Cathode Bias can operate in manual Bias?? or it is just suited to auto bias??
I dont read this link yet, but it seems good stuff:
http://www.diyaudio.com/forums/tubes-valves/90671-self-bias-fixed-bias-big-tubes.html
Thanks for the link. I am looking for new commentary as I was told some existed. The tubes noted at the beginning of the link like the 845 and have such high bias voltages (greater than -100V) that developing that bias over a cathode resistor would produce a lot of heat.
Cathode bias is the technical (better) term for autobias, a term which is confusing as it implies something automatic going on. However in Europe it appears they commonly call cathode bias auto bias. For our discussion here I would rather not use that term. I suggest we use the proper terms:
1.
Fixed Bias for amps with bias pots that put a negative voltage on the grid and the cathode is very close to ground. In these we measure the "bias" as a small voltage drop across a resistor typically 1 ohm which gives 1 mV per mA. However saying our "bias is 50 mV" as people do is incorrect. The correct statement is that our idle current is 50 mA which we measure for convenience across a 1 ohm resistor. (other resistor values may be used but they are generally 10 ohms or less). The correct use of the term bias in a tube amp is to say " I have adjusted the negative grid voltage (the bias) on the tube so that the idle current is 50mA and for this tube in this circuit that bias voltage is -35 volts (typical for a EL-34). If we put a KT-88/6550 in that socket to achieve the same idle current we will have to increase the negative bias to -45 volts (typically) and if we don't the tube will draw excessive current and run red hot. Keep in mind that a given tube within the same type can easily be 10 volts more or less negative so our bias pot allows us to adjust for that also.
My RM-9 allows the use of a wide range of rather different tubes for no other reason than I have a large range on the bias pot that encompasses all. Why other manufacturers don't to this I have no idea. Although David Manley admitted that
he didn't want people putting any tubes other than the ones he wanted into
his amplifiers. Such was the topic of the Manley-Modjeski letters in Stereophile a long time ago.
Sorry for the long definition but its all in the language. If you start speaking this correctly it will likely enhance your knowledge of what is going on.
2.
Cathode Bias (sometimes called auto bias, especially in Europe where tubes are called valves) This method uses a large, often high power (10-20 watt) resistor in between the cathode and ground. If the tube is an EL-34 it is chosen to develop 35 volts across it at 50 mA. If the tube is a KT-88 the resistor has to be larger in value and size to develop 45 volts for 50 mA. This is why there are some amps that have a switch in the back to choose the other resistor when you choose these different tube types. It's all terribly simple and one does not then have to provide pots and jacks and a negative bias power supply. The amp runs a bit hotter for the extra watts (only 5 watts per tube) and takes care of itself. The way it does this is rather interesting. On the one hand you can simply say that the cathode resistor degenerates the DC gain of the tube. (There is almost always a capacitor bypassing this resistor to restore the AC gain.) By doing so as different tubes are substituted in the amplifier their differences will be degenerated (made smaller) to the point that they all run close enough to the same current (10% is good enough) to not degrade the performance of the amplifier. Matched tubes are still a good idea but the matching is less critical. The down side of cathode bias is that at full power the cathode voltage rises, reducing the current in the tube, reducing power and increasing distortion. However if we are playing the amplifier at low to moderate levels we do not reach this point. Note that all cathode bias amps are not equal. The Dyna ST-35 and many of that period used a single cathode resistor for all four output tubes. This is a horrible idea and requires very closely matched tubes. I am aware that some mods correct this design flaw that was caused by economy if nothing else.
3.
Servo or computer driven bias (for want of a better term). Here an additional solid state circuit that is not in the signal path will monitor the tube current and hold it steady. However this circuit has some time constant and can be confused by the music. There are likely some schemes where a computer of sorts measures and adjusts the bias when the signal is not present or very low.
Servo bias compares the cathode current to a reference value and holds the tube there by adjusting the negative voltage on the grid. Although simple to make, this circuit will make the grid more negative during loud passages and will need time to recover on soft passages. I don't like the idea of my bias moving around all the time. A servo is no better than cathode bias, though it does eliminate the cathode resistor and can have a longer time constant though not long enough to pass John Atkinsons (or any other) amplifier tests. How well these work is totally up to the designer but like any computer if something goes wrong you have no control over it. Personally I am not interested in either of these schemes.
There are the three. Any others?
Poll