fixed bias circuits and pot wipers

[Don[_4_]]

I have seen two methods of using a pot in fixed bias circuits.
In one method, the wiper carries the voltage to the grid, and if the wiper
were to fail, the bias voltage goes to zero,
The other method has the wiper connected to another terminal of the pot. If
the wiper fails, the bias goes up to a high negative voltage. This last method
seems best to me. Am I missing something here? Why would the first method be
used? -Don

[Patrick Turner]

On Mar 30, 2:10*pm, Don > wrote:
> I have seen two methods of using a pot in fixed bias circuits.
> In one method, the wiper carries the voltage to the grid, and if the wiper
> were to fail, the bias voltage goes to zero,
> The other method has the wiper connected to another terminal of the pot. If
> the wiper fails, the bias goes up to a high negative voltage. This last method
> seems best to me. Am I missing something here? Why would the first method be
> used? -Don

Read my power amp pages about 5050, 8585 amps and all will be
revealed.

http://www.turneraudio.com.au/8585-amp-october-2006.html
http://www.turneraudio.com.au/Integrated5050.htm

Patrick Turner.

[mick]

On Tue, 29 Mar 2011 23:16:39 -0500, flipper wrote:

> On Tue, 29 Mar 2011 23:10:17 -0400, Don > wrote:
>
>>I have seen two methods of using a pot in fixed bias circuits. In one
>>method, the wiper carries the voltage to the grid, and if the wiper were
>>to fail, the bias voltage goes to zero, The other method has the wiper
>>connected to another terminal of the pot. If the wiper fails, the bias
>>goes up to a high negative voltage. This last method seems best to me.
>>Am I missing something here? Why would the first method be used? -Don
>
> Could be that a lot of people don't consider fault conditions but the
> plain ole wiper tap is simple and linear.
>
> Consider that in a simple 'traditional' configuration the bias pot is
> last in a series of resistors beginning with a plain ole RC filter.
> Wiper tied to one end makes it a 'variable resistor', as opposed to a
> tap, so, depending on how everything else is arranged, it isn't linear
> and impedance changes with position, which can bias shift more than just
> the one being adjusted.


That's not likely though, is it? The grid is a highZ input, so the bias
chain isn't loaded all that much (ok, I'm leaving out class AB1 etc).
Suitable values, with the variable resistor at the earthy end of the
chain, should give a reasonable bias adjustment. A pot would suffer from
the same linearity problem anyway. Just use lower values. A bad wiper
connection will just bias the valve to cut-off then.

--
Mick (Working in a M$-free zone!)
Web: http://www.nascom.info
Filtering everything posted from googlegroups to kill spam.

[Don[_4_]]

In article <e753972e-1a9b-4e3f-9d52-
>, says...
>On Mar 30, 2:10*pm, Don > wrote:
>> I have seen two methods of using a pot in fixed bias circuits.
>> In one method, the wiper carries the voltage to the grid, and if the wiper
>> were to fail, the bias voltage goes to zero,
>> The other method has the wiper connected to another terminal of the pot. If
>> the wiper fails, the bias goes up to a high negative voltage. This last
method
>>
>> seems best to me. Am I missing something here? Why would the first method
be
>> used? -Don
>
>Read my power amp pages about 5050, 8585 amps and all will be
>revealed.
>
> http://www.turneraudio.com.au/8585-amp-october-2006.html
>http://www.turneraudio.com.au/Integrated5050.htm
>
>Patrick Turner.

It seems to me that in your 8585 amp, the pot wiper is connected to the grid,
and if failure occurs,
zero bias is on the grid. But your protection circuitry for the power tubes
will sense the problem, and cut off the amp.
So, Robert's your mother's brother, or things are a-ok.

The 50-50 amp seems to have "fixed" fixed bias, that's to say, the bias is set
with resistors, and is not adjustable.
The bias can be balanced, so that each tube has equal current. The pot wiper
that does the balancing goes to ground.
If the wiper were to fail, the output tubes would still have a neg. voltage
bias on the grids.

Do I understand correctly?
-Don

[Big Bad Bob]

On 03/29/11 20:10, Don so witilly quipped:
> I have seen two methods of using a pot in fixed bias circuits.
> In one method, the wiper carries the voltage to the grid, and if the wiper
> were to fail, the bias voltage goes to zero,
> The other method has the wiper connected to another terminal of the pot. If
> the wiper fails, the bias goes up to a high negative voltage. This last method
> seems best to me. Am I missing something here? Why would the first method be
> used? -Don
>

inexpensive solution. Use a high reliability potentiometer and there
shouldn't be a problem. Cheap carbon 'open to the air' ones are noisy
anyway, so hi-fi equipment will probably use decent quality wirewound
pots that are sealed against the elements. Such a device is likely to
last indefinitely, longer than the tubes or transformers, even if it's
never adjusted again.

[Big Bad Bob]

On 03/30/11 01:47, Patrick Turner so witilly quipped:
> On Mar 30, 2:10 pm, > wrote:
>> I have seen two methods of using a pot in fixed bias circuits.
>> In one method, the wiper carries the voltage to the grid, and if the wiper
>> were to fail, the bias voltage goes to zero,
>> The other method has the wiper connected to another terminal of the pot. If
>> the wiper fails, the bias goes up to a high negative voltage. This last method
>> seems best to me. Am I missing something here? Why would the first method be
>> used? -Don
>
> Read my power amp pages about 5050, 8585 amps and all will be
> revealed.
>
> http://www.turneraudio.com.au/8585-amp-october-2006.html
> http://www.turneraudio.com.au/Integrated5050.htm
>
> Patrick Turner.

interesting use of constant current sources via transistors, even in the
plate circuits of the triodes, and the P-P 6BQ5 driver setup with
constant (combined) current on the 2 6BQ5 cathodes.

[Patrick Turner]

On Mar 31, 5:37*am, mick > wrote:
> On Tue, 29 Mar 2011 23:16:39 -0500, flipper wrote:
> > On Tue, 29 Mar 2011 23:10:17 -0400, Don > wrote:
>
> >>I have seen two methods of using a pot in fixed bias circuits. In one
> >>method, the wiper carries the voltage to the grid, and if the wiper were
> >>to fail, the bias voltage goes to zero, The other method has the wiper
> >>connected to another terminal of the pot. If the wiper fails, the bias
> >>goes up to a high negative voltage. This last method seems best to me.
> >>Am I missing something here? Why would the first method be used? -Don
>
> > Could be that a lot of people don't consider fault conditions but the
> > plain ole wiper tap is simple and linear.
>
> > Consider that in a simple 'traditional' configuration the bias pot is
> > last in a series of resistors beginning with a plain ole RC filter.
> > Wiper tied to one end makes it a 'variable resistor', as opposed to a
> > tap, so, depending on how everything else is arranged, it isn't linear
> > and impedance changes with position, which can bias shift more than just
> > the one being adjusted.
>
> That's not likely though, is it? The grid is a highZ input, so the bias
> chain isn't loaded all that much (ok, I'm leaving out class AB1 etc).
> Suitable values, with the variable resistor at the earthy end of the
> chain, should give a reasonable bias adjustment. A pot would suffer from
> the same linearity problem anyway. Just use lower values. A bad wiper
> connection will just bias the valve to cut-off then.
>
> --
> Mick * * * * * * * * * * *(Working in a M$-free zone!)
> Web:http://www.nascom.info
> Filtering everything posted from googlegroups to kill spam. * *- Hide quoted text -
>
> - Show quoted text -

Bias circuits are prone to ageing and failure. I like to use 25mm dia
"heavy duty" 10k wire wound pots rated for 2 watts for the bias pot.
Bean counters don't, and use the cheapest small crap junk they can
find. I like to always place a bypassing resistance of say 470k from a
wiper to the -bias supply or - end of the pot wherever the bias comes
from a wiper on a pot. If wiper goes open, tube gets an extra -Vdc
bias so tube cuts off, THD increases, owner detects something is
wrong, but smoke is avoided - unless a coupling cap has failed to
become a low resistance.

Patrick Turner.

[Patrick Turner]

On Mar 31, 7:04*am, Don > wrote:
> In article <e753972e-1a9b-4e3f-9d52-
> >, says....
>
>
>
>
>
> >On Mar 30, 2:10*pm, Don > wrote:
> >> I have seen two methods of using a pot in fixed bias circuits.
> >> In one method, the wiper carries the voltage to the grid, and if the wiper
> >> were to fail, the bias voltage goes to zero,
> >> The other method has the wiper connected to another terminal of the pot. If
> >> the wiper fails, the bias goes up to a high negative voltage. This last
> method
>
> >> seems best to me. Am I missing something here? Why would the first method
> be
> >> used? -Don
>
> >Read my power amp pages about 5050, 8585 amps and all will be
> >revealed.
>
> >http://www.turneraudio.com.au/8585-amp-october-2006.html
> >http://www.turneraudio.com.au/Integrated5050.htm
>
> >Patrick Turner.
>
> It seems to me that in your 8585 amp, the pot wiper is connected to the grid,
> and if failure occurs,
> zero bias is on the grid. But your protection circuitry for the power tubes
> will sense the problem, and cut off the amp.
> So, Robert's your mother's brother, or things are a-ok.
>
> The 50-50 amp seems to have "fixed" fixed bias, that's to say, the bias is set
> with resistors, and is not adjustable.
> The bias can be balanced, so that each tube has equal current. The pot wiper
> that does the balancing goes to ground.
> If the wiper were to fail, the output tubes would still have a *neg. voltage
> bias on the grids.
>
> Do I understand correctly?
> -Don- Hide quoted text -
>
> - Show quoted text -

Indeed you understand me correctly. I have lost count ofthe number of
times my protection circuits have stopped smoke production because of
the use of NOS tubes someone has bought in good faith. Maybe they were
50 years old and never been used. But after sitting on a shelf for 50
years a tube's glass-pin joint may have fatigued and the tube goes
gassy, causing bias failure. ( If you could preserve a new bride for
50 years before leading her gently by the hand to the bedroom to
consumate the marriage you would have to expect some difficulties,
including equipment failure, and once in the kitchen a NOS bride might
have even more serious problems ).

COWPAT = 1/N squared,
where COWPAT means Chance Of Working Perfectly Any Time,
N is the number of things which could dissallow functionality, or
things you didn't consider properly, or the mistakes you made, unknown
un-knowables, and perhaps a spanner which was dropped into a crank
case which you didn't know was there until you pressed the 'start'
button, or some old tube which awas destined to fail 1 month after a
50 year delay to "use consummation", or some inadvertent circuit
fault.

There is often a 100% chance of a circuit failing to work even with
one one un-considered thing. Even things that were carefully
considered can be wrong, and you'll find outabout when you flick the
on switch, or measure the circuit. Perfect working = maximally
optimised working and many things need to be got right.

I never like to assume anything is true until I proove it to myself,
and then I ask who else will believe me? So I have to proove I can
proove to others its right, not just to me.

The only way I stop **** happening is to be my own hardest critic.

Patrick Turner.

[Patrick Turner]

On Mar 31, 10:11*am, Big Bad Bob <BigBadBob-at-mrp3-
> wrote:
> On 03/30/11 01:47, Patrick Turner so witilly quipped:
>
>
>
>
>
> > On Mar 30, 2:10 pm, > *wrote:
> >> I have seen two methods of using a pot in fixed bias circuits.
> >> In one method, the wiper carries the voltage to the grid, and if the wiper
> >> were to fail, the bias voltage goes to zero,
> >> The other method has the wiper connected to another terminal of the pot. If
> >> the wiper fails, the bias goes up to a high negative voltage. This last method
> >> seems best to me. Am I missing something here? Why would the first method be
> >> used? -Don
>
> > Read my power amp pages about 5050, 8585 amps and all will be
> > revealed.
>
> > *http://www.turneraudio.com.au/8585-amp-october-2006.html
> >http://www.turneraudio.com.au/Integrated5050.htm
>
> > Patrick Turner.
>
> interesting use of constant current sources via transistors, even in the
> plate circuits of the triodes, and the P-P 6BQ5 driver setup with
> constant (combined) current on the 2 6BQ5 cathodes.- Hide quoted text -
>
> - Show quoted text -

Some may find my circuitry rather queer, and sneer at my use of
transistors which I maintain act as complicit slaves to every tiny
whim of current supply to the kingly tubes who rule masterfully over
the task of voltage amplification. The transistor is often seen as an
ugly ungracious troglodyte whose attempts to amplify do poison all
music; but when asked to work as a constant current slave they excel.

But next time you build an amp, remember you always could do worse
than try my ideas. If you do try something I posted, have a THD meter
standing by, make the tests, then have a listen, and the THD meters
will smile with nice numbers and your ears will be delighted.

Patrick Turner.

[John L Stewart]

On Mar 31, 7:04*am, Don > wrote:
> In article <e753972e-1a9b-4e3f-9d52-
> >, says....
>
>
>
>
>
> >On Mar 30, 2:10*pm, Don > wrote:
> >> I have seen two methods of using a pot in fixed bias circuits.
> >> In one method, the wiper carries the voltage to the grid, and if the wiper
> >> were to fail, the bias voltage goes to zero,
> >> The other method has the wiper connected to another terminal of the pot. If
> >> the wiper fails, the bias goes up to a high negative voltage. This last
> method
>
> >> seems best to me. Am I missing something here? Why would the first method
> be
> >> used? -Don
>
> >Read my power amp pages about 5050, 8585 amps and all will be
> >revealed.
>
> >http://www.turneraudio.com.au/8585-amp-october-2006.html
> >http://www.turneraudio.com.au/Integrated5050.htm
>
> >Patrick Turner.
>
> It seems to me that in your 8585 amp, the pot wiper is connected to the grid,
> and if failure occurs,
> zero bias is on the grid. But your protection circuitry for the power tubes
> will sense the problem, and cut off the amp.
> So, Robert's your mother's brother, or things are a-ok.
>
> The 50-50 amp seems to have "fixed" fixed bias, that's to say, the bias is set
> with resistors, and is not adjustable.
> The bias can be balanced, so that each tube has equal current. The pot wiper
> that does the balancing goes to ground.
> If the wiper were to fail, the output tubes would still have a *neg. voltage
> bias on the grids.
>
> Do I understand correctly?
> -Don- Hide quoted text -
>
> - Show quoted text -

Indeed you understand me correctly. I have lost count ofthe number of
times my protection circuits have stopped smoke production because of
the use of NOS tubes someone has bought in good faith. Maybe they were
50 years old and never been used. But after sitting on a shelf for 50
years a tube's glass-pin joint may have fatigued and the tube goes
gassy, causing bias failure. ( If you could preserve a new bride for
50 years before leading her gently by the hand to the bedroom to
consumate the marriage you would have to expect some difficulties,
including equipment failure, and once in the kitchen a NOS bride might
have even more serious problems ).

COWPAT = 1/N squared,
where COWPAT means Chance Of Working Perfectly Any Time,
N is the number of things which could dissallow functionality, or
things you didn't consider properly, or the mistakes you made, unknown
un-knowables, and perhaps a spanner which was dropped into a crank
case which you didn't know was there until you pressed the 'start'
button, or some old tube which awas destined to fail 1 month after a
50 year delay to "use consummation", or some inadvertent circuit
fault.

There is often a 100% chance of a circuit failing to work even with
one one un-considered thing. Even things that were carefully
considered can be wrong, and you'll find outabout when you flick the
on switch, or measure the circuit. Perfect working = maximally
optimised working and many things need to be got right.

I never like to assume anything is true until I proove it to myself,
and then I ask who else will believe me? So I have to proove I can
proove to others its right, not just to me.

The only way I stop **** happening is to be my own hardest critic.

Patrick Turner.

The lower silicon rectifier in the 5050 negative supply is shown wrong way around. Regards, John

[Patrick Turner]

On Apr 1, 3:52*am, John L Stewart <John.L.Stewart.
> wrote:
> Patrick Turner;928708 Wrote:
>
>
>
>
>
> > On Mar 31, 7:04*am, Don > wrote:-
> > > In article <e753972e-1a9b-4e3f-9d52-
> > > >,
> > says....
>
> > >-
> > > >On Mar 30, 2:10*pm, Don > wrote:-
> > > >> I have seen two methods of using a pot in fixed bias circuits.
> > > >> In one method, the wiper carries the voltage to the grid, and if
> > the wiper
> > > >> were to fail, the bias voltage goes to zero,
> > > >> The other method has the wiper connected to another terminal of the
> > pot. If
> > > >> the wiper fails, the bias goes up to a high negative voltage. This
> > last--
> > > method
> > >--
> > > >> seems best to me. Am I missing something here? Why would the first
> > method--
> > > be--
> > > >> used? -Don--
> > >-
> > > >Read my power amp pages about 5050, 8585 amps and all will be
> > > >revealed.-
> > >-
> > > >http://www.turneraudio.com.au/8585-amp-october-2006.html
> > > >http://www.turneraudio.com.au/Integrated5050.htm-
> > >-
> > > >Patrick Turner.-
>
> > > It seems to me that in your 8585 amp, the pot wiper is connected to
> > the grid,
> > > and if failure occurs,
> > > zero bias is on the grid. But your protection circuitry for the power
> > tubes
> > > will sense the problem, and cut off the amp.
> > > So, Robert's your mother's brother, or things are a-ok.
>
> > > The 50-50 amp seems to have "fixed" fixed bias, that's to say, the
> > bias is set
> > > with resistors, and is not adjustable.
> > > The bias can be balanced, so that each tube has equal current. The pot
> > wiper
> > > that does the balancing goes to ground.
> > > If the wiper were to fail, the output tubes would still have a *neg.
> > voltage
> > > bias on the grids.
>
> > > Do I understand correctly?
> > > -Don- Hide quoted text -
>
> > > - Show quoted text --
>
> > Indeed you understand me correctly. I have lost count ofthe number of
> > times my protection circuits have stopped smoke production because of
> > the use of NOS tubes someone has bought in good faith. Maybe they were
> > 50 years old and never been used. But after sitting on a shelf for 50
> > years a tube's glass-pin joint may have fatigued and the tube goes
> > gassy, causing bias failure. ( If you could preserve a new bride for
> > 50 years *before leading her gently by the hand to the bedroom to
> > consumate the marriage you would have to expect some difficulties,
> > including equipment failure, and once in the kitchen a NOS bride might
> > have even more serious problems ).
>
> > COWPAT = 1/N squared,
> > where COWPAT means Chance Of Working Perfectly Any Time,
> > N is the number of things which could dissallow functionality, or
> > things you didn't consider properly, or the mistakes you made, unknown
> > un-knowables, and perhaps a spanner which was dropped into a crank
> > case which you didn't know was there until you pressed the 'start'
> > button, or some old tube which awas destined to fail 1 month after a
> > 50 year delay to "use consummation", or some inadvertent circuit
> > fault.
>
> > There is often a 100% chance of a circuit failing *to work even with
> > one one un-considered thing. Even things that were carefully
> > considered can be wrong, and you'll find outabout when you flick the
> > on switch, or measure the circuit. Perfect working = maximally
> > optimised working and many things need to be got right.
>
> > I never like to assume anything is true until I proove it to myself,
> > and then I ask who else will believe me? So I have to proove I can
> > proove to others its right, not just to me.
>
> > The only way I stop **** happening is to be my own hardest critic.
>
> > Patrick Turner.
>
> The lower silicon rectifier in the 5050 negative supply is shown wrong
> way around. * *Regards, John
>
Its all fixed up now, I re-drawed the schematic.
I must re-draw the others soon while I go though all pages.

Patrick Turner.

[Big Bad Bob]

On 03/31/11 00:29, Patrick Turner so wittily quipped:
> On Mar 31, 5:37 am, > wrote:
>> On Tue, 29 Mar 2011 23:16:39 -0500, flipper wrote:
>>> On Tue, 29 Mar 2011 23:10:17 -0400, > wrote:
>>
>>>> I have seen two methods of using a pot in fixed bias circuits. In one
>>>> method, the wiper carries the voltage to the grid, and if the wiper were
>>>> to fail, the bias voltage goes to zero, The other method has the wiper
>>>> connected to another terminal of the pot. If the wiper fails, the bias
>>>> goes up to a high negative voltage. This last method seems best to me.
>>>> Am I missing something here? Why would the first method be used? -Don
>>
>>> Could be that a lot of people don't consider fault conditions but the
>>> plain ole wiper tap is simple and linear.
>>
>>> Consider that in a simple 'traditional' configuration the bias pot is
>>> last in a series of resistors beginning with a plain ole RC filter.
>>> Wiper tied to one end makes it a 'variable resistor', as opposed to a
>>> tap, so, depending on how everything else is arranged, it isn't linear
>>> and impedance changes with position, which can bias shift more than just
>>> the one being adjusted.
>>
>> That's not likely though, is it? The grid is a highZ input, so the bias
>> chain isn't loaded all that much (ok, I'm leaving out class AB1 etc).
>> Suitable values, with the variable resistor at the earthy end of the
>> chain, should give a reasonable bias adjustment. A pot would suffer from
>> the same linearity problem anyway. Just use lower values. A bad wiper
>> connection will just bias the valve to cut-off then.
>>
>> --
>> Mick (Working in a M$-free zone!)
>> Web:http://www.nascom.info
>> Filtering everything posted from googlegroups to kill spam. - Hide quoted text -
>>
>> - Show quoted text -
>
> Bias circuits are prone to ageing and failure. I like to use 25mm dia
> "heavy duty" 10k wire wound pots rated for 2 watts for the bias pot.
> Bean counters don't, and use the cheapest small crap junk they can
> find. I like to always place a bypassing resistance of say 470k from a
> wiper to the -bias supply or - end of the pot wherever the bias comes
> from a wiper on a pot. If wiper goes open, tube gets an extra -Vdc
> bias so tube cuts off, THD increases, owner detects something is
> wrong, but smoke is avoided - unless a coupling cap has failed to
> become a low resistance.

would a fast-blow fuse soldered between B+ and the output transformer be
more effective? The idea is that if it's a soldered-in fuse, a decent
tech would ask "why did this fuse blow" and investigate before just
replacing it. You hope. In any case fuses are cheap, and possibly more
justifiable to bean counters, if you call it a safety feature and a form
of liability insurance against smoking/flaming components.

[Big Bad Bob]

On 03/31/11 01:06, Patrick Turner so wittily quipped:
>> interesting use of constant current sources via transistors, even in the
>> plate circuits of the triodes, and the P-P 6BQ5 driver setup with
>> constant (combined) current on the 2 6BQ5 cathodes.

> Some may find my circuitry rather queer, and sneer at my use of
> transistors which I maintain act as complicit slaves to every tiny
> whim of current supply to the kingly tubes who rule masterfully over
> the task of voltage amplification.

sounds good to me. The thing that does the amplifying is the tube, and
the bias circuitry simply helps it work better. Complaining about a
transistor used as a constant current source is like complaining about a
resistor, since the transistor should not be adding any [detectable]
distortion or noise in that capacity.

> The transistor is often seen as an ugly ungracious troglodyte whose
> attempts to amplify do poison all music; but when asked to work as a
> constant current slave they excel.

absolutely - it works around the problems associated with cascode tube
circuits.

> But next time you build an amp, remember you always could do worse
> than try my ideas. If you do try something I posted, have a THD meter
> standing by, make the tests, then have a listen, and the THD meters
> will smile with nice numbers and your ears will be delighted.

No doubt. You're applying modern technology to old-school designs and
getting the best of both worlds [or at least that appears to be your goal].

I'm not above using combined solid state and tubes where it makes sense.
If what you want is the benefit of tube preamps, but the output stage
doesn't matter, then a few triodes with a solid state output would give
you what you want. Several guitar amplifiers do things like that. If
THD is way below audible in the power stage, it can be transistors OR
tubes and it wouldn't make a difference in the sound. Having the tubes
in the preamp would give you the 'tube preamp' sound, and that has its
own characteristic distortion quality [like a Marshall amp]. In other
cases tube output with tube rectifier may be a component of the sound,
where the voltage drop on the tube rectifier actually creates a kind of
'attack/sustain' effect on the sound [in this case distortion is in the
output stage, not the preamp], like an old Fender amp that only puts out
15W but sounds GREAT when you overdrive it. So you 'turn everything up
to 11' [ok 10 it's a Fender, not a Marshall] to get the full effect.
Anyway, for that sound the 'tube everything' is necessary. So yeah,
depending on what you want to do, use tubes, or solid state, whatever
makes sense.