Re: Battery bias -- positioning the battery for best sound



Andre Jute wrote:

Patrick Turner wrote:
Andre Jute wrote:

Battery bias -- positioning the battery for best sound -- a discussion
document with illustrations

The purpose of battery bias is to remove the soundshaping element of
the
capacitor from the cathode circuit of any tube stage. However, John
Byrns
points out that the battery itself is a soundshaping element and should
be
removed as far from the cathode as possible. Three possibilities are:

A. Batttery in Cathode
B. Batttery in Grid Leak
C. Battery in Grid

Which is the superior choice and why?

Text and illustrations from
http://members.lycos.co.uk/fiultra/Discussion%20Circuits.htm

If you look the battery in gridleak bias method, then methinks there may be
a mistake because the
battery is placed between wiper of the input potentiometer and 0V, and the
grid of the
417 input tube is also from wiper, so effectively the grid is shunted to 0V
via the battery.
Batteries are low impedance voltage supplies, so that grid leak schematic
would only make silence
if built into a circuit.

Ha! I altered that drawing; download a new copy at
http://members.lycos.co.uk/fiultra/Discussion%20Circuits.htm
It still doesn't work well, of course but that is the point of a
discussion document, to let people point out why something doesn't work
and why, so that it becomes an instrument of instruction. My own
preference, A, the battery in the cathode, works perfectly.

The image you have there now for the grid leak method ould flatten the
battery
soon,
BTW, the image is a strange one on my screen; it tries to appear far too
big, but
I am not
good at PC displays....

The battery in the cathode does allow for battery recharge if it is a
rechargable
type,
but the voltage must be just right, or else you may have to positively
or
negatively bias the brid to compensate...

Battery in series is OK because grid current will take a huge time to
discharge
the battery,
but in 10 years that battery type may not be available.




Grid leak biasing was in the past known as bias a bias voltage generated
across a large value
biasing resistance; there is no need for a battery at all, since the very
small grid current
flowing in a bias resistor is enough to generate the bias voltage if that
bias R is say 10Mohms.
And the bias voltage tends to self regulate, the more anode current, the
more grid current so bias increases and the anode voltage tends not to
swing so negatively.
Many old radios used grid leak bias on tubes which had a grounded cathode
for a 1st audio triode,
and with two anodes of diodes for detection, 6AV6, etc.
But with real grid leak bias you need a capacitor to couple from the grid
to the source,
and a low leakage cap at that.
The 10M bias resistance does produce a lot of noise so the cap coupling to
the source and source impedance
should be valued to shunt the noise of the high resistance of the grid leak
resistor.

One of the superb things about the simplest tube amps I like is that
they are to a very large extent self-adjusting. The transfer mechanism
is almost biological in its adaptability, in the way that from two
voltages on the anode and the grid it determines operating current. And
autobias is another wonderful self-adjustment. Wonders never cease for
us tubies.

Yes, but in a Bogen amp I recently completely re-designed and re-wired
for a
customer the second 12AX7 gain triode in the phono amp did have grid
leak bias,
so the coupling cap could be a low value and still have a long time
constant.
Rg was a high value Rg, but the anode voltage was only +20V, ie, there
was too
little Ig, so hardly any bias voltage.
Grid leak bias is a bean counter special, a cheap nasty way to bias
anything. Its
unpredictable sometimes.



Frankly, I hear nothing wrong with 1,000 uF caps as bypass caps on cathode
circuits,

Andre Schmeets wants to discuss caps. You guys should open a separate
thread. For a start, you and he could explain what is important, eg
ESR.

Perhaps it has been talked over before many times. The caps I use for
bypassing
all have much wider F range than ones we may have bought 30 years ago.
The nonlinearity of a 100uF Nichicon electro or most other generic
electros
is not a major nor even minor cause of sonic problems; MUCH bigger
things must be
attacked first before
worrying about the caps in the chain, IMHO.





I don't like using batteries at all, mainly becaiuse their life isn't long
perhaps; its another darn thing to replace/fix.

In a DIYer's amp battery life is more than long enough. The batteries
in my T68bis/T39 KISS proto have now served four years. Anyway, what
can they blow up? One or two 417A. which (despite John Stewart's dark
mutterings and my dark fears about the *future* supply of this tube)
are presently plentiful and cheap. Without even asking what he paid, I
bet I've never paid as much for a 417A as John paid for his E280CC.

I have zero problem with diyers. Please feel free to use what you like
and please
enjoy
what you hear, and enjoy your own theorizing about why things sound
different.
But all flavours of audio are available from diyers, flavours of
opinions as
well, and they differ,
sometimes leading to a serious clash which I don't want to aste time on.
So we all have our own truth, some a budhist, christian, muslim, jew,
God is
honoured by all.
So where I see ppl say Black Gates are the ONLY way to go, or Auricaps,
I raise
my hat,
I say enjoy, but I will still do it my way.

417A is a mighty little tube; I must try one sometime.

Did you know there was an E280F? its a frame grid pentode one up from
E180F.

The 280F is a seriously good pentode for some apps, and makes a serious
triode....





For those not liking cap couplings or electros, just use a divider to make
a negative fixed bias from a bias supply,
then apply the correctly adjusted bias via a 470k or whatever to the input
grid.

Fixed bias off any supply other than a battery is another soundshaping
element. And John says outright that battery is a soundshaping element,
and Chris probably suspects it.

Hmm, "sound shaping" sounds like amplifier "voicing".
I never know where to start to voice things; my mind is a pool of
thoughts
acumulated over
10 years of trials and errors, and I build intuitively, but with a close
eye on
the meters.

The timbre, sound shaping, voice, whatever I get in things I make is a
matter of
chance though because whatever anyone makes, several or many items in a
signal
path
all combine for a total effect, and so trying assign the sonics
perceived to a
change of just one part is difficult.
The sound of one hand clapping is hard to hear.




If the input source is from a CD player with opamp outputs which are say
400 ohms Rout,
then any noise or sonic behaviours of the bias supply filter caps is
attenuated by a factor of 400/470,000,
and I don't believe anyone could level any criticism about caps in the darn
signal path.

The other method involves unbypassed cathode resistances. No caps or
battereies used.
Nothing to squeal about.

Ha! In your dreams.

Oh, but there is! Where you have an unbypassed Rk, the effective output
resistance of the tube
can be much higher than just the Ra of the tube plus DC supply R plus
following cap coupled load all in parallel.
Witha well grounded cathode, the lowest of these 3 resistances is the Ra of
the triode, so
Rout is just a little lower than Ra, and hence the HF losses due to Miller
and stray C or cable C
affecting the gain stage is minimal; with unbypassed Rk, the Ra is
effectively increased
from just Ra to Ra + ( [µ + 1] x Rk), so in a 6SN7 where Ra could be 10k at
4 mA of Ia, if the 2.2k Rk is unbypassed,
Ra' effective is 10k + ( 21 x 2k2 ) = 56.2k, a huge increase which will
allow substantial losses
and loss of slam and dynamics.

Thanks, Patrick. It was particularly these large by-effects that I was
hoping to elicit by publishing a discussion document.

The benefits of the unbypassed Rk in an input stage are that THD is about
halved, but really at average levels
the THD is negligible anyway.

The truth is than unbypassed Rk sounds a bit chill to many people.
You're making my point, that these are not value-free decisions, that
to experienced audiophiles and especially DIYers and even more
especially those into high sensitivity speaks and therefore amps of
modest but sweet output these topological and component decisions make
a difference to the sound.

I would say that some of the engineering choices and their technical
results
do co-relate with subjective perceptions.



If the gain pot were placed after the input tube of an amp, then a 1V
signal from a CD player
is amplified to say 10Vrms+, then reduced to 1V again by the pot after the
triode input tube.
Althoughthe SNR is improved this way, THD/IMD is increased 10+ times, and
yet very few ppl
complain; they still like the SET sound even though THD at avearge levels
is raised from say
0.05% in the input tube with pot before the gain tube to 0.5%++ with pot
after the gain tube.

Vanishing THD is an overrated pastime for people who haven't put their
brains in gear since they left school. The question should always be,
What other advantage is there to an alteration that changes the THD,
either in its level or its composition between harmonics?

THD and the far more insidious resulting IMD is a cause for concern at
least.

With tube gear the THD/IMD can be high compared to state of art NFB
controlled SS
circuits,
But we hear more of the natural warmth and emotion conveyed by the
tubes.

I better get out my flame suit......

I heard a concert of mainly Liszt last night, some very complex dark
music,
difficult to play,
a challenge no doubt, and what the heck was going on in Liszt's mind?
I have no idea, but I thought that to re-produce the power and majesty
of that
grand
does take some real doing; the venue could only seat 20....





Where do the sonic effects of using capacitors fit into the scheme of
things?
Awkwardly, methinks.
I have a friend who swears there are no other good caps other than
Auricaps.
Lord knows what is so marvellous about them.
But he is relaxed after placing them in any amp he buys ( from china )
and is ill at ease with other folks amps when he finds out there are no
Auricaps within, just
630V Wima polyprops at a 20 dB lower price.

WTF! Those Wima caps from the RS catalogue are absolutely brilliant,
and cheap too.

Well that's how some people are constructed; there isn't any real logic.
Logic itself leads not to the truth by the way....
I am calm, just listen I say to him, and yes, i say to him, for you I
would only
give you
auricaps if I make an amp for you. But he actually likes tha last phono
stage I
did,
he said it was better than anything else he heard, and coming from such
a
difficult
person to please I nearly fell over.


The ear is subject to tension, or mental stress. muscles control the bones
between the ear drum
and other ear bits within, and if an audiophile is obsessive about
capacitors, then he needs to
have the right type of caps present to feel at ease before his ear muscles
can relax and let the real dynamics
of the music work the bones in his ears without the limiting action by
tense muscles.

Those Christian Brothers at your school were Jesuits in drag!

And my sisters were taught by nuns.

But whoever taught us didn't stop us enjoying sin as a pastime.



To see if there was anything in this outside the realms of audio (I
have Lowther PM6A well-treated with Ennemoser C37 Lack), I borrowed a
violin from Dieter Ennemoser, the controversial Austrian luthier (the
other instrument makers once had him jailed, now his instruments are in
the national museum) and showed it to local teachers and players and
luthiers. Dieter makes pink and blue violins, but you have to look
really hard to see the tint; it's not a Liberace job, more a very pale
wash over the natural colour of well-stained and varnished wood.
Several rated violinists told me the Ennemoser violin sounded better
for being tinted.

Nobody really knows what the best thing is to paint onto wood excited by
strings
we
scrape with horse hair.
Those old Italians knew something we may have forgotten.....



In my most recent line preamp, I used a humble 12AU7 as the gain element,
and I don't care what they say about 12AU7, its one of the most gloriously
accurate yet warm but not insipid triodes of all time.

Accuracy is a matter of very careful inplementation with a 12AU7. I
like the 12AU7 sound (I used it in my T113, a favourite amp) but for
unchallenged precision give me a 6SN7 any day.

True, and you can't deny a well found 6CG7 has its merits, but stocks of
Siemans
6CG7
are like rocking horse poo.



I am happy to use a well bypassed cathode and a transistor CCS to deliver
DC to the anode,

Been listening to Allen Wright again, have you then? His books are on
my shelf right next to the RDH.

Allen hates bypassing mostly, he hates polystyrene caps, he hates a
lotta things.

OK, so he's human, full of hate about some things. Are not we all?

But with some things like fet cascode phono circuits, I can share his
wonderments....



then have an open loop gain of 15 since the only ac load is a 100k or 50K
cap coupled gain pot
in parallel with the shunt feedback resistor to get the gain from 15 down
to say 4.
Does anyone complain about the FB, the coupling cap, the electro bypass,
the darn squalid state constant current source? not a living soul
complains; the consensus is that it makes the better sound compared to
simpler techniques.

An excuse among the microwatters is that those transistors lie outside
the signal path. It's bull of course; nothing is outside the signal
path.

Well although the tranny IS in the signal path, its effect in technical
terms of
THD is
beneficial, because the tube is relieved of having to produce power,
voltage x
current,
in a damn resistor used for merely conveying DC to the anode. The power
in the
Rdc
is entirely wasted, and THD provoking, and the R limits the amount of DC
able to
be
used in the triode concerned because an R has a voltage drop across it.
The CCS wins conditions for the tube to allow it to work more happily.
I find CCS sound better, and the ethereal atmosphere in music is better
conveyed.

The CCS with one lone MJE350 can have a finite resistance of say 20Meg
ohms,
and although it IS in the signal path, it also bows out, since its
effect is so
utterly
unoticeable; the CCS is the slave behind the scenes who works to keep
its tube
master happy
without causing any kind of fuss.

Perhaps pentodes are all female though; using CCS loads on pentode
signal stages
is fraught with problems of too much gain from the mistress.....

To which a pentode would say, ah, but all you triode men have too much
negative
feedback in you....


So as a result of all my experiences along my audiological life path I can
conclude that the
pet theories about best biasing cannot always be considered examples of the
ultimate truths about
how to set up a triode. The truth for anyone is what they find sounds best
after they have tried it out,
free of being affected by theories which have very little technical
authority; ie, not a stitch of real
evidence about distortions/artifacts to back them up.

Well, that's all right for us, because we build the gear we listen to
and can change it. But the poor *** who buys his stuff is at the
mercy of some manufacturer's pet theory.

Well worse, he buys what the bean counter specified....

And if he shops on the High
Street, he is at the mercy of the guys who think THD is THD, never mind
how it is constituted.

Indeed. 5% thd is horrible from SET the guys say.
But at 1/2 a watt from 300B? I just cross the street when i see the guys
a
comin....



Forget measurements, Patrick. This is about amps so simple, and so
silent already, that you can hear the sound of a single component. In
this case it is about a stage simplified to the point where all the
components will "sound" in order to choose among them. I know where I
will put my battery: where I've always had it. But it would be foolish
not to put up a discussion document and listen to what everyone else
has to say, especially those with relevant experience.

Each unto their own truth....



I am not one who does think biasing makes much difference if the working
point of a triode is the same either way and if the engineering is
basically OK. The triode will just do what triodes do, and caps and biasing
won't alter than much.

Then why fix the operating point of the triode with a CCS?

Well with the CCS in the anode circuit, current is fixed, but because of
the NFB
within
the triode, the anode voltage will stay put once the cathode resistance
is choden
right
or fixed grid bias is applied at the right voltage. The anode current
can be
trimmed as well....
Its better than an anode resistance.
Sure things will drift over years, but there are no batteries to go
flat, leak
gunk, or be replaced by dopey
owners who can't/won't find the right replacement.



For those who do think biasing affects the sound, then they should read all
about the Melquiades amplifiers whch are discussed in a forum at
http://www.goodsoundclub.com/Forums/ShowPost.aspx?PostID=2139

I am not overly impressed with Melquiades claims about biasing, or that it
could make any real sonic difference,
and in fact I think Mr Melquiades is unduly drawing attention to his bias
methods that is different again to anything
mentioned here at r.a.t, since I just don't see what difference his method
makes.
Methinks he just likes being in the limelight, and the attention he gets,
and he taught the cat in the images of his amps to
solder up his circuits and advise about the sound quality :-)

My cat doesn't presume to judge the sound. She is more concerned with
staying alive so that the cat on the landing below, the understudy cat,
does not inherit her comfortable seat in my study. She studies my
electrostats suspiciously...

But I have not heard his system about which he makes all these claims; it
is a triamped horn based thinge...
He has all sorts of ways of talking about what he's doing, it surely would
make make amusing reading to some,
I watch, I read, I wonder on the tapestry of life, all sorts of ways to do
an audio circuit,
but who really knows best?

God has ordered a 212E amp from me. SET of course. I'm trying to
persuade him to take one of my T113 Push-Pull triode-linked EL34. He
thinks he knows. I think I know. Who knows best is an irrelevant
question because next week I might change my mind.

Oh well, when you get called up There to please explain your position,
watch out for the God Of Triodes who will meet you, they say he is quite
grumpy.....

Not many 212E around, but KR make some nice 1610.
A couple of 845 instead will nearly satisfy a God they say.


Patrick Turner.




Patrick Turner.

Gee, I signed this letter earlier.

Ciao.

or directly as a pdf (recommended) from
http://members.lycos.co.uk/fiultra/Battery%20positon%209Mar06.pdf
or directly as a jpg from
http://members.lycos.co.uk/fiultra/Battery%20positon%209Mar06.jpg

Andre Jute
Visit Jute on Amps at http://members.lycos.co.uk/fiultra/
"wonderfully well written and reasoned information
for the tube audio constructor"
John Broskie TubeCAD & GlassWare
"an unbelievably comprehensive web site
containing vital gems of wisdom"
Stuart Perry Hi-Fi News & Record Review
.

Quantcast