| Author |
Message |
Ian Iveson
Guest
|
 Posted:
Mon Oct 17, 2005 4:43 am Post subject:
Re: Battery bias directly to grid |
|
|
At the foot of the attenuator? Assuming source blocks DC, and can
tolerate it.
cheers, Ian
"Andre Jute" <fiultra@yahoo.com> wrote in message
news:1129479453.815805.139670@z14g2000cwz.googlegroups.com...
| Quote: | I've posted a circuit for discussion at:
http://members.lycos.co.uk/fiultra/Discussion%20Circuits.htm
What are the implications of this circuit?
What, theoretically, makes it better than the standard battery
bias
application in the cathode circuit?
Will the batteries charge as efficiently in this position as in
the
cathode circuit; what are the effects of the resistors topping and
tailing the battery set?
What are this circuit's dangers?
Are there alternatives that work better or are safer?
Andre Jute
|
|
|
| Back to top |
|
 |
Jon Yaeger
Guest
|
| Posted:
Mon Oct 17, 2005 4:43 am Post subject:
Re: Battery bias directly to grid |
|
|
in article 1129488747.358162.79260@g44g2000cwa.googlegroups.com, Andre Jute
at fiultra@yahoo.com wrote on 10/16/05 2:52 PM:
| Quote: |
I constantly use battery bias in the cathode circuit. It gives you
better bass, as all kinds of fixed bias do. But John Byrns, the last
time we discussed the sound shaping of capacitors, said that the
battery is itself a soundshaping element and should be removed from the
cathode into the grid circuit.
|
*** I'm not an engineer, so I don't have the benefit of backing up my
opinions with good quantitative data.
Nonetheless, if simplicity is your goal I can't imagine how you could
benefit from putting something as complex and as variable as a NiCad in the
signal path (could it be considered an "active" passive device?)
NiCads have a non-linear discharge curve and internal impedance increases as
charge decreases. This changes the setpoint of your input level control and
linearity overall. NiCads also have the potential to actually reverse
polarity if they are allowed to discharge fully. You can imagine what a
positive grid bias would do to the current drawn by the tube . . . .
As pointed out by others, there is no mechanism for charging the cell.
NiCads are supposed to be charged with DC (pulsating DC is fine) but I
believe A/C is inadvisable.
What kind of artifacts, distortions, or noise are introduced by a cell in
line with a signal? This ought to be observable with a scope and a signal
source. I would certainly know what the waveforms look like before I did
anything else (although I personally would not employ one in this manner).
I wish someone would kindly explain the alleged benefit of using four grid
stopper resistors where one would do. On the face of things, it would seem
to add complexity and noise and little else.
Jon
| Quote: |
I used direct grid battery bias briefly about fifteen years ago but it
involved me in screaming match with the Magnequest Scum. That was
before I learned that whatever I did they turned into a screaming
match, and that they and their hangers-would lie about electronics to
keep the screaming match going. I can't even remember what it sounded
like.
But questions remain to be answered before I entrust valuable tubes to
the topology. It is much more useful to work them out in RAT than to
put my mind in gear and hug my findings to me, wouldn't you say? (Or
has everyone now given up on the ARRL dream of sharing information,
which motivated the founders and early members of RAT?)
Battery bias directly in the grid circuit between pot wiper and the
tube grid uses the downwards leg of a pot as the grid leak. What would
happen in case of a loose connection? In fact, what would happen if
your suggestion of a battery running out of juice came true?
Are there any other questions to be answered with this circuit?
Andre Jute
|
|
|
| Back to top |
|
|
Andre Jute
Guest
|
| Posted:
Mon Oct 17, 2005 4:43 am Post subject:
Re: Battery bias directly to grid |
|
|
Gentlemen:
Thanks for all your replies. To summarize:
There are three ways of inserting battery bias into a stage:
1. Into the cathode circuit in place of cathode resistor and bypass
cap. This is the standard way. The battery trickle-charges and battery
life pretty much equals shelf life. I've done this for years and my
observation is that batteries in this position last better than those
used in, for instance, my Apple wireless mouse and wireless keyboard,
which are fully discharged and recharged very often. Some purists claim
that the battery itself is a sound-shaping element and wish to remove
it elsewhere.
2. Into the grid leak circuit in place of the resistor there. The
battery in this position will discharge via the attenuator, probably
quite slowly. I tried this twice, both times briefly, the second time
recently. There was something indefinable, possibly subliminal, in the
sound that I didn't quite like, and that was enough for me to return
the battery to the cathode circuit. (I have a single-tube amp in which
the single tube magnifies such minor differences, if used with highly
sensitive speakers like horns.) The schematic of the KISS Ultrafi amp
published while I tried this position for the battery has now been
withdrawn and replaced with the latest schematic, which uses the
battery in the cathode circuit and has other significant improvements
(higher plate load on the 417A, stiffer power supply).
3. Between the wiper of the attenuator and the grid of the tube. In
this position battery tube will discharge, possibly most quickly of all
these possibilities. Probably best with a bypass cap, which defeats the
purposes of simplicity, low parts count, and of course of losing the
soundshaping caps. I haven't tried this and now won't bother.
It's pretty clear from the discussion that battery bias in the cathode
circuit is the only topology among these with anywhere near the
failsafe capability of autobias, plus additional sonic advantages in
tighter bass (or as low as your speakers permit your amp to go).
I'm sure you fellows will indulge me in the observation that knowing
what not to do is quite as valuable as knowing what is the right thing
to do!
Thanks all.
Andre Jute
Andre Jute wrote:
| Quote: | I've posted a circuit for discussion at:
http://members.lycos.co.uk/fiultra/Discussion%20Circuits.htm
What are the implications of this circuit?
What, theoretically, makes it better than the standard battery bias
application in the cathode circuit?
Will the batteries charge as efficiently in this position as in the
cathode circuit; what are the effects of the resistors topping and
tailing the battery set?
What are this circuit's dangers?
Are there alternatives that work better or are safer?
Andre Jute |
|
|
| Back to top |
|
|
Patrick Turner
Guest
|
| Posted:
Mon Oct 17, 2005 2:57 pm Post subject:
Re: Battery bias directly to grid |
|
|
Andre Jute wrote:
Thanks for the schematic.
What you have here is a low impedance dc voltage source in a high impedance
part of the circuit, the grid circuit. So the battery's effect is truly
negligible,
And would be when driving an output tube into grid current.
Except who would bother using a 35v battery for grid bias
to get the effect of fixed bias?
Ah, but you could use a couple of 90V batteries for direct coupling of the
driver anode to the following grid, except that it all is rather unstable
stable
because battery voltage doesn't last over many years.
But in a preamp or power amp input tube the variations over time with battery
voltage
is less of a problem.
In the 417 schematic, there is no method by which the battery can be charged.
Its just expected to sit there at 2.4V charge level, in a cathode, current
charges the battery
when the amp is on.
The battery in the 417 schematic may have some stray C to 0V but again the 20k
attenuator
determines the R to 0V of the grid circuit and C would not much affect the HF.
Why 4 x 220 ohm R in parallel?
Wouldn't a single 2k7 or 1k0 be a more appropriate grid stopper?
Patrick Turner.
| Quote: |
Patrick Turner wrote:
Andre Jute wrote:
I've posted a circuit for discussion at:
http://members.lycos.co.uk/fiultra/Discussion%20Circuits.htm
What are the implications of this circuit?
What, theoretically, makes it better than the standard battery bias
application in the cathode circuit?
Will the batteries charge as efficiently in this position as in the
cathode circuit; what are the effects of the resistors topping and
tailing the battery set?
What are this circuit's dangers?
Are there alternatives that work better or are safer?
Andre Jute
Unfortunately the images won't appear in the page in either Netscape
or IE.
I have to say I have yet to use a battery in a tube amp for biasing,
but they can work well and unlike even large value electrolytic caps
used for bypassing an Rk the battery remains a low impedance as F falls.
So the battery acts to render the amp a fixed bias amp, with possibly
better bass extension.
But for me the use of batteries isn't warranted since i can easily
achieve a bandwidth down to 5Hz without much phase shift
so any improvement due to batteries might be an illusion.
And the idea of a battery leaking junk liquid all over the internals
isn't nice...
If I could see more of the details I might be able to comment
more appropriately.
Patrick Turner. |
|
|
| Back to top |
|
|
Patrick Turner
Guest
|
| Posted:
Mon Oct 17, 2005 3:09 pm Post subject:
Re: Battery bias directly to grid |
|
|
bruce seifried wrote:
| Quote: | In article <1129479453.815805.139670@z14g2000cwz.googlegroups.com>,
"Andre Jute" <fiultra@yahoo.com> wrote:
I've posted a circuit for discussion at:
http://members.lycos.co.uk/fiultra/Discussion%20Circuits.htm
What are the implications of this circuit?
What, theoretically, makes it better than the standard battery bias
application in the cathode circuit?
Will the batteries charge as efficiently in this position as in the
cathode circuit; what are the effects of the resistors topping and
tailing the battery set?
What are this circuit's dangers?
Are there alternatives that work better or are safer?
Andre Jute
I note that the circuit you're describing is quite similar to an amp
design you have published on your website, the Type 39, Series VI "The
Tradition":
http://members.lycos.co.uk/fiultra/T39-KISS-300B-Ultrafi-crct.jpg
I assume you've actually built this circuit and tried it, so you should
know that rather then the circuit *charging* the batteries, they will
eventually be *discharged* by the input attenuator, especially in low
gain settings. You are also sending dc out the input jack to the
preceding piece of gear, which, in the case of a transformered output
circuit, will discharge the batteries at *high* gain settings.
Changing the gain setting will also tend to send a transient pulse into
your speakers.
If you're going to use this topology, you'll need to bite the bullet and
use a cap between the pot and the batteries/grid.
|
If the input grid appears slightly negative with respect to the bias supply,
then minute negative grid current can flow and very slowly discharge the
battery.
If the grid become a little positive, I flow is charging to the battery,
but this occurs when the tube is old and past its use by date.
But the transients you speak of don't occur any more than with any other
biasing method.
Patrick Turner.
|
|
| Back to top |
|
|
Patrick Turner
Guest
|
| Posted:
Mon Oct 17, 2005 3:28 pm Post subject:
Re: Battery bias directly to grid |
|
|
Jon Yaeger wrote:
| Quote: | in article 1129488747.358162.79260@g44g2000cwa.googlegroups.com, Andre Jute
at fiultra@yahoo.com wrote on 10/16/05 2:52 PM:
I constantly use battery bias in the cathode circuit. It gives you
better bass, as all kinds of fixed bias do. But John Byrns, the last
time we discussed the sound shaping of capacitors, said that the
battery is itself a soundshaping element and should be removed from the
cathode into the grid circuit.
*** I'm not an engineer, so I don't have the benefit of backing up my
opinions with good quantitative data.
Nonetheless, if simplicity is your goal I can't imagine how you could
benefit from putting something as complex and as variable as a NiCad in the
signal path (could it be considered an "active" passive device?)
|
The battery is a one ohm output impedance dc voltage source device in series
with a
circuit of meany megohms impedance, so there is utterly negligible ac current
trying to flow in the 1 ohm, so the non linear battery impedance is of
zero concern.
| Quote: |
NiCads have a non-linear discharge curve and internal impedance increases as
charge decreases. This changes the setpoint of your input level control and
linearity overall. NiCads also have the potential to actually reverse
polarity if they are allowed to discharge fully. You can imagine what a
positive grid bias would do to the current drawn by the tube . . . .
|
The non linear behaviour is of no concern.
The other concerns about battery discharge are valid, but hey,
ppl are supposed to check their batteries wherever they have them.
I would rather use a fixed bias or cathode bias.
Using some of the better quality electro caps and resistors are blameless imho.
| Quote: |
As pointed out by others, there is no mechanism for charging the cell.
NiCads are supposed to be charged with DC (pulsating DC is fine) but I
believe A/C is inadvisable.
What kind of artifacts, distortions, or noise are introduced by a cell in
line with a signal? This ought to be observable with a scope and a signal
source.
|
Not in the case of the schematic Andre has presented.
The signal current in the battery is simply too low to develop a
voltage across the low battery impedance.
There would be more chance of battery artifacts in the cathode circuit where the
impedance looking into the cathode = the anode RL / tube gain.
A little gutless wonder, the 12AX7 with a 100k RL and gain of 70
would have Rkin = 1,400 ohms, very much less than the grid input.
I think you'd find that that battery related distortions were a tiny fraction of
those
naturally occuring in the triode.
| Quote: | I would certainly know what the waveforms look like before I did
anything else (although I personally would not employ one in this manner).
I wish someone would kindly explain the alleged benefit of using four grid
stopper resistors where one would do. On the face of things, it would seem
to add complexity and noise and little else.
Jon
|
Maybe someone has shares in a resistor company. Oops, that's unkind, Vishay now
owns nearly all
the resistor making companies.
It may look like noise is better reduced because you have 4 resistors in
parallel.
But noise is determined by the value of resistance more than the type of
resistor.
10 x 550 ohm R would be no quieter than 4 x 220 ohm or 1 x 55 ohm.
220ohms has twice the noise as 55 ohms.
5,500 ohms has 10 times the noise as 55 ohms.
Its a square rule thinge here.
Patrick Turner.
| Quote: |
I used direct grid battery bias briefly about fifteen years ago but it
involved me in screaming match with the Magnequest Scum. That was
before I learned that whatever I did they turned into a screaming
match, and that they and their hangers-would lie about electronics to
keep the screaming match going. I can't even remember what it sounded
like.
But questions remain to be answered before I entrust valuable tubes to
the topology. It is much more useful to work them out in RAT than to
put my mind in gear and hug my findings to me, wouldn't you say? (Or
has everyone now given up on the ARRL dream of sharing information,
which motivated the founders and early members of RAT?)
Battery bias directly in the grid circuit between pot wiper and the
tube grid uses the downwards leg of a pot as the grid leak. What would
happen in case of a loose connection? In fact, what would happen if
your suggestion of a battery running out of juice came true?
Are there any other questions to be answered with this circuit?
Andre Jute
|
|
|
| Back to top |
|
|
Andre Jute
Guest
|
| Posted:
Mon Oct 17, 2005 5:09 pm Post subject:
Re: Battery bias directly to grid |
|
|
Patrick Turner wrote:
| Quote: | Why 4 x 220 ohm R in parallel?
Wouldn't a single 2k7 or 1k0 be a more appropriate grid stopper?
Patrick Turner.
|
It's just one of those things you have to know if you want to use the
417A/5842. There are four grid pins (thanks, Chris, that is indeed what
I meant, blushing with shame at being caught out in such an obvious
imprecision) and it is no good just clipping unused pins because the
internal bit of wire from pin to grid is also an antenna. You have to
get the resistor right up to the pin, and that is easier with four
resistors in parallel to four pins than one resistor (somehow attached
to all four the grid pins? without making extensions that will
aggravate the problem?).
Next thing, the value of the grid stopper. The most important bit of
expertise you require with the 417A is that on its grid side you want
to use only the lowest value resistance for any purpose whatsoever. A
50R gridstopper does the biz, so you want to stop there. (Heh-heh. What
do they say about puns and weakness of mind?) I have in fact
experimented with big stoppers but found no advantage. 50R 200R max is
pretty well established practice in the microwatter community. The four
by 220R I took a long time ago from Steve Bench, who has a lot of
experience with the 417A on the test bench and is not bamboozled by
audiophoolery.
Andre Jute |
|
| Back to top |
|
|
Andre Jute
Guest
|
| Posted:
Mon Oct 17, 2005 5:27 pm Post subject:
Re: Battery bias directly to grid |
|
|
Patrick Turner wrote:
| Quote: | Jon Yaeger wrote:
I wish someone would kindly explain the alleged benefit of using four grid
stopper resistors where one would do. On the face of things, it would seem
to add complexity and noise and little else.
Jon
Maybe someone has shares in a resistor company. Oops, that's unkind, Vishay now
owns nearly all
the resistor making companies.
It may look like noise is better reduced because you have 4 resistors in
parallel.
But noise is determined by the value of resistance more than the type of
resistor.
10 x 550 ohm R would be no quieter than 4 x 220 ohm or 1 x 55 ohm.
220ohms has twice the noise as 55 ohms.
5,500 ohms has 10 times the noise as 55 ohms.
Its a square rule thinge here.
Patrick Turner.
|
You two are talking about resistor noise (Johnson Noise, flicker, all
kinds of crap I have long since forgotten the names of that the
engineers will tell you about (1)). As I've explained elsewhere in this
thread, I couldn't care less about the miniscule so-called noise of the
four tiny resistors. I'm interested in keeping RF out of my amp. I
want to listen to my own music, not every badly implemented PA system
in town! Or the Martians, for that matter.
Andre Jute
(1) Though I'm not so sure all of these meet Patrick's square rule
algorithm. |
|
| Back to top |
|
|
bruce seifried
Guest
|
| Posted:
Mon Oct 17, 2005 8:26 pm Post subject:
Re: Battery bias directly to grid |
|
|
In article <4353775E.5B755A29@turneraudio.com.au>,
Patrick Turner <info@turneraudio.com.au> wrote:
| Quote: | But the transients you speak of don't occur any more than with any other
biasing method.
Patrick Turner.
|
With the Dact 20K input attenuator, and a +2.4 volt dc source connected
to its wiper, you will see something on the order of a few tens of
millivolts being generated when changing the gain setting at the low end
of the attenuator. Feeding this to the input of a dc-coupled power amp
will give you low level, audible transients. Not a deal breaker, but not
good design practice.
Switching the attenuator from its lowest gain setting to off, or vice
versa, will give you a 2.4 volt step signal into your input. I would
call this a rather large transient, and not especially good for the
health of any speaker attached.
-bruce seifried |
|
| Back to top |
|
|
Patrick Turner
Guest
|
| Posted:
Mon Oct 17, 2005 11:15 pm Post subject:
Re: Battery bias directly to grid |
|
|
Andre Jute wrote:
| Quote: | Patrick Turner wrote:
Why 4 x 220 ohm R in parallel?
Wouldn't a single 2k7 or 1k0 be a more appropriate grid stopper?
Patrick Turner.
It's just one of those things you have to know if you want to use the
417A/5842. There are four grid pins (thanks, Chris, that is indeed what
I meant, blushing with shame at being caught out in such an obvious
imprecision) and it is no good just clipping unused pins because the
internal bit of wire from pin to grid is also an antenna. You have to
get the resistor right up to the pin, and that is easier with four
resistors in parallel to four pins than one resistor (somehow attached
to all four the grid pins? without making extensions that will
aggravate the problem?).
Next thing, the value of the grid stopper. The most important bit of
expertise you require with the 417A is that on its grid side you want
to use only the lowest value resistance for any purpose whatsoever. A
50R gridstopper does the biz, so you want to stop there. (Heh-heh. What
do they say about puns and weakness of mind?) I have in fact
experimented with big stoppers but found no advantage. 50R 200R max is
pretty well established practice in the microwatter community. The four
by 220R I took a long time ago from Steve Bench, who has a lot of
experience with the 417A on the test bench and is not bamboozled by
audiophoolery.
|
Well that all seems clear.
The 417 is a hi gm triode and like other hi gM tubes I have tried these
could be RF unstable.
I have not actually tried the 417A.
But 6J6 and others tend to be a bit oscillatory.
One of the early limitations of triodes was their Miller capacitance.
Combined with long leads which are inductive enough to cause problems at
say 100MHz,
the C&L get together with the high gm and you have an oscillator when you
don't want it and didn't plan it.
Anyone using high gm j-fets such as 2sk147, 2sk369 in cascode with a triode
like a 6DJ8
or a triode strapped high gm pentode can also run into RF problems on a
phono front end.
Short leads are mandatory.
The "antenna" action of the extra grid connections may not actually be
antenna action.
oscillation is more likely to be caused by the queer behaviour at RF due to
stray C and L
at RF.
Adding resistance to the network of stray C&L is often enough to
stop phase shift occuring at RF.
Anything which can oscillate, does oscillate, unless something to stop it
is used.
Patrick Turner.
|
|
| Back to top |
|
|
Sander deWaal
Guest
|
| Posted:
Mon Oct 17, 2005 11:21 pm Post subject:
Re: Battery bias directly to grid |
|
|
Patrick Turner <info@turneraudio.com.au> said:
| Quote: | But 6J6 and others tend to be a bit oscillatory.
|
Once you get past that stage, it's a fine audio tube.
| Quote: | Anything which can oscillate, does oscillate, unless something to stop it
is used.
|
Murphy's Law # 113: oscillators won't start, amplifiers will
oscillate.
Amendment: the latter will oscillate in frequencies just beyond the
range of your test gear.
--
"Audio as a serious hobby is going down the tubes."
- Howard Ferstler, 25/4/2005 |
|
| Back to top |
|
|
Sander deWaal
Guest
|
| Posted:
Mon Oct 17, 2005 11:31 pm Post subject:
Re: Battery bias directly to grid |
|
|
"Ruud Broens" <broensr@wanadoo.nl> said:
| Quote: | : Murphy's Law # 113: oscillators won't start, amplifiers will
: oscillate.
: Amendment: the latter will oscillate in frequencies just beyond the
: range of your test gear.
:
You know your Murphy's - got an irish pub in town, too ? ;-)
|
Thank's sockpuupet Rewd for, admitting you have a drinking problem.
--
"Audio as a serious hobby is going down the tubes."
- Howard Ferstler, 25/4/2005 |
|
| Back to top |
|
|
Ruud Broens
Guest
|
| Posted:
Mon Oct 17, 2005 11:33 pm Post subject:
Re: Battery bias directly to grid |
|
|
"Sander deWaal" <nospam@wanadoo.nl> wrote in message
: >Anything which can oscillate, does oscillate, unless something to stop it
: >is used.
:
:
: Murphy's Law # 113: oscillators won't start, amplifiers will
: oscillate.
: Amendment: the latter will oscillate in frequencies just beyond the
: range of your test gear.
:
You know your Murphy's - got an irish pub in town, too ? ;-)
R.
:
: "Audio as a serious hobby is going down the tubes."
: - Howard Ferstler, 25/4/2005 |
|
| Back to top |
|
|
Patrick Turner
Guest
|
| Posted:
Mon Oct 17, 2005 11:50 pm Post subject:
Re: Battery bias directly to grid |
|
|
Andre Jute wrote:
| Quote: | Patrick Turner wrote:
Jon Yaeger wrote:
I wish someone would kindly explain the alleged benefit of using four grid
stopper resistors where one would do. On the face of things, it would seem
to add complexity and noise and little else.
Jon
Maybe someone has shares in a resistor company. Oops, that's unkind, Vishay now
owns nearly all
the resistor making companies.
It may look like noise is better reduced because you have 4 resistors in
parallel.
But noise is determined by the value of resistance more than the type of
resistor.
10 x 550 ohm R would be no quieter than 4 x 220 ohm or 1 x 55 ohm.
220ohms has twice the noise as 55 ohms.
5,500 ohms has 10 times the noise as 55 ohms.
Its a square rule thinge here.
Patrick Turner.
You two are talking about resistor noise (Johnson Noise, flicker, all
kinds of crap I have long since forgotten the names of that the
engineers will tell you about (1)). As I've explained elsewhere in this
thread, I couldn't care less about the miniscule so-called noise of the
four tiny resistors. I'm interested in keeping RF out of my amp. I
want to listen to my own music, not every badly implemented PA system
in town! Or the Martians, for that matter.
Andre Jute
(1) Though I'm not so sure all of these meet Patrick's square rule
algorithm.
|
I've mentioned the RF likelyhood in another post.
The equivalent series input noise resistance is important in an amp devoted to a
moving coil cartridge.
There is much about noise resistance in RDH4, and you'd think that a 6J6 would solve
our problems if we used one for MC input.
In practice it does not, due to microphony and a tendency to oscillate at RF.
The triode can be imagined to be a model where you have a perfect triode that
generated no noise, then having an equivalent R in series with the grid that does
produce the noise of the triode.
The input resistance = 2.5 / gm.
So a 12AX7 with gm = 0.0016A/V has ENR = 2.5 / 0.0016 = 1.56k.
This resistance will produce about 2uV of noise in the band between 50Hz and 20 kHz.
If you test a 12AX7 with a grounded input, you should find about 140uV of noise at
its
anode with about 150k as the RL because its a result of the input noise being
amplified,
plus the shot noise, which isn't high compared to input noise.
So if 2mV of signal is applied from a moving magnet cart the SNR
must be at least 1,000 : 1 = 60 dB, which is only just ok for vinyl replay.
The cart itself also generates noise, but less than the 47k input bias R of the amp.
With no cart connected, the noise from the amp is greatest
and 47k is about 30 times more than the noise resistance of the 12AX7.
But the noise will be increased by the square root of the series resistance,
so you will get a noise increase of sq.rt 30 x 2uV = 10.9 uV.
Many devices are worse in practice than a 12AX7; many transistor amps are
very poor, or were very poor, but the mechanism by which bjts develop noise is
different to a triode.
The j-fet is much better, with my amateur handbook saying that ENR = 0.7 / gm,
and when one uses a 2sk369 at 5mA, gm = 0.4A/V, so
input noise is theoretically 17.5 ohms, about 100 times less than a 12AX7,
and so the noise is about 0.2uV instad of the whopping 2 uV.
Thus the circuits using a j-fet at their input can be very quiet indeed and direct
connection to an
MC cart is possible.
There is much merit in the cascode circuitry using a fet and triode as implemeted by
Allen Wright
at http://www.vacuumstate.com
My use of the 2sk369 with a triode did give me 20 dB less noise.
And the flicker noise is absent.
Using a high gm triode such as the 417A or a 6C45pi
is supposed to reduce noise, since the ENR is lower
because it = 2.5 / gm, so when gm is higher, you are supposed to get less noise.
But after testing dozens of tubes one is very lucky to have any follow the formula
for noise.
In a revised Quad preamp I have a paralleled 6DJ8 and the gm must be
8 times the gm of half 12AX7, and one would expect noise to be
about 1/3 of the 12AX7, but in fact its 3 dB greater! damn.
So much for the theory.
I have not tested the 6C45pi I have for trial yet.
Its gm is about the same as having 3 x 6DJ8 all paralleled together.
The noise of 55 ohms is going to be much greater than the input noise of a good j-fet
amp.
Its where one tries to use a low output MC that noise becomes important.
If you have one of the Ortophons with only 100uV of rated output, you need a quiet
amp.
But if you have a Shure V15, no problem, it has a rated 5mV of signal output.
Line level stages or inputs to power amps can have much higher series Rin
because the input signal is usually above 200mV.
Patrick Turner. |
|
| Back to top |
|
|
Ruud Broens
Guest
|
| Posted:
Mon Oct 17, 2005 11:53 pm Post subject:
Re: Battery bias directly to grid |
|
|
"Sander deWaal" <nospam@wanadoo.nl> wrote in message
news:aer7l15emrfs1gaudpa0h88visfpcobarp@4ax.com...
: "Ruud Broens" <broensr@wanadoo.nl> said:
: >:
: >You know your Murphy's - got an irish pub in town, too ? ;-)
:
:
: Thank's sockpuupet Rewd for, admitting you have a drinking problem.
:
: --
it's called a drinking challenge :-)
but you know me, it's just coffee
B. |
|
| Back to top |
|
|
|
|
|
|
Profile
Log in to check your private messages
Log in