I'm really not an expert on tube technology. I understand that BIAS is a way of restricting current flow. Tubes need to be biased to run cooler and to remain in the linear region.

1- When a tube is biased, is it completely turned off or you just partially reduce the current flow ?

2- if the tube is not completely turned off, how can you get rid of the noise when there is no signal at the grid ?

thanks in advance
pat

To get a grip on this, it may help to study some characteristic curves---typically, plate current vx. plate voltage for a series of different grid voltages.

Basically, the grid voltage controls the plate current in a non-linear fashion. The concept of "bias" has to do with setting the starting point for what will happen when an AC signal is applied to the grid.

It's apparent that you're thinking of audio power output tubes, although bias (properly, grid bias) applies to any and all amplifier tubes, voltage and power amplifier tubes alike.

Audio output stages may fall into three classes:

In a Class A amplifier, the tube is biased so that plate current always flows in the tube, even with the largest signal cycle to be applied to the control grid, which is never driven to cutoff. Single tubes or pairs of tubes in push-pull may be operated in Class A.

In a Class B amplifier, the tube is biased to cutoff and amplifies only half of a signal cycle. This requires a pair of tubes in push-pull for satisfactory (i.e., distortion-free) results. One tube amplifies the top half of the signal; the other tube amplifies the bottom half, and the two halves are "married" in the output transformer.

In a Class AB amplifier, the tube is biased somewhere between Class A and Class B. At low signal levels, it behaves like a Class A stage; at higher levels, it behaves like a Class B; thus, like Class B, a pair of push-pull tubes is required.

Class A is the most inefficient mode of output tube operation (although typically the most distortion-free). Bias is comparatively low, and the output tubes run hot as hell.

Class B is far more efficient, and capable of much greater output with the same tubes than Class A. With the tubes biased to cutoff, they run cool. On the other hand, a Class B stage is harder to design than a Class A stage, and far less common. It is touchier to handle, and much less forgiving of parts deterioration due to age; furthermore, it is totally unforgiving of abuse.

Again, Class AB falls between the two. Most tube-type stage amps and many high-fidelity amps and receivers are built to Class AB designs (particularly, class AB1, in which control grids are NEVER positive with respect to the cathodes).

I can't guess what sort of "noise" you're referring to. There are (too) many different kinds of noise, with more potential sources than kinds. Chances are, however, that the source has nothing to do with your output stage, and thus represents an entirely different question to be dealt with in a different context.

Hope this helps.

Larry

Use earplugs?

PM sent.

Larry

The purpose of biasing is not to make tubes run cooler, though they sure do get hot PDQ when the bias is incorrect!

If you consider a simple diode, the cathode emits electrons which are attracted to the plate if it is more positive than the cathode, and a flow of current will result. Should the plate be more negative than the cathode, it repels the electrons and no current flows. So how we bias the plate determines what the diode will do.

In an amplifier tube, you've got one or more grids in addition to the plate. All of the elements impact the flow of current from cathode to plate. Consequently, they all have to be biased correctly if the tube is to function as intended. Whether the tube is biased so the flow is completely cut off, going full blast (a condition known as 'saturation'), or somewhere inbetween, depends on the application.

The usual way the various conditions are denoted is by "class" of amplifier. The input is assumed to be a sinusoidial AC waveform, which you'll recall from geometry, can be derived from a circle. Like a circle, it starts out at 0 degrees and rotates until it reaches 360 degrees, after which it starts over again.

A Class-A amplifier is one which amplifies the full 360 degrees of the waveform. Practically this is done by biasing a tube so that when no signal is present, the plate current is halfway between cutoff and saturation. As long as the input voltage does not approach either limit, the output should be a reasonable facsimile of the input. Class-A amplifiers offer the lowest distortion, but because half the plate current flows even when no input is present, they are the least efficient. Best case efficiency is only 25% where capacitive coupling is used. Consequently, class-A circuits are used mostly for single-ended voltage amplifiers and where only modest amounts of power are to be handled.

Class-B amplifiers are also often referred to as "push-pull" circuits. At a minimum, two tubes are used. Each one is biased into cutoff, but just barely, so conduction occurs for 180 degrees of input cycle. The phase of the input is inverted for one of the tubes so the full 360 degrees of waveform is covered. Since the tubes only conduct when there is signal to amplify, efficiency is much better, 78.5% best case. So class-B circuits find wide application where there is moderate or high power to be handled. However, it is very difficult (with tube technology) to design or build circuits in such a way as to precisely start and stop conducting at exactly 0 and 180 degrees. As a consequence, class-B designs have higher distortion than class-A circuits.

Class-AB amplifiers are modified class-B circuits biased in such a way as to allow the tubes to conduct further than 0 and 180 degrees, but less than the full 360 degrees of input waveform. Efficiency is lower than straight class-B, but distortion is also significantly less. Hence these circuits find wide application in audio work.

Class-C amplifiers may be single ended or push-pull. In them, the tube(s) are biased deeply into cutoff, so they conduct for less than 180 degrees of input waveform. Efficiency is very high--as much as 90%--but distortion is also very high. Consequently, class-C amplifiers are typically found in transmitters, where the distortion in the RF output can be dealt with by tuned circuits after the amplifier.

An example of a situation where tubes are often biased into saturation is in oscillator circuits.

With regard to tube noise, this is a huge subject with lots of "gotchas." In any operating vacuum tube, a certain amount of random noise will be generated due to the effects of heat and electron motion on the elements inside the tube. The only way to eliminate or turn off that noise is to replace the tube with a transistor! Noise getting in from other sources, such as the components or wiring connected to the grid circuit of a tube, are separate issues.

Pat, I hope my earlier PM clarified things for you.

Let me know if I can be of any further help.

Larry

Most noise other than 120 HZ residual hum is due to poor design, restoration, or resistors past their prime. If you have already recapped, your best bang for the buck would be to do a wholesale replacement of all carbon resistors which couldn't cost much. Of course you should blast out all the controls, switches and tube sockets with deoxit.