Fixed Bias: the ups and downs

[alexmoose]

Hey, I was looking into making a self bias using a CCS for an EL34 amp, but I just found a 24v transformer that I ripped out of an iced tea machine (the fruits of dumpster diving), and I noticed that after being rectified, its just around the right voltage for a fixed bias on an EL34.

I wanted to know what the advantages, and dis-advantages of a fixed bias? or will this iced-tea machine transfomer (SEC: 24v, 2 amps) not work?

[arnoldc]

I'm going to tag along in this topic if alex don't mind.

Fixed bias on SET- what are the advantages and disadvantages?

[Yvesm]

At first, no more #$£%§* cathode bypass cap

The better bypass is a short straight piece of wire

Yves.

[arnoldc]

Hi Yves, yes no more BlackGate or whatever flavor of the month for the bypass, lower B+ too, but what about the sound?

[analog_sa]

Good point ArnoldC. The sound is different. With some tubes in some setups i prefer fixed bias with its better bass and dynamics while in others it becomes too harsh.

With enough effort and tweaking both biasing schemes can be made to sound great.

And btw the voltage for fixed bias needs to be very, very clean - you can hear every component of the biasing network.

[Yvesm]

Agreed.

I beleive that the impedance and the speed of the B+ source also plays a great role.
Some "drawback" in it - more or less hidden in auto bias - may be "revealed" with fixed bias.

"Beans counters compromises" other says "Good engineering practices"

Yves.

[Miles Prower]

Originally posted by alexmoose
Hey, I was looking into making a self bias using a CCS for an EL34 amp, but I just found a 24v transformer that I ripped out of an iced tea machine (the fruits of dumpster diving), and I noticed that after being rectified, its just around the right voltage for a fixed bias on an EL34.

I wanted to know what the advantages, and dis-advantages of a fixed bias? or will this iced-tea machine transfomer (SEC: 24v, 2 amps) not work?

The advantage of fixed bias is almost always lower THD. If you derive that fixed bias from a negative rail and a DC coupled cathode follower, you also get much improved transient overdrive behaviour. The main problem with cathode bias is that a transient overdrive that increases the current also increases the bias and pushes the operation into a more nonlinear portion of the loadline. This distortion can persist until that extra voltage leaks off. This will degrade the sound.

The main disadvantages is the need for a negative rail (really NBD with solid state diodes in the main PS) and that it is not a "set it and forget it" proposition. You will need to keep an eye on current balance periodically and adjust as required. It is also less forgiving of inherent tube mismatch, and will always require touch up if changing tubes.

Regardless, I much prefer fixed bias.

[arnoldc]

guys, I'm looking at some schematics of Antique Sound Labs fixed bias 2A3, 300B amps and their bias circuit is sooooo simple- just a diode, a cap and a couple of resistors.

What will make a good bias circuit? Regulated Tube or MOSFET or whatever?

[Majestic]

I've not played around with SE enough to be able to offer an informed opinion.

But, under static conditions; fixed bias seems to offer
- less measurable distortion
- more efficiency (less heat wasted)
- the opportunity to finally remove that darned sonically-deleterious bypass cap

However...
- It's also been stated that fixed bias takes longer to recover from overload, and that this is one reason why cathode-bias is purported to sound better, even despite the now missing bypass-cap. For more on this see: "The Fixed-Bias Story" by Herbert Ravenswood, February 1958, Radio-Electronics magazine, pages 47-49. (ask me if you need a PDF copy)

As with all things, YMMV.
My suspiscion is that some amps in some settings will sound better to some listeners, and others won't.

[Yvesm]

Originally posted by arnoldc
guys, I'm looking at some schematics of Antique Sound Labs fixed bias 2A3, 300B amps and their bias circuit is sooooo simple- just a diode, a cap and a couple of resistors.

What will make a good bias circuit? Regulated Tube or MOSFET or whatever?

Primarly, a lo R and Z one.
Use at least a decent bleeder.

Majestic : An hi R or Z bias source ceratainly does not help for fast recovery from an overload !
Serie regulators are the worst in this situation cos they can't "suck" the grid current.
Shunt ones had my preference.

Be aware to DO NOT regulate the bias voltage without stabilizing plate voltage too. Not so easy to do.
With tetrodes or penthodes, just the screens need to be regulated.

Consider also simply changing the existing cathode resistor for a voltage source like zener or led chain As SY says !

Yves.