When we bias a tube amp what are we measuring?

[RogerMod]

There seems to be some confusion about biasing a tube amplifer. I would like to ask the readers here to give their thought on what are we adjusting and what are we measuring when we adjust a tube amplifer's bias.

If one were discussing his amplifier with another person what would be the proper language to use?

Why do we care about bias so much? How does it affect sound? How does it affect tube life?

When you measure bias do you put your meter on a mV scale or a mA scale?

Even it you find this too much to answer at least give it some thought.... or just do what the manufacturer says to do.

 

[thorpej]

Bias refers to the difference in voltage between the control grid and the cathode, or Vgk, at quiescent conditions. What this ultimately controls is the amount of quiescent current that flows through the tube, and where along the load line you center the operating point.

When someone refers to biasing "hot", it means "more current" (and a less-negative Vgk). When biasing "cold", it means "less current" (and a more-negative Vgk).

The term bias point is used loosely ... if someone is referring to "biased the tubes at 50mA", that's a measurement of the quiescent current. If they say "biased at 2.6V", well, maybe that means they adjusted a bias control until they read 2.6V at the test point... if they say "biased at -35V", maybe they mean they set the negative voltage on the control grids to -35V; context is everything.

We care about bias because it relates directly to the performance of the amplifier as well as tube life. Bias too cold, and your amp will sound like s**t. Bias too hot, and your tubes won't last (and your amp could sound like s**t.)

 

[kward]

Biasing matters greatly because it's the center point around which the output signal swings. Get the biasing wrong (too hot or too cold) and the output signal may not be able to swing to its full potential as designed by the manufacturer, or it may cause significantly more distortion than expected by the manufacturer. So it's really important to get the biasing set right so the best performance of the amp can be obtained.

The technically correct definition of a bias point for a triode a set of three values that define the tube's quiescent (zero signal) operating point, and for a pentode, it's a set of four values that define the tube's quiescent operating point.

For a triode, the bias point is defined as:
a) The current at the plate
b) The voltage drop across the tube from plate to cathode
c) The voltage differential between cathode and control grid

For a pentode, the bias point is defined as the above three, plus
d) the voltage differential between the screen grid and the cathode

All three (triode) or four (pentode) must be simultaneously stated to completely specify the "bias point" or "quiescent point" for the tube in question.

For the output stage of a power amp, amp manufactures will sometimes provide a user adjustable mechanism to set the output stage bias by allowing the voltage difference between cathode and control grid to be adjusted within a certain predetermined range, which once set, will automatically set the other values.

The terms "cathode bias" and "fixed bias" are also used routinely to describe the method by which the voltage differential between cathode is control grid is provided. Regardless of which method is used, the above three (triode) or four (pentode) parameters must always be set for the tube to be biased. "Fixed bias" means the voltage is applied to the grid that is more negative than the cathode voltage, and sometimes this voltage is made adjustable, so you might hear the term "adjustable fixed bias." Whereas "cathode bias" means the cathode voltage is made more positive than the control grid voltage by placing a resistor in the cathode circuit that when current flows through it, it creates a voltage potential at the cathode that is more positive than the voltage at the control grid. In both cases, the thing being manipulated is the voltage differential between cathode and control grid.

 

[sKiZo]

Especially important in a push/pull amp design. The more closely matched and balanced electronically the power tubes in each pair are, the better the amp will sound.

PS - the limiting factors on how high you can bias over the mfg's recommendation are tube life and how strong the iron is. Push them higher than what the transformer is rated for, and you're asking for trouble. That said, I run some tubes "hot" with better results. The force is strong in my iron, and I'm running KT120's in an amp originally designed for KT88's. I'll never develop the full potential of the tubes, but I like the way they sound, and they seem more comfortable and robust at the higher bias. Can't really say if there's an impact on tube life, as even then, the tubes aren't anywhere near their limits.

 

[sKiZo]

I'm not confused ...

(or am I??) <G>

 

[RogerMod]

It sets the high and low peak grid signal voltage values for clipping and/or cutoff, at a given plate/screen voltage.

Actually bias does not determine those points unless the driver is very wimpy and not up to its job.

 

[RogerMod]

Biasing matters greatly because it's the center point around which the output signal swings. Get the biasing wrong (too hot or too cold) and the output signal may not be able to swing to its full potential as designed by the manufacturer, or it may cause significantly more distortion than expected by the manufacturer. So it's really important to get the biasing set right so the best performance of the amp can be obtained.
.

In a single ended amplifier I will agree with you. In a push pull amplifier this is not true. Lets say we take any push pull AB amplifier and bias it rather cold (more negative grid) or bias it hot all the way to class A. As long as the power supply holds up there will be no appreciable difference in power. I know many people think bias affects output power but it does not. This can be easily observed on the bench. It does affect distortion of course.

 

[RogerMod]

Especially important in a push/pull amp design. The more closely matched and balanced electronically the power tubes in each pair are, the better the amp will sound.

PS - the limiting factors on how high you can bias over the mfg's recommendation are tube life and how strong the iron is. Push them higher than what the transformer is rated for, and you're asking for trouble. That said, I run some tubes "hot" with better results. The force is strong in my iron, and I'm running KT120's in an amp originally designed for KT88's. I'll never develop the full potential of the tubes, but I like the way they sound, and they seem more comfortable and robust at the higher bias. Can't really say if there's an impact on tube life, as even then, the tubes aren't anywhere near their limits.

How strong which iron is?

 

[RogerMod]

Heeeheeee... See page two of "How and when to tube roll," this thread is a spinoff. You are now a specimen.

Oh, Im here to see what another group thinks about bias. What people dont know about bias in push pull amplifiers is "we make it up" there is no precise formula or black magic. Its a simple trade off between distortion and tube life with a little bit of damping thrown in.

 

[sKiZo]

How strong which iron is?

The power transformer. That has to be able to supply enough juice to keep the power tubes happy, as well as the drivers and storage, while still leaving a comfortable reserve. Many amps are built to a price point, and don't have a lot of reserve available, and even a small bump up in bias can make surprisingly large demands that the stock transformer just can't handle.

The old Dynacos are good examples. The stock iron was just enough to power up a quad of EL34's - anything else and you take your chances, unless you've upgraded with newer more robust components. Notice the height of the newer iron ...

 

[primosounds]

Oh, Im here to see what another group thinks about bias. What people dont know about bias in push pull amplifiers is "we make it up" there is no precise formula or black magic. Its a simple trade off between distortion and tube life with a little bit of damping thrown in.

RogerMod, are you like the "sensei" who sits quietly while the students scramble to answer the question of the day? And see if they are worthy of attaining further enlightenment? I think i will keep my thoughts to myself and let "Sifu" explain further.

 

[RogerMod]

Granted the stock ST-70 power transformer runs hot. My dear friend Bruce DePlama worked for David Hafler and admitted David was tough on the suppliers. He would argue over a few pennies in a transformer for the preamps and got a bit in trouble there. When you consider he offered the kit for $99 its amazing even in that day. Thats about $650 in todays money and one would be hard pressed to do it for that today.

However this introduction has nothing to do with bias. I do not agree that "even a small bump up in bias can make surprisingly large demands that the stock transformer just can't handle". Do you mean that if the idle current gets too high the transformer will get too hot? If it does, just turn it down man. Even though the transformer runs hot the survival rate is high even to this day.

The bias current is small compared to the full power current that the power supply must deliver. If it can deliver that current it can certainly handle small changes in bias. For example the RM-9 has very large iron, EI-150 x 2.75 inch stack for power and outputs. The idle current runs 120 mA per channel and the full load current at 100 watts out is 500 mA. So I care not if the bias moves around a bit when the power supply can do that. Although I find some amplifiers have somewhat unstable bias it is generally not the fault of the iron but the circuitry. In the RM-9 the bias supply compensates for line variations so that the idle current does not rise with line voltage.

There is not problem running any power tube in there with a 1.5 amp heater as long as the idle current is properly adjusted. That includes the KT88 family which has the same heater demand as the EL34. BTW I believe I see a 5U4 in the left hand picture which does stress the transformer over the proper 5AR4.

And unfortuantely more bias means less current, which is really all I came here to say. I never though it would cause such fervor.

 

[RogerMod]

RogerMod, are you like the "sensei" who sits quietly while the students scramble to answer the question of the day? And see if they are worthy of attaining further enlightenment? I think i will keep my thoughts to myself and let "Sifu" explain further.

Teaching audio, or perhaps anything, requires the student to have a open mind. Do you know this Zen story?

A Cup of Tea
Once, a long time ago, there was a wise Zen master. People from far and near would seek his counsel and ask for his wisdom. Many would come and ask him to teach them, enlighten them in the way of Zen. He seldom turned any away.One day an important man, a man used to command and obedience came to visit the master. “I have come today to ask you to teach me about Zen. Open my mind to enlightenment.” The tone of the important man’s voice was one used to getting his own way.

The Zen master smiled and said that they should discuss the matter over a cup of tea. When the tea was served the master poured his visitor a cup. He poured and he poured and the tea rose to the rim and began to spill over the table and finally onto the robes of the wealthy man. Finally the visitor shouted, “Enough. You are spilling the tea all over. Can’t you see the cup is full?”

The master stopped pouring and smiled at his guest. “You are like this tea cup, so full that nothing more can be added. Come back to me when the cup is empty. Come back to me with an empty mind.”

*To read longer version of this tea story, another version I encourage you to check out is on Osho.com

 

[primosounds]

Teaching audio, or perhaps anything, requires the student to have a open mind. Do you know this Zen story?

Thank you for sharing, and that is a very nice story that we in an audio forum, should always be aware of.
One of my College teachers was fond of saying , "not all wisdom comes from the same house". Another way of espousing a similar view. Cheers.

 

[kward]

Roger, I invite you to reread my post without making assumptions. You will note I have said nothing nor implied anything about output power as it relates to biasing.

 

[RogerMod]

Roger, I invite you to reread my post without making assumptions. You will note I have said nothing nor implied anything about output power as it relates to biasing.

If not output power, what does this mean? "Biasing matters greatly because it's the center point around which the output signal swings. Get the biasing wrong (too hot or too cold) and the output signal may not be able to swing to its full potential "

Im not trying to make anyone wrong here, just trying to correct a few misconceptions. I have spoken to many owners who think bias affects power. It all too prevalent. Should I let people continue to think that or help them get the proper information?

 

[triode17]

Hello Roger. What we are setting is the Q Point on the tube's Load Line. If the Idle Current is too low, then the tube will run closer to cut off than it should, causing the output signal to clip earlier on negative peaks ( I believe). If the Idle Current is too high (what I call overbiased), then the tube will run closer to saturation, causing the positive output peaks to clip earlier.So then yes, the idle current does affect power.
Then, it ultimately comes down to setting the Idle Power Dissipation between longevity and low distortion. I follow the 70% Rule.... I think you know the rest.

 

[kward]

"swinging to its full porential" I meant to apply to both A and AB1. In an AB1 amp there is a portion where both tubes are conducting, so changing the bias affects the output swing allowed where both tubes are conducting. I hope that helps clarify.

 

[nj pheonix]

The power transformer. That has to be able to supply enough juice to keep the power tubes happy, as well as the drivers and storage, while still leaving a comfortable reserve. Many amps are built to a price point, and don't have a lot of reserve available, and even a small bump up in bias can make surprisingly large demands that the stock transformer just can't handle.

The old Dynacos are good examples. The stock iron was just enough to power up a quad of EL34's - anything else and you take your chances, unless you've upgraded with newer more robust components. Notice the height of the newer iron ...

I understand your point but the case you're making is somewhat skewed in that you're showing a picture of a stock ST70 next to a new VTA ST120.
Apples to oranges. The single GZ34 is a greater limiting factor than the PT in that design.

No inference on the bias topic.
Sorry about the sidetrack

 

[dcgillespie]

Maybe it would be better to address the power output issue by defining it as "undistorted power output", with the term undistorted defined by the absence of any visible clipping or crossover distortion.

Dave