Is there any=20 real advantage to cathode bias? - AudioKarma.org Home Audio Stereo = Discussion=20 Forums

Find all posts by Rod = Beauvex

=20

#2 = =09

04-02-2012, 02:09 PM

=20

battradio<= /A> 3D"battradio

Join Date: Mar 2009

Location: Near ST. Louis MO

Posts: 4,540

=09 =20

It , is self limiting , the = more=20 current the tube draws , the more bias it produces .In some ways = it more=20 linear than fixed bias .

=20

__________________
3D""

I just hate that when you wake up and find out = you are=20 Conti the brain damaged robot and even worse than that the perfect = audio=20 system you had is gone !
Mark

= =09

=20

battradio

Find all posts by = battradio

=20

#3 = =09

04-02-2012, 02:30 PM

=20

for_p1

Join Date: Jan 2012

Location: NJ

Posts: 3,113

=09 =20

It provides negative = feedback either=20 at DC (when bypassed by large cap) or at operation frequencies. = Thus it=20 either stabilizes operation point or reduces distortion. =

=20

= =09

=20

for_p1

Find all posts by = for_p1

=20

#4 = =09

04-02-2012, 03:24 PM

=20

mhardy6647 3D"mhardy6647

Join Date: Sep 2002

Location: New England

Posts: 46,738

=09 =20

cheap and simple to = implement, too.

=20

__________________
Preamps need drive, not = gain. - G.=20 Kaufman

=20

= =09

=20

mhardy6647

Find all posts by = mhardy6647

=20

#5 = =09

04-02-2012, 04:04 PM

=20

gadget73

Join Date: Dec 2006

Location: Southern NJ

Posts: 21,227

=09 =20

Its very reliable. No = worries about=20 the bias supply failing and causing the tubes to self-destruct and = no=20 worries about someone mis-adjusting it and causing damage. =

=20

__________________
solid stuff, tube stuff, its = all good=20 stuff

=20

= =09

=20

gadget73

Find all posts by = gadget73

=20

Audiokarma

=20 =20 =

=20

#6 = =09

04-02-2012, 04:42 PM

=20

Celt

=20

Yoko Phono

Join Date: Nov 2002

Location: Paragould, Arkansas

Posts: 37,083

=09 =20

Best way to do it, imho. =

=20

__________________
3D""

Let me live in the house = beside the=20 road and be a friend to man.

= =20

= =09

=20

Celt

Find all posts by = Celt

=20

#7 = =09

04-02-2012, 06:12 PM

=20

for_p1

Join Date: Jan 2012

Location: NJ

Posts: 3,113

=09 =20

Quote:

Originally Posted by Celt

Best way to do it,=20 imho.

No, the best way to = do it is=20 fixed bias with servo. You can do a lot of things with it, like = signal=20 dependent bias point.

=20

= =09

=20

for_p1

Find all posts by = for_p1

=20

#8 = =09

04-02-2012, 11:31 PM

=20

dcgillespi= e 3D"dcgillespie

=20

Fisher SA-100 Clone

Join Date: Mar 2010

Location: Ball Ground, GA

Posts: 5,667

=09 =20

Would not a signal = dependent bias=20 point not be just another way to describe cathode = bias?

Dave

=20

= =09

=20

dcgillespie

Find all posts by = dcgillespie

=20

#9 = =09

04-02-2012, 11:38 PM

=20

BuzzK

=20

AK Member

Join Date: Nov 2009

Posts: 1,386

=09 =20

Dave,

What's your = opinion on=20 cathode bias vs. fixed?

Some say the midrange is clearer = with=20 cathode bias, but not the bass or top end. I know there are = probably no=20 universals, but . . . . ?

= =20

= =09

=20

BuzzK

Find all posts by = BuzzK

=20

#10 = =09

04-03-2012, 12:02 AM

=20

Shadowdog<= /A> 3D"Shadowdog

Join Date: May 2009

Location: Flin Flon, Manitoba

Posts: 1,641

=09 =20

It also depends on the = primary=20 impedance of the output transformers in a given circuit which type = can be=20 used. Cathode bias favors a higher primary so if you converted = some amps=20 to fixed it may not be a benefit sonic wise. Credit goes to Dave = on his=20 many posts and paper on his EFB mod and answering my = questions.

Randy

=20

= =09

=20

Shadowdog

Find all posts by = Shadowdog

=20

Audiokarma

=20 =20 =

=20

#11 = =09

04-03-2012, 01:12 AM

=20

audiodon

Da due lund lund lund

Join Date: Oct 2008

Location: Arlington, MA

Posts: 9,479

=09 =20

When a cathode biased amp = starts to=20 run out of steam, it asks too much of the power supply, limiting = the=20 available low end and distorting the high end.
My favorite = Fisher tube=20 integrated is cathode biased.
Some call a cathode biased amp a = working=20 man's amp.
A Harman Kardon 250 is supposed to be cathode = biased, but=20 I've never seen one up close.
I hadn't thought of it as a = midrange amp,=20 but I do like my midrange, and I count on a powered sub for the = lows, so=20 that makes sense to me.

=20

__________________
If you want something done, ask = a busy=20 person.
My reference system: http://www.audiokarma.org/forums/showthread.php?t=3D364= 201&highlight=3Da

=20

= =09

=20

audiodon

Find all posts by = audiodon

=20

#12 = =09

04-03-2012, 02:04 AM

=20

Shadowdog<= /A> 3D"Shadowdog

Join Date: May 2009

Location: Flin Flon, Manitoba

Posts: 1,641

=09 =20

From what I've read if you = use an=20 output transformer with greater primary impedance than you would = for fixed=20 bias you should be able to minimize the negatives you mentioned. = Eico did=20 that with the HF-87/89, and looks like most big big amps are and = were=20 back in the day fixed bias

Many small amps (6V6/ EL84 = etc.) with=20 cathode bias probably could benefit in the conversion to fixed = bias as=20 they have output transformers with the right primary impedance I = would=20 think.

Randy

=20

= =09

=20

Shadowdog

Find all posts by = Shadowdog

=20

#13 = =09

04-03-2012, 09:04 AM

=20

dcgillespi= e 3D"dcgillespie

=20

Fisher SA-100 Clone

Join Date: Mar 2010

Location: Ball Ground, GA

Posts: 5,667

=09 =20

The original question posed = is so=20 general that it invites a number of possible answers. So the = general=20 answer to the general question is that no doubt the real advantage = of=20 cathode bias is the economic one.

From a strict standpoint = of=20 circuit theory however, while there are situations where cathode = bias can=20 perform as well as fixed bias, it can never perform better than = fixed=20 bias. But even that statement needs a qualifier.

If we = confine the=20 definition of performance to just power output and distortion, = then with=20 all else equal, cathode bias can never provide the same power = output as=20 traditional fixed bias can, because of the voltage loss across the = cathode=20 resistor, or whatever device is placed in the cathode circuit to = raise its=20 voltage to be more positive than that of grid #1 (the control = grid), all=20 relative to circuit ground. Any voltage lost across a cathode bias = device=20 is voltage that cannot otherwise appear (or be dropped) across the = OPT=20 primary winding, and therefore represents a loss of power = output.

Regarding distortion, the only time cathode bias can produce the = same low=20 distortion levels as fixed bias can, is with a true class A = push-pull=20 amplifier. In such an amplifier, the voltage across a cathode bias = resistor remains as stable as if fixed bias were used, because the = current=20 draw through the two output tubes is exactly equal and opposite at = all=20 times through out the entire power output range of the amplifier.=20 Therefore, in this scenario, distortion will be equally low which = ever=20 bias scheme is used.

None of this is to imply that cathode = bias=20 designs deliver "bad" performance. While they still cannot = ultimately=20 deliver the same performance as that of an equivalent fixed bias = design,=20 they can deliver quite good performance when designed correctly. = The words=20 "designed correctly" means that adjustments have to be made to = account for=20 the fact that cathode bias is to be used, where as with fixed bias = designs, no such design corrections are necessary. =

Virtually all=20 consumer oriented vacuum tube audio amplifiers of push-pull design = operate=20 in class AB1. That means that after a very low power output level = has been=20 achieved, the operation of the output stage changes from one of = class A,=20 to that of class B, and does so automatically. When the switch to = class B=20 operation commences, the current draw through the two output tubes = becomes=20 anything but equal and opposite, while the total current draw for = the two=20 tubes starts to rise dramatically. Therefore, if cathode bias is = used, the=20 bias voltage across the bias resistor will start to rise and = fluctuate=20 significantly as well. To keep the cathode bias voltage as stable = as=20 possible -- very necessary to achieve the lowest possible = distortion -- a=20 capacitor is placed across the resistor to minimize the = fluctuations. In=20 practice, it works pretty well. But the capacitor hardly makes the = output=20 stage operate exactly like that of a fixed biased output stage. =

While the cathode bypass cap can work to limit bias fluctuations = across=20 the bias resistor due to fluctuations in peak current flow as = power output=20 increases (and therefore work to maximize available power output), = it=20 cannot limit the increase due to the average increase in current = flow.=20 That is why the cathode bias voltage will always increase in a = cathode=20 biased amplifier as power output increases. This bias increase due = to=20 increased average current flow through the output stage then = shifts the=20 operating conditions of the stage, requiring the OPT primary = impedance to=20 be adjusted to reflect best performance at the point the operating = conditions shift to under dynamic (i.e. power output producing)=20 conditions. Because the shift is to one of less maximum possible = current=20 flow, the OPT primary impedance must be raised to accommodate this = shift=20 to maintain reasonable distortion levels. Raising the primary = impedance=20 also means an increased turns ratio in the OPT, meaning that on = top of the=20 power output lost through the cathode bias resistor, even more = power is=20 lost due to the raised OPT impedance required. But even though = power=20 output is reduced, with proper adjustment to the load line, = distortion=20 performance can actually be quite good.

There is an = additional=20 downside to cathode bias however, and that is that typical = quiescent=20 currents must be raised to account for the inevitable shift in = operating=20 conditions that cathode biasing produces. Raising the quiescent = current=20 helps to minimize the overall shift that occurs, maximizing = overall=20 performance from this type of bias scheme. That is why so many = cathode=20 biased amplifiers run the tubes right at the limits of practical=20 dissipation levels. By comparison, equivalent fixed bias designs = can=20 usually operate at lower quiescent levels to achieve the same = levels of=20 overall performance. While higher quiescent current levels will = affect=20 tube life somewhat, it does not affect it nearly as much as might = be=20 expected, since the tubes were ultimately designed to pass peak = currents=20 far in excess of the quiescent currents drawn -- even in cathode = biased=20 designs.

Ultimately then, power output is always reduced = with=20 cathode bias, but distortion can still be made to be quite = reasonable at=20 the further expense of increase quiescent current draw. However, = as long=20 as enough usable low distortion power can be safely generated with = cathode=20 bias, then the practical sides of cathode biasing come into play; = i.e. the=20 auto quiescent current adjustment feature, economy, etc. Since the = 6BQ5=20 family of tubes can safely develop a very respectable 13-14 watts = RMS=20 power output with cathode bias, and because so many wonderfully = efficient=20 speakers exist that work so well with this power level, it's no = wonder=20 then that with all things considered, the cathode biased output = stage=20 (particularly using 6BQ5 type tubes) was by far the most popular = design of=20 all.

Hopefully, this will help explain the design = differences and=20 the considerations required of the two approaches -- and even help = to=20 dispel some myths along the way!

I hope this = helps!

Dave

=20

= =09

=20

dcgillespie

Find all posts by = dcgillespie

=20

#14 = =09

04-03-2012, 10:17 AM

=20

BuzzK

=20

AK Member

Join Date: Nov 2009

Posts: 1,386

=09 =20

Thanks, Dave. Great post. = Any=20 chance it could become a sticky?

= =20

= =09

=20

BuzzK

Find all posts by = BuzzK

=20

#15 = =09

04-03-2012, 05:13 PM

=20

jaymanaa

More Wag Less Bark

Join Date: Mar 2006

Location: OZ

Posts: 51,218

=09 =20

Quote:

Originally Posted by dcgillespie = 3D"View

The original question = posed is so=20 general that it invites a number of possible answers. So the = general=20 answer to the general question is that no doubt the real = advantage=20 of cathode bias is the economic one.

From a strict=20 standpoint of circuit theory however, while there are = situations=20 where cathode bias can perform as well as fixed bias, it can = never=20 perform better than fixed bias. But even that statement = needs a=20 qualifier.

If we confine the definition of = performance to=20 just power output and distortion, then with all else equal, = cathode=20 bias can never provide the same power output as traditional = fixed=20 bias can, because of the voltage loss across the cathode = resistor,=20 or whatever device is placed in the cathode circuit to raise = its=20 voltage to be more positive than that of grid #1 (the = control grid),=20 all relative to circuit ground. Any voltage lost across a = cathode=20 bias device is voltage that cannot otherwise appear (or be = dropped)=20 across the OPT primary winding, and therefore represents a = loss of=20 power output.

Regarding distortion, the only time = cathode=20 bias can produce the same low distortion levels as fixed = bias can,=20 is with a true class A push-pull amplifier. In such an = amplifier,=20 the voltage across a cathode bias resistor remains as stable = as if=20 fixed bias were used, because the current draw through the = two=20 output tubes is exactly equal and opposite at all times = through out=20 the entire power output range of the amplifier. Therefore, = in this=20 scenario, distortion will be equally low which ever bias = scheme is=20 used.

None of this is to imply that cathode bias = designs=20 deliver "bad" performance. While they still cannot = ultimately=20 deliver the same performance as that of an equivalent fixed = bias=20 design, they can deliver quite good performance when = designed=20 correctly. The words "designed correctly" means that = adjustments=20 have to be made to account for the fact that cathode bias is = to be=20 used, where as with fixed bias designs, no such design = corrections=20 are necessary.

Virtually all consumer oriented = vacuum tube=20 audio amplifiers of push-pull design operate in class AB1. = That=20 means that after a very low power output level has been = achieved,=20 the operation of the output stage changes from one of class = A, to=20 that of class B, and does so automatically. When the switch = to class=20 B operation commences, the current draw through the two = output tubes=20 becomes anything but equal and opposite, while the total = current=20 draw for the two tubes starts to rise dramatically. = Therefore, if=20 cathode bias is used, the bias voltage across the bias = resistor will=20 start to rise and fluctuate significantly as well. To keep = the=20 cathode bias voltage as stable as possible -- very necessary = to=20 achieve the lowest possible distortion -- a capacitor is = placed=20 across the resistor to minimize the fluctuations. In = practice, it=20 works pretty well. But the capacitor hardly makes the output = stage=20 operate exactly like that of a fixed biased output stage. =

While the cathode bypass cap can work to limit bias = fluctuations=20 across the bias resistor due to fluctuations in peak current = flow as=20 power output increases (and therefore work to maximize = available=20 power output), it cannot limit the increase due to the = average=20 increase in current flow. That is why the cathode bias = voltage will=20 always increase in a cathode biased amplifier as power = output=20 increases. This bias increase due to increased average = current flow=20 through the output stage then shifts the operating = conditions of the=20 stage, requiring the OPT primary impedance to be adjusted to = reflect=20 best performance at the point the operating conditions shift = to=20 under dynamic (i.e. power output producing) conditions. = Because the=20 shift is to one of less maximum possible current flow, the = OPT=20 primary impedance must be raised to accommodate this shift = to=20 maintain reasonable distortion levels. Raising the primary = impedance=20 also means an increased turns ratio in the OPT, meaning that = on top=20 of the power output lost through the cathode bias resistor, = even=20 more power is lost due to the raised OPT impedance required. = But=20 even though power output is reduced, with proper adjustment = to the=20 load line, distortion performance can actually be quite=20 good.

There is an additional downside to cathode bias = however, and that is that typical quiescent currents must be = raised=20 to account for the inevitable shift in operating conditions = that=20 cathode biasing produces. Raising the quiescent current = helps to=20 minimize the overall shift that occurs, maximizing overall=20 performance from this type of bias scheme. That is why so = many=20 cathode biased amplifiers run the tubes right at the limits = of=20 practical dissipation levels. By comparison, equivalent = fixed bias=20 designs can usually operate at lower quiescent levels to = achieve the=20 same levels of overall performance. While higher quiescent = current=20 levels will affect tube life somewhat, it does not affect it = nearly=20 as much as might be expected, since the tubes were = ultimately=20 designed to pass peak currents far in excess of the = quiescent=20 currents drawn -- even in cathode biased designs. =

Ultimately=20 then, power output is always reduced with cathode bias, but=20 distortion can still be made to be quite reasonable at the = further=20 expense of increase quiescent current draw. However, as long = as=20 enough usable low distortion power can be safely generated = with=20 cathode bias, then the practical sides of cathode biasing = come into=20 play; i.e. the auto quiescent current adjustment feature, = economy,=20 etc. Since the 6BQ5 family of tubes can safely develop a = very=20 respectable 13-14 watts RMS power output with cathode bias, = and=20 because so many wonderfully efficient speakers exist that = work so=20 well with this power level, it's no wonder then that with = all things=20 considered, the cathode biased output stage (particularly = using 6BQ5=20 type tubes) was by far the most popular design of = all.

Hopefully, this will help explain the design differences and = the=20 considerations required of the two approaches -- and even = help to=20 dispel some myths along the way!

I hope this = helps!

Dave

And I was going to = answer the=20 thread title with a simple "No". 3D""=20

I have to do some reading on your EFB = circuit. It kept=20 popping up last night whilst I was studying up on regulated screen = supplies.

May favorite cathode bias circuit if you can call = it=20 that, is the garter.

=20

__________________
"Only the dead have seen the = end of=20 war." Plato

=20

= =09

=20

jaymanaa