would seem like the motivation for auto bias.

I know nothing technically to take a position on this, just trying to understand.

 

[solderdude]

Technics made an effort by offering amplifiers with 'computer class-A'
More of a marketting gimmick though but the signal and temeperature were measured and bias current was adjusted.
The class-A part of course wasn't class-A but new Class-A.

 

[sergeauckland]

Did you mean to say one issue with fixed bias? As the rest of the write-up seems to make a case for auto bias that wouldn't care if it is pure sine wave or music. So that would seem like the motivation for auto bias.

I know nothing technically to take a position on this, just trying to understand.

Unless we have a problem with terminology, I understand auto-bias to mean where the bias voltage is generated by the current flowing through a cathode resistor, i.e. it happens automatically. It's also been called self-bias.

Fixed bias is where there is a separate bias supply, and no cathode resistor or if there is one, of very low value, just sufficient to allow a measurement to be made of the cathode current.

S.

 

[Vasr]

Unless we have a problem with terminology, I understand auto-bias to mean where the bias voltage is generated by the current flowing through a cathode resistor, i.e. it happens automatically. It's also been called self-bias.

Fixed bias is where there is a separate bias supply, and no cathode resistor or if there is one, of very low value, just sufficient to allow a measurement to be made of the cathode current.

S.

May be I understood the terminology wrong. Rather than a specific implementation methodology, I understand auto-bias as not being set to a fixed bias value as is typically done with a bias setting on a Class AB amp but that it would vary based on the situation. So I would have thought auto-bias would be more immune to the type of signal going through than a fixed bias setting than it be an issue for auto-bias but not one for fixed bias. I should go and read up.

 

[sergeauckland]

May be I understood the terminology wrong. Rather than a specific implementation methodology, I understand auto-bias as not being set to a fixed bias value as is typically done with a bias setting on a Class AB amp but that it would vary based on the situation. So I would have thought auto-bias would be more immune to the type of signal going through than a fixed bias setting than it be an issue for auto-bias but not one for fixed bias. I should go and read up.

Auto bias / self-bias depends on the valve drawing current through a cathode resistor. This current creates a voltage and so lifts the cathode voltage from ground. As the grid is nominally at ground potential and the cathode is positive, that sets up negative bias for the valve, as the grid is at a lower voltage than the cathode. As this bias voltage depends on the amount of current flowing through the valve, it varies with the signal going through the valve, which is one reason for the cathode bypass capacitor, to smooth out variations in the bias voltage due to audio. (the other reason for bypassing the cathode resistor is to maintain gain.)

With fixed bias, the cathode is grounded (or very close to it) and the grid is held negative by a separate supply that is then immune to variations caused by the signal. Apart from the cost of the separate supply, fixed bias has the risk of serious damage to the output valves and output transformer if the bias supply fails, which is why sensible fixed bias designs have some sort of safety device that cuts off the supply if the bias fails.
Auto-bias / Self-bias doesn't have a supply to fail and so is safer, the chances of the cathode resistor going short-circuit is very small, resistors tend to fail open circuit.

S.

 

[SIY]

Unless we have a problem with terminology, I understand auto-bias to mean where the bias voltage is generated by the current flowing through a cathode resistor, i.e. it happens automatically. It's also been called self-bias.

Fixed bias is where there is a separate bias supply, and no cathode resistor or if there is one, of very low value, just sufficient to allow a measurement to be made of the cathode current.

S.

That’s one form of automatic bias. But there are more modern versions which don’t have the disadvantages of poor efficiency and susceptibility to blocking distortion.

 

[FeddyLost]

Today i thought that if difference somehow present even at youtube a/b comparisons, it's a good idea to set up good recorder and change amps/sources/cables and then analyze resulting waveforms, find correlation, compare them a/b/source and do everything as usually, but without listening.
Even the best portable recorder is much cheaper than power analyzer for amp output.

 

[sergeauckland]

That’s one form of automatic bias. But there are more modern versions which don’t have the disadvantages of poor efficiency and susceptibility to blocking distortion.

I'm afraid my knowledge of valves stopped in 1970, it was SS after that, with one brief foray in the early 1980s for a bit if fun, but strictly with classic techniques.

S

 

[SIY]

I'm afraid my knowledge of valves stopped in 1970, it was SS after that, with one brief foray in the early 1980s for a bit if fun, but strictly with classic techniques.

S

I have an irrational love of tubes. There, I confessed!

 

[MakeMineVinyl]

One issue with auto bias, which was recognised even in the 1950s, is that a class AB1 output valve can be biased either for continuous sine wave or speech and music duties. Distortion will be very different under the two conditions. The reason is that with speech and music (assuming reasonable dynamic range) the average current drawn by the valve is low, close to its quiescent bias level. With continuous tone, the current draw is much higher, and so the bias voltage will rise due to the bias being derived by the current through the cathode resistor. An amplifier designed for sine-wave duty will have the wrong bias (too low) on speech and music, and one designed for speech and music will have too much bias on tones. That makes it difficult to measure distortion accurately, although manufacturers who understood this at the time, would create a fixed bias so the amp could be measured. The cathode bypass capacitor created a long(ish) time constant that smoothed out the bias variations on speech and music, but that wouldn't work for tone which by its nature is continuous.

High power valve amps typically had fixed bias using a separate supply that didn't have the issue with bias shifting with signal, but this added considerably to cost so wasn't used on most lower powered amps.

S.

This implementation is on a solid-state amplifier.

 

[MakeMineVinyl]

I have an irrational love of tubes. There, I confessed!

Although I work around solid-state amplifiers all day my home system uses tubes, with the exception of the subwoofer amplifier.

 

[SIY]

Although I work around solid-state amplifiers all day my home system uses tubes, with the exception of the subwoofer amplifier.

My first published article in AudioXpress was a tube preamp. My tube stuff has, unfortunately, not been turned on since we moved to Arizona, due to the heat load.

One day, after we’re done here and move somewhere with a more amenable climate, I’ll get back to my Super Power amp...

 

[DonH56]

I shut all my tube gear down years ago and sold some of it. I played with them much more in college and shortly after, before Life and Work took up all available time. I sold a few pieces, basically hobbyist items. I got jazzed and built a tube preamp with differential circuits all the way through, thinking it would offer much lower distortion and noise. It had fancy bias circuits, nicely stabilized feedback to linearize the output, etc. It was quieter, cleaner, and measured much better than the ARC and CJ tube preamps we had in the shop for me to compare. Sold one; too many folk said it "sounded too solid-state". At one point I built a little mixer to add back even-order HD at a fairly high level and they thought it then sounded like a tube preamp. I gave up.

I do miss those big output tubes in my old Ham rig, and of course helping swap those big tubes in 10~100 kW RF transmitters was fun.

 

[Sal1950]

I shut all my tube gear down years ago and sold some of it.

I had to sell my big VTL monoblock amps when I left Chicago for Florida, heat just would have been too big a problem. Plus I had to sell all my old classic tube gear collection, just no room for it here. All I have left is a working 1925 Atwater Kent Model 20 radio, but I haven't fired it up in years, and a 1939 Zenith "Racetrack" radio, also working. I do fire that one up now and then to keep the caps formed and think back to the family who probably sat around it all through WW II listening to the Roosevelt "Fire Side Chats" and all the rest, a real bit of history.
I love tubes, brings me back to my mis-spent youth when I built a linear power amp for my CB using 4 6146s. Had to wait till late at night to use it, it would bleed thru all the neighbors TV and really tick them off, then they'd complain to my dad and I'd get a good wack. LOL

 

[restorer-john]

All I have left is a working 1925 Atwater Kent Model 20 radio, but I haven't fired it up in years, and a 1939 Zenith "Racetrack" radio, also working.

Sal, they are beautiful. The Atwater is just gorgeous. Don't ever get rid of them.

 

[cistercian]

I have an irrational love of tubes. There, I confessed!

My favorite small treasure as a boy was a copy of the pamphlet "The care and feeding of power tetrodes"
The 4-125 was seen on the cover. Eitel Mccullough. (Eimac)
Published in 1951. I loved the 1955 radio handbook that was part of the home library too. Dad saved his
Navy ET coursebooks and I inhaled them too. It was all tubes, all the time. All of my early electronic projects were
tubes scavenged from old TV's and radios. I still have a decent selection of TV power transformers!
I was sad when the best US plants that made them closed.

 

[SIY]

My favorite small treasure as a boy was a copy of the pamphlet "The care and feeding of power tetrodes"
The 4-125 was seen on the cover. Eitel Mccullough. (Eimac)
Published in 1951. I loved the 1955 radio handbook that was part of the home library too. Dad saved his
Navy ET coursebooks and I inhaled them too. It was all tubes, all the time. All of my early electronic projects were
tubes scavenged from old TV's and radios. I still have a decent selection of TV power transformers!
I was sad when the best US plants that made them closed.

I think I still have the Navy course book around here somewhere. Large format paperback with a blue cover... But my most treasured classic text is Landee, Davis, and Albrecht's "Electronic Designers Handbook." Amazingly comprehensive.

 

[cistercian]

I think I still have the Navy course book around here somewhere. Large format paperback with a blue cover... But my most treasured classic text is Landee, Davis, and Albrecht's "Electronic Designers Handbook." Amazingly comprehensive.

I don't have that one. I do have a copy of "the origin of Radar" written by the man who invented the T/R switch for the early WW2 Radars.
It is an epic text with pictures of tube ring oscillators being pulsed at 15kv for high peak power and to raise the frequency limit
by reducing transit time of the electrons in the tube. Absolutely wild edge of the tech at the time. They used Eimac tubes!
It is a little yellow book and was given to me by the HS librarian. I checked it out until the card was full and was the only person who had ever
checked it out! So it was gifted to me. I have to dig it out and read it again! Author Robert Morris Page .

 

[restorer-john]

It is a little yellow book and was given to me by the HS librarian. I checked it out until the card was full and was the only person who had ever
checked it out! So it was gifted to me.

My school librarian did the same with Towers International Transistor Selector! Nobody had ever borrowed it before and I kept borrowing it. He also gave me about 5 years worth of Electronics Australia magazines. I still have them complete with the school stamp and binder ring holes punched through all the pages. I had to employ an early version of mental error correction on schematics where the holes (3 per page near the spine) took out component values.