6SN7 et al mu follower distortion
Ian Bell
followers using tubes kindly loaned or donated by group members (Thanks
Peter and Matthew). Here is a summary of the results.
Basic set up is similar to Morgan Jones with 8mA standing current, the
bottom triode biassed to about Vg=-3V and the top pentode CCS replaced
by a normal mu follower CF set to Vg=-1.3V, with 10K between the tubes
and a 320V supply and the output via a 0.1uF into 100K load.
I tested distortion at a variety of levels at 200Hz, 2KHz and 20KHz. As
the distortion was identical at all three frequencies the majority of
tests were done only at 2KHz.
I tested 4x6CG7 (Matsushita), 2x7N7 (Sylvania) and 8x6SN7 (RCA) at
output voltages between 1 and 50V rms.
All the tubes produced remarkably similar results. The variation between
types was not greater than the variation within a type although the best
tube of all was a 6SN7. Typical THD readings were:
2V rms 0.04%
10V rms 0.2%
20V rms 0.4%
50V rms 1.0%
At 50V rms we are pretty close to grid current but since the oscillator
used has a low output impedance no grid current distortion was observed.
A subsequent test feeding the oscillator via a 120K series resistor
showed no sign of grid current up to 25V rms output (as far as the
oscillator output would go).
I was curious why Jones chose to bias his test rig at Vg=-3.4Volts but
with tubes with mu about 20 and with 20Vrms output you need to bias
several volts -ve to be sure to be away from grid current. This does not
mean that this bias point produces the lowest distortion. Simulation
showed biassing the lower triode also at about 1.3V should give even
lower distortion but of course the simulation does not simulate grid
current. So I re-biassed the bottom tube to 1.3V and check the
distortion. With a 120K in series with the oscillator, grid current
distortion began at 6Vrms output. Below that the distortion was
exceptionally good. At 5V rms is was just 0.04%. Distortion at 2V rms
was hard to measure as it expect it approaches the limit of both my
distortion test set and the oscillator distortion. I would estimate it
to be no more than 0.02% at 2Vrms.
Cheers
Ian
Andre Jute
Nice work, Ian. There's a sense in which it is a mistake to work with
the 6SN7 and its workalikes because it is such a silent tube that the
differentials will also be small and undramatic, but on the other hand
it gives you a real feeling for what a good tube responds like, and
how subtle changes can have a big impact on results, which is probably
the most important thing to learn in playing along the loadline and
transfer curves. You should also check out the tests with the 6S?7
done by Lynn Olson and his pal Matt Kamna (a measurement expert), and
by the ever-excellent Steve Bench.
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
Ian Bell
Yes, the 6SN7 and derivatives do seem to deserve their reputation. I get
the feeling later tubes tended to go for high mu, smaller (cheaper)
construction, sod the distortion and use loads of global NFB.
You should also check out the tests with the 6S?7
> done by Lynn Olson and his pal Matt Kamna (a measurement expert),
I don't think I have seen those. I'll try find them.
and
> by the ever-excellent Steve Bench.
>
Yes, I am well aware of Steve's excellent work.
Cheers
Ian
Andre Jute
http://www.nutshellhifi.com/library/FindingCG.html will take you
directly to the tests.
Worth reading around the Nutshell site. I used to know LynnO on the
Joenet in decades gone by, when it seemed everyone interested in tube
hi-fi knew everyone else. -- AJ
Ian Bell
Thanks for that Andre. I had come across it before but not bookmarked
it. It is a pity he did no test on his 6SN7 in a mu follower set up!
He also makes the usual point about NFB altering distortion spectra but
fails to back it up with measurements. Surprising since he had all that
lovely kit available to do it with.
Cheers
Ian
tubegarden
Ian Iveson
Ian Bell wrote:
Of course?
* Duncan Amplfication Generic Triode Model (Spice 3F4
Implementation)
* Copyright (C)1997-2002 Duncan Amplfication
* Unauthorised Commercial use prohibited
* Please refer to documentation at http://www.duncanamps.com
.SUBCKT NH6SN7GTB A G K
* ANODE MODEL
BLIM LI 0 V=(URAMP(V(A)-V(K))^ 1 )* 0.0037
BGG GG 0 V=V(G)-V(K)- 0
BRP1 RP1 0 V=URAMP(-V(GG)* 0.02 )
BRP2 RP2 0 V=V(RP1)-URAMP(V(RP1)-0.999)
BRPF RP 0 V=(1-V(RP2)^ 2 )+URAMP(V(GG))* 0.002
BGR GR 0 V=URAMP(V(GG))-URAMP(-(V(GG)*(1+V(GG)*
0.006167 )))
BEM EM 0 V=URAMP(V(A)-V(K)+V(GR)* 19.2642 )
BEP EP 0 V=(V(EM)^ 1.4 )*V(RP)* 0.0000189
BEL1 EL1 0 V=URAMP(V(EP))
BEL EL 0 V=V(EL1)-URAMP(V(EL1)-V(LI))
BLD LD 0 V=URAMP(V(EP)-V(LI))
BAK A K I=V(EL)
* GRID MODEL
BGF GF 0 V=(URAMP(V(G)-V(K)- 0 )^1.5)* 0.000213
BG G K I=V(GF)+V(LD)
* CAPS
CAK A K 0.0000000000007
CGK G K 0.0000000000024
CGA G A 0.0000000000039
.ENDS
> So I re-biassed the bottom tube to 1.3V and check the
> distortion. With a 120K in series with the oscillator,
> grid current distortion began at 6Vrms output. Below that
> the distortion was exceptionally good. At 5V rms is was
> just 0.04%. Distortion at 2V rms was hard to measure as it
> expect it approaches the limit of both my distortion test
> set and the oscillator distortion. I would estimate it to
> be no more than 0.02% at 2Vrms.
Thanks, Ian. Nice to see actual measurements of real
circuits.
I wonder if you measured at any intermediate operating
points, in between the two you mention? Just wondering if
the change in distortion is a trend or a blip.
Are you able to vary the HT voltage, to explore other parts
of the safe operating area? Some idea of a trend would be
good there too.
How does simulated total distortion compare with your real
measurements?
cheers, and happy new year, Ian
Ian Bell
Ah, well, perhaps not. I have tried very many models and all but one
fail to accurately model the normal operating region where the grid is
negative wrt the cathode. The one that does model the normal operating
region accurately (probably because it is based on The Audio Designers
Tube Register curves which are the measured curves of actual tubes) does
not include a grid model. Maybe I can make a hybrid and nick the grid
model from the Duncan Amps models.
Yes I did, I just gave a sample in the summary. I measured several tubes
in 1V increments from 1V to 10V rms and then in 5V increments
thereafter. As expected the distortion is very close to being directly
proportional to signal level as expected until the onset of grid current
when it rises steeply. Only the lower level measurements are suspect as
they are approximately double the distortion of the oscillator itself.
> Are you able to vary the HT voltage, to explore other parts
> of the safe operating area? Some idea of a trend would be
> good there too.
>
Not at the moment. 320V is as high as I can get with HT and that's just
high enough for the circuit used. I suspect if I could raise the HT I
could get even lower figures. However I was regularly achieving 0.4% THD
at 20Vrms which is at -48dB relative to the fundamental. For most tubes,
Morgan Jones only achieved -50dB second harmonic with a pentode CCS
plate load at 19.5V rms so I do not expect there is much improvement to
be gained by raising the HT. Personally, I was surprised I got
distortion figures so close to Jones' with such a simple circuit which
speaks volumes for the inherent linearity of the 6SN7 family.
At present both tube halves sit comfortably within the SOAR. The bottom
tube has a plate/cathode voltage around 140V which at 8mA is less than
1.2W dissipation and the top one has about 100V across it which is 800mW
both of which are well within the SOAR of the 6CG7. I would be more
concerned about exceeding heater/cathode voltages than SOAR. At present
the heaters are raised to +75V (6CG7 has max +Ve dc of 100) adn the top
cathode is at about 220V (6CG7 has max -ve dc of 200V which would allow
cathode to go to 275V) so there's no much margin on either one.
> How does simulated total distortion compare with your real
> measurements?
>
As I mentioned above, one model reproduces the distortion levels very
accurately except for grid current. All the others I have tried are way out.
> cheers, and happy new year, Ian
>
>
An A Happy New Year to you to.
Cheers
ian
Patrick Turner
If you stick to building real circuits and measuring them you will
always be able to tweak it up to get the widest V swing, symetrical
clipping, and lowest THD widest BW and lowest Rout as all or some of
these things might matter to you.
The 10k between each 6sn7 triode section is just high enough so that the
bottom tube sees a load of approximately the top tube open loop gain x
10k without output any load connected. In your case, if you need 1Vrms
to exist between grid and cathode of top triode for a cathode output
voltage of say 18V, then OLG = 18/1 = 18.
Then the bottom tube sees a load = 18 x 10k, or 180k which is about
9Ra, and which is approaching the load for minimum THD. The triode anode
load is a CCS, and were you to use a CCS instead of the 10k, then THD
would maybe drop 6dB, but because your figures are fairly good so far,
the complexity if an added CCS isn't neccessary.
I like to operate the top tube with a fixed bias via 1M from a bias
supply, then cap couple the top grid to the bottom anode.
The grid bias voltage you should have in your case should **ALWAYS** be
well above thet -1.3V you've been quoting. Bias for to and bottom
triodes should be close to about -5V which implies that if you have Ia =
5mA, then there's 50V across the 10k, and Ea = 135V for both triodes. A
quick draw of a load line on anode curves with a load slope of 180k will
soon tell you what sort of swing you'll get and how little the swing
will be if the Ia is higher, and Eg lower.
The other thing is that loading the top tube cathode does slightly make
the circuit work as a SRPP, but the effect is minor.
Try measuring the THD with and without a load, and with a real world
range of loads lower than the one you tried.
To get truer THD results, THD of your oscillator always should be 1/10
of the minimum THD you will expect to be able to measure.
Therefore say you wish to measure down to 0.01%, then the oscillator THD
should be 0.001%.
If the oscillator THD is high, and is 2H, then it either adds to the THD
of the triodes which is predominantly 2H, or it cancels, so you won't
ever get an accurate idea of THD as the level is tested at lower and
lower voltages where you may be using the amplifier.
I got mightily fed up with poor measurements and wrong measurements when
I started so I built a simplr Wien bridge oscillator for 1kHz using
opamps and a "grain of wheat" light bulb, and trimmed it to operate only
at 1kHz with about 20 cycles of F adjustment with both fine and course
adjust pots. It makes about 0.2% THD at 1kHz, and once the F is set, the
F is very stable. Some oscillators have an anoyingly variable F which
means you'll have difficulty nulling the F out when you test the output
of the DUT.
I then build a discrete component amp with BJTs to raise the 1V of the
oscillator to make 4 voltage ranges, 0-1, 0-2, 0-4, 0-8V, so that on the
lowest range, only small fraction of the oscillator is amplified to make
1V of output. The voltage amp is set up to act as a bandpass filter with
a very simple R&C NFB network which gives gain = 8x at 1kHz, but much
less at 2kHz and 500Hz and further away from the 1kHz. The result is
that the THD at the amp output in the 1V range is 0.001%. I have an
output level potentionmeter of 5k which produces less THD than the
preceeding gear.
I also made a bridged T tunable null filter, and another 1.6Khz to 10kHz
bandpass amp to amplify the output from the bridged T to see really low
levels of THD. I also build a high input impedance buffer to accept
signals from the types of circuit you are doing so the load of the
nulling filter didn't affect what i was trying to measure. I also fitted
switchable hum filters.
I think I spent about 6 weeks building and rebuilding that oscillator
and tweaking it. But it taught me all about analog measuring, and
building low noise circuits, and about just how difficult it is the get
right.
The result is that I don't have to estimate what THD might be while I
measure. What I measure IS THE REAL DEAL within 10%, ie, if I measure
0.01%, it indicates THD is between 0.009% and 0.011%, and the
measurement is good enough.
I don't have a schematic online of my seld designed test gear but I am
sure anyone here could build their own THD measuring gear after reading
old Wireless World mags or searching around like I did 14 years ago. And
BTW, its easier to get good results buidling such gear by just using
opamps rather than attempting anything with tubes. I also made a passive
LC filter using air cored L and C carefully set up with two LC parallel
resonant sections fed by 4.7k R, so that although the insertion loss is
12dB, the output signal has THD another 20dB lower than the oscillator
produces, so THD < 0.001%. But wherever you have coils in THD measuring
gear, you cannot use iron cores because you'll get iron core distortion
much larger than what you are trying to measure. Even an iron box to
shield the coils causes THD. So you have to keep the LC filter away from
iron, and well away from any mains wiring. The oscillator and following
amps need to be fed by a remote PS with good rail filtering.
> >
> > Thanks, Ian. Nice to see actual measurements of real
> > circuits.
> >
> > I wonder if you measured at any intermediate operating
> > points, in between the two you mention? Just wondering if
> > the change in distortion is a trend or a blip.
> >
>
> Yes I did, I just gave a sample in the summary. I measured several tubes
> in 1V increments from 1V to 10V rms and then in 5V increments
> thereafter. As expected the distortion is very close to being directly
> proportional to signal level as expected until the onset of grid current
> when it rises steeply. Only the lower level measurements are suspect as
> they are approximately double the distortion of the oscillator itself.
!
>
> > Are you able to vary the HT voltage, to explore other parts
> > of the safe operating area? Some idea of a trend would be
> > good there too.
> >
>
> Not at the moment. 320V is as high as I can get with HT and that's just
> high enough for the circuit used. I suspect if I could raise the HT I
> could get even lower figures. However I was regularly achieving 0.4% THD
> at 20Vrms which is at -48dB relative to the fundamental. For most tubes,
> Morgan Jones only achieved -50dB second harmonic with a pentode CCS
> plate load at 19.5V rms so I do not expect there is much improvement to
> be gained by raising the HT. Personally, I was surprised I got
> distortion figures so close to Jones' with such a simple circuit which
> speaks volumes for the inherent linearity of the 6SN7 family.
0.4% at 20Vrms isn't too bad, but you should get about 1/2 that, or no
more than 0.1% at 10V.
This is about the limit for SE triode signal amps without any loop NFB.
>
> At present both tube halves sit comfortably within the SOAR. The bottom
> tube has a plate/cathode voltage around 140V which at 8mA is less than
> 1.2W dissipation and the top one has about 100V across it which is 800mW
> both of which are well within the SOAR of the 6CG7.
This explains your 1.3V bias. I never use a single 1/2 6SN7 at 8 mA
because I doubt there is any sonic benefit and you should get at least
the same low THD with 4mA and the lower bias. When the anode load
becomes high, it doesn't matter if the load line is down in the more
curved regions of Ra because the load line is close to horizontal and
intersects the Ra curves at about the same points in their curvatures.
>I would be more
> concerned about exceeding heater/cathode voltages than SOAR. At present
> the heaters are raised to +75V (6CG7 has max +Ve dc of 100) adn the top
> cathode is at about 220V (6CG7 has max -ve dc of 200V which would allow
> cathode to go to 275V) so there's no much margin on either one.
You can get over the heater bias problem by using two single triodes,
say 6J5, or 2 x 6SN7 tubes, and bias bottom heaters at 0V and the top
heaters at close to whatever the cathode Ek is.
>
> > How does simulated total distortion compare with your real
> > measurements?
> >
>
> As I mentioned above, one model reproduces the distortion levels very
> accurately except for grid current. All the others I have tried are way out.
Hmm, I have never wasted time modelling or simulating things beyond what
I can achieve in my head or with a ruler and set of Ra curves. When you
actually build it, its the real thing.
I did try to make an output stage for a phono amp using a 12AT7 for gain
and loaded with a CCS, then a direct coupled 6CG7 as a normal output
cathode follower. It measured well. But I preferred the sound of the
12AT7 when used as a bottom tube and 6CG7 at the top tube as a
ต-follower. It also measured well. And is wasted less PSU power. See the
"Rocket" schematic about 1/4 the way down the page at
http://www.turneraudio.com.au/preamp-rocket-phono-2005.html
Patrick Turner.
PS, I find it difficult to be happier at Xmas or NY than at any other
time of the year.
But try not to do anything I would not enjoy.....
Andre Jute
>The other thing is that loading the top tube cathode does slightly make
>the circuit work as a SRPP, but the effect is minor.
Actually, any circuit which works even partially like an SRPP is a
good circuit, especially if the advantage of the SRPP it mimics is a
low impedance. I'm not always so sure that the mu-follower is truly
worth the additional complication over an SRPP for anyone except the
purist who can, as a consolation for the extra expense and
complication, mutter to himself over and over, "Yes, but an SRPP isn't
a real current source and a mu-follower is an exemplary current
source... It's true, it's true!" An SRPP with a single extra resistor
to arrange a voltage lift can in most instances be arranged to work
for practical purposes as well, and probably more reliably, than a mu-
follower.
Okay, I don't want to sound like a cost accountant, but someone must
be realistic.
Quite a bit of other good stuff, some of it at least arguable on
grounds of taste, some of it indubitably right, snipped in the
interest of bandwidth.
Andre Jute
The tubes tend to make people believe in a god, and SS leads them to
the
devil. -- Patrick Turner
Ian Bell
Tweaking may be necessary but it is not sufficient IMHO. It is effective
only with that particular set of components. If you build another
another one just the same it won't perform just the same and you will be
need to tweak it again to achieve similar performance. To me that's
poor design practice. Better to understand what is going one so a
circuit with repeatable performance can be built without tweaking.
> The 10k between each 6sn7 triode section is just high enough so that the
> bottom tube sees a load of approximately the top tube open loop gain x
> 10k without output any load connected. In your case, if you need 1Vrms
> to exist between grid and cathode of top triode for a cathode output
> voltage of say 18V, then OLG = 18/1 = 18.
>
> Then the bottom tube sees a load = 18 x 10k, or 180k which is about
> 9Ra, and which is approaching the load for minimum THD. The triode anode
> load is a CCS, and were you to use a CCS instead of the 10k, then THD
> would maybe drop 6dB, but because your figures are fairly good so far,
> the complexity if an added CCS isn't neccessary.
>
Jones achieved only 2dB better figures with a CCS so I would be doubtful
that a 6dB improvement could be had.
> I like to operate the top tube with a fixed bias via 1M from a bias
> supply, then cap couple the top grid to the bottom anode.
>
> The grid bias voltage you should have in your case should **ALWAYS** be
> well above thet -1.3V you've been quoting.
For the bottom tube probably yes but it does not seem to affect the top
triode performance.
Bias for top and bottom
> triodes should be close to about -5V
Why 5V ??
which implies that if you have Ia =
> 5mA, then there's 50V across the 10k, and Ea = 135V for both triodes. A
> quick draw of a load line on anode curves with a load slope of 180k will
> soon tell you what sort of swing you'll get and how little the swing
> will be if the Ia is higher, and Eg lower.
>
I don't think so. The curves are much more bunched together when Ia is
5mA. My tests at lower values of Ia gave higher distortion levels.
> The other thing is that loading the top tube cathode does slightly make
> the circuit work as a SRPP, but the effect is minor.
>
> Try measuring the THD with and without a load, and with a real world
> range of loads lower than the one you tried.
>
I have but I did not mention them in the summary. Distortion at 20V rms
2KHz into various loads was:
100K load 0.41%
25K load 0.46%
9.9K load 0.52%
By measuring loaded and unloaded output levels, the output impedance was
calculated as 830 ohms.
Agreed.
> If the oscillator THD is high, and is 2H, then it either adds to the THD
> of the triodes which is predominantly 2H, or it cancels, so you won't
> ever get an accurate idea of THD as the level is tested at lower and
> lower voltages where you may be using the amplifier.
>
Agreed.
Er, half of 0.4% is 0.2% or did I miss something?
> This is about the limit for SE triode signal amps without any loop NFB.
>
>> At present both tube halves sit comfortably within the SOAR. The bottom
>> tube has a plate/cathode voltage around 140V which at 8mA is less than
>> 1.2W dissipation and the top one has about 100V across it which is 800mW
>> both of which are well within the SOAR of the 6CG7.
>
> This explains your 1.3V bias. I never use a single 1/2 6SN7 at 8 mA
> because I doubt there is any sonic benefit and you should get at least
> the same low THD with 4mA and the lower bias.
I disagree. My 8mA bias is with the bottom tube at just over -3V so grid
current is not an issue. THD is definitely measurably lower at 8mA Ia
than at lower currents.
When the anode load
> becomes high, it doesn't matter if the load line is down in the more
> curved regions of Ra because the load line is close to horizontal and
> intersects the Ra curves at about the same points in their curvatures.
>
>
>> I would be more
>> concerned about exceeding heater/cathode voltages than SOAR. At present
>> the heaters are raised to +75V (6CG7 has max +Ve dc of 100) adn the top
>> cathode is at about 220V (6CG7 has max -ve dc of 200V which would allow
>> cathode to go to 275V) so there's no much margin on either one.
>
> You can get over the heater bias problem by using two single triodes,
> say 6J5, or 2 x 6SN7 tubes, and bias bottom heaters at 0V and the top
> heaters at close to whatever the cathode Ek is.
Yes, I plan to use them in pairs and had considered having both bottom
triodes in one 6SN7 and the top pair in another with separate heater
supplies.
>>> How does simulated total distortion compare with your real
>>> measurements?
>>>
>> As I mentioned above, one model reproduces the distortion levels very
>> accurately except for grid current. All the others I have tried are way out.
>
> Hmm, I have never wasted time modelling or simulating things beyond what
> I can achieve in my head or with a ruler and set of Ra curves. When you
> actually build it, its the real thing.
>
Unfortunately Ra curves are not a very good indicator of what real tubes
actually do. Manufacturers data sheets are a long way from what the
tubes actually available today actually do. The most reliable set of
curves I have found are those created from measurements of the tubes of
today as found in the Audio Designers Tube Register. The spice models
based on these same sets of curves give simulated results very close to
those I measured. Other spice models I have found to give very
inaccurate results. Equally, manufactures data sheets are only suitable
for making the crudest of estimations of THD in low level preamplifiers.
> I did try to make an output stage for a phono amp using a 12AT7 for gain
> and loaded with a CCS, then a direct coupled 6CG7 as a normal output
> cathode follower. It measured well. But I preferred the sound of the
> 12AT7 when used as a bottom tube and 6CG7 at the top tube as a
> ľ-follower. It also measured well.
You say they 'measured well'; care to put some actual figures to that
statement?
Cheers
ian
Ian Bell
> On Jan 2, 11:46 am, Patrick Turner <i...@turneraudio.com.au> wrote:
>
>> The other thing is that loading the top tube cathode does slightly make
>> the circuit work as a SRPP, but the effect is minor.
>
> Actually, any circuit which works even partially like an SRPP is a
> good circuit, especially if the advantage of the SRPP it mimics is a
> low impedance. I'm not always so sure that the mu-follower is truly
> worth the additional complication over an SRPP
If you want low distortion then the mu follower is much better than the
SRPP; definitely worth the extra complexity.
for anyone except the
> purist who can, as a consolation for the extra expense and
> complication, mutter to himself over and over, "Yes, but an SRPP isn't
> a real current source and a mu-follower is an exemplary current
> source... It's true, it's true!" An SRPP with a single extra resistor
> to arrange a voltage lift can in most instances be arranged to work
> for practical purposes as well, and probably more reliably, than a mu-
> follower.
OK, then show me an SRPP design with a 6SN7 that does significantly
better than 0.4% at 20Vrms out into 10K at 2KHz.
>
> Okay, I don't want to sound like a cost accountant, but someone must
> be realistic.
>
Indeed, and if you want to save a few pennies at the expense of a whole
lot of distortion then SRPP is the way to go.
> Quite a bit of other good stuff, some of it at least arguable on
> grounds of taste, some of it indubitably right, snipped in the
> interest of bandwidth.
>
LOL
Cheers
Ian
Andre Jute
You're missing my point, Ian, probably unsurprisingly. You're
interested in pre-amps and therefore need to extract the maximum of
silence as well as a big gain. I use CD and integrated two stage amps
(mostly); I start with 2Vrms of clean signal. Virtually any 6SN7
circuit that is competently built is silent enough for me, though in
fact I use high-mu 417A/5842 almost exclusively these days. And I have
many other trade-offs if I want addtional silence, as I usually do. I
can trade tube longevity by running the tube at very high voltage and
current. I can and do move the quiescent operating point well away
from any possibility of grid current. I can and do flatten the
loadline along which the signal swings by very high loads so that a
real accountant would tear out his hair at the "waste" -- like
building a 10 litre V16 engine for a car and then choking it back with
the silencers to a whisper and a miserable 200bhp when it is capable
of 600bhp. Everyone else makes a 300B give 8W; mine deliver 3.8W...
Andre Jute
Refinement beyond price
Ian Iveson
>>>
>>> I was curious why Jones chose to bias his test rig at
>>> Vg=-3.4Volts but with tubes with mu about 20 and with
>>> 20Vrms output you need to bias several volts -ve to be
>>> sure to be away from grid current. This does not mean
>>> that this bias point produces the lowest distortion.
>>> Simulation showed biassing the lower triode also at
>>> about 1.3V should give even lower distortion but of
>>> course the simulation does not simulate grid current.
>>
>> Of course?
>>
>
> Ah, well, perhaps not. I have tried very many models and
> all but one fail to accurately model the normal operating
> region where the grid is negative wrt the cathode. The one
> that does model the normal operating region accurately
> (probably because it is based on The Audio Designers Tube
> Register curves which are the measured curves of actual
> tubes) does not include a grid model. Maybe I can make a
> hybrid and nick the grid model from the Duncan Amps
> models.
That would probably be OK but watch out for "LD" that links
the grid model to the anode model. You may need to rename
it, or add it's derivation into your anode model.
I say probably because in reality grid current has a
relationship to anode current, probably, because I guess it
decreases perveance (less free electrons available so anode
current should fall such that total cathode current remains
the same, all other things being equal). But it should be a
negligible error where grid current is small.
Right, I got that bit, and I can see why you expected what
you got. What I was wondering, though, was whether you had
tried a range of bias currents in between the two you
mention, to establish the relationship between standing
current and distortion for the same signal. I got the
impression that you were asserting that lower standing
current is related to lower distortion, but it's not very
clear...perhaps I'm missing something. The question is
slightly ambiguous, because of variation in mu and therefore
in gain for different standing currents, but that shouldn't
be so great as to prevent reasonable comparison. There is
also the problem of ensuring that the bottom anode remains
set at half the supply voltage. Two points aren't enough to
establish a trend. I ask partly because it seemed to be the
question that you set out with: why did MJ bias at such a
level?
I realise that standing current is also related to maximum
signal capability, but still wonder if I want to amplify a
small signal, where a low standing current would be
sufficient to stay far enough clear of grid current, will I
always get greater distortion by raising the standing
current? Or have I got it the wrong way round?
Whichever, I'm still wondering about sweet spots.
This comes as a pretty big suprise. Do you mean that you
have measured simulated distortion and compared it to your
real measurements? This makes a sweep of distortion level v
standing current even more interesting.
Could you post a copy of your model, please? AFAIK, the
generic DM model underlying the 6SN7 I posted is the most
accurate and sophisticated available, with one possible
exception that you probably couldn't use on your simulator.
Considering the DM triode comes with a grid model that had
been omitted from previous, inferior triode models, and that
DM is the originator of nearly all available models and I
would guess *all* in common use, I surmise that *all* of the
improvement in the model you have used is due to the more
appropriate data used to generate it.
It is therefore not unlikely that fitting the most recent DM
model to that data would result in a model more accurate
than the one you have, particularly in the BRH quarter of
the SOAR, IIRC.
Which is why I would appreciate a copy of the model text. If
it is similar to the DM generic model, it may be possible to
transfer common parameters, and leave any extra ones as they
are. If not, then the data the model was fitted to would be
useful, although a complete refit is a tedious business.
Ian
Ian Bell
> Ian Bell wrote
>>>> I was curious why Jones chose to bias his test rig at
>>>> Vg=-3.4Volts but with tubes with mu about 20 and with
>>>> 20Vrms output you need to bias several volts -ve to be
>>>> sure to be away from grid current. This does not mean
>>>> that this bias point produces the lowest distortion.
>>>> Simulation showed biassing the lower triode also at
>>>> about 1.3V should give even lower distortion but of
>>>> course the simulation does not simulate grid current.
>>> Of course?
>>>
>> Ah, well, perhaps not. I have tried very many models and
>> all but one fail to accurately model the normal operating
>> region where the grid is negative wrt the cathode. The one
>> that does model the normal operating region accurately
>> (probably because it is based on The Audio Designers Tube
>> Register curves which are the measured curves of actual
>> tubes) does not include a grid model. Maybe I can make a
>> hybrid and nick the grid model from the Duncan Amps
>> models.
>
> That would probably be OK but watch out for "LD" that links
> the grid model to the anode model. You may need to rename
> it, or add it's derivation into your anode model.
>
Thanks for the tip. I'll watch out for that.
> I say probably because in reality grid current has a
> relationship to anode current, probably, because I guess it
> decreases perveance (less free electrons available so anode
> current should fall such that total cathode current remains
> the same, all other things being equal). But it should be a
> negligible error where grid current is small.
>
Yes, and re-reading RDH4 on grid current there seems to be several
contributory factors and I doubt the spice model includes them all.
What I did do was:
1. Try a lot of tubes at 8mA and -3.4V (bottom triode) and 8mA 1.3V (top
triode) and got consistent results between tube types and also got
distortion ruoghly proportional to output level.
2. Tried a few tubes at 8mA and 1.3V both top and bottom triodes and got
lower distortion at lower levels but grid current at higher levels. This
presumably means that a lower bias voltage means lower distortion at a
given current though it is by no means conclusive.
3,. Following Patrick's comment I did a quick test yesterday with both
triodes biassed at 5mA and -5V (actually turned out to be 4.55mA and
-4.55V). I got exactly the same distortion figures at 2V and 20V rms as
in the original experiment (1 above). Frankly I expected the distortion
to go up at the lower current but it did not - this is however one test
on one tube only.
> I realise that standing current is also related to maximum
> signal capability, but still wonder if I want to amplify a
> small signal, where a low standing current would be
> sufficient to stay far enough clear of grid current, will I
> always get greater distortion by raising the standing
> current? Or have I got it the wrong way round?
I am not sure but I would have expected distortion to decrease as
standing current increases simply because you move away from the region
where the curves bunch. And if you look at the spacing between curves
they seem to me to be less evenly spaced as the bias voltage gets more
negative which does explain why test #2 gave lower distortion
>
> Whichever, I'm still wondering about sweet spots.
>
Me too. Test 3 seems to show that nearly halving the standing current
makes no difference to distortion and leads to a bias voltage a long way
from grid current problems.
> This comes as a pretty big surprise. Do you mean that you
> have measured simulated distortion and compared it to your
> real measurements?
Yes, the spice model based on the The Audio Designers Tube Register
curves gives THD results that are extremely close to the measurements.
This makes a sweep of distortion level v
> standing current even more interesting.
>
Yes, that might be easier to achieve in a simulation than on the bench.
> Could you post a copy of your model, please? AFAIK, the
> generic DM model underlying the 6SN7 I posted is the most
> accurate and sophisticated available, with one possible
> exception that you probably couldn't use on your simulator.
>
> Considering the DM triode comes with a grid model that had
> been omitted from previous, inferior triode models, and that
> DM is the originator of nearly all available models and I
> would guess *all* in common use, I surmise that *all* of the
> improvement in the model you have used is due to the more
> appropriate data used to generate it.
>
That would be my conclusion too.
> It is therefore not unlikely that fitting the most recent DM
> model to that data would result in a model more accurate
> than the one you have, particularly in the BRH quarter of
> the SOAR, IIRC.
>
Possibly. The model I have apparently used quite a sophisticated curve
fitting model so I am not sure why the DM model should be inherently better.
> Which is why I would appreciate a copy of the model text. If
> it is similar to the DM generic model, it may be possible to
> transfer common parameters, and leave any extra ones as they
> are. If not, then the data the model was fitted to would be
> useful, although a complete refit is a tedious business.
No problem and I just realised I told a big fib. The model is apparently
taken from the GE data sheet not the The Audio Designers Tube Register
curves as I stated earlier - I wonder where I got that from?? - must
check that out. Anyway, here is the 6SN7 portion of the file:
*6SN7 LTSpice model from GE 6SN7 datasheet
.subckt 6sn7 P G K
Bp P K
I=(0.02003791851m)*uramp(V(P,K)*ln(1.0+(-0.07740549711)+exp((4.618036737)+(4.618036737)*((20.85288965)+(-110.4389272m)*V(G,K))*V(G,K)/sqrt((28.13407639)**2+(V(P,K)-(7.118597372))**2)))/(4.618036737))**(1.380047579)
Cgp G P 4.0pF
Cgk G K 2.6pF
Cpk P K 0.7pF
.ends 6sn7
Note the preamble to the set of models is as follows:
*Generated by Joel Tunnah using Curve Captor v0.9.1
*UPDATED 02/07/06
*Contents:
*6SN7
*12AX7
*12AU7
*12AT7
*ECC88
*6C4Pi
Cheers
Ian
>
> Ian
>
>
Iain Churches
"Ian Bell" <ruffr...@yahoo.com> wrote in message
news:gje4b4$ps2$1...@localhost.localdomain...
Good work Ian.
The THD figs you obtained match precisely those I have
measured with 6CG7 in mu follower topology. I have
built several such preamps.
I found the most consistent tubes to be Siemens and
Westinghouse both with 0.02% at 1kHz and 2Vrms
output for a 125mV input. (24dB gain)
You might like to take one step further now, and compare
the distortion profile of tubes that have similar THD.
Cheers
Iain
Ian Bell
So far I have tried a couple of 7N7 kindly loaned by Matt, and a bunch
of 6SN7s given to me by Peter. Their THD figures are remarkably
consistent. I have also tried four Matsushita 6CG7 tubes I got from
Colomor and these are as good as the N7 tubes and quieter too - I think
I'll be buying a few more of these. I am keeping an eye out for other
makes of 6CG7 but so far they are rather few and far between and so far
not a Siemens to be seen amongst them.
> You might like to take one step further now, and compare
> the distortion profile of tubes that have similar THD.
>
> Cheers
> Iain
>
Thanks for the kind words. Yes, I would like to delve deeper and to that
end I have just purchased an HP 3581A Wave Analyser which should let me
take a closer look at the distortion spectra.
Cheers
Ian
Patrick Turner
Andre Jute wrote:
>
> On Jan 2, 11:46 am, Patrick Turner <i...@turneraudio.com.au> wrote:
>
> >The other thing is that loading the top tube cathode does slightly make
> >the circuit work as a SRPP, but the effect is minor.
>
> Actually, any circuit which works even partially like an SRPP is a
> good circuit, especially if the advantage of the SRPP it mimics is a
> low impedance. I'm not always so sure that the mu-follower is truly
> worth the additional complication over an SRPP for anyone except the
> purist who can, as a consolation for the extra expense and
> complication, mutter to himself over and over, "Yes, but an SRPP isn't
> a real current source and a mu-follower is an exemplary current
> source... It's true, it's true!" An SRPP with a single extra resistor
> to arrange a voltage lift can in most instances be arranged to work
> for practical purposes as well, and probably more reliably, than a mu-
> follower.
The pure SRPP works OK for most folks if they like simplicity, but let's
consider some facts.
For SRPP, where Rk of the top and bottom triode are equal, Rout is less
than the Ra of one triode, but several times that of a pure CF. If you
move to µ-follower where the Rk between top and bottom triode equals at
least half the Ra for one triode then the Rout plummets to be much
closer to that of a CF. The penalty paid for the µ-foll is an extra Rk
of at least Ra/2, and a bias resistor, and a capacitor to couple the
bottom triode anode to the top triode grid. Maybe $2.00.
With SRPP, one can tailor the Rk between the two tubes and the load to
nearly equalize the ac anode current, but there will always be slightly
more Iac in one tube than in the other when a load is connected so full
series PP operation with complete cancelation of 2H never really
happens. And instead, after mucking about with loads and Rk, usually you
find the tubes not very well loaded, and the load each sees is not many
times Ra, and you get considerable 3H instead of the more benign 2H.
With µ-foll, you simply never have to worry about making the damn
circuit work to produce minimal 2H cancelation. Its you have series PP
where the top tube acts as a "bootstrapped follower", ie, quasi CF with
low THD, and the bottom tube has a very high ac load many times Ra and
many more times Ra than the SRPP could ever offer. Thus the bottom tube
develops more gain and lower THD/IMD than most other arrangements and
you get the saving of not having to waste dc power and generate
distortion in cathode load resistors. Let's not try to escape the
miniscule cost of a few R&C parts to get a µ-foll to work better than
its primitive cousin the SRPP. In µ-foll there is mimimized load ac flow
in the bottom gain triode, since it is buffered by the CF above it.
I think the µ-foll sounds better.
I have also used triode gain stages with CCS dc supply, and a following
R load many times Ra and sonic results are the best. I really don't like
resistances used for dc conveyance to a triode in critical circuits
because they create unecessary distortion.
One of my classic examples is this good sound preamp at
http://www.turneraudio.com.au/line-preamp-2003.html
There's a 1/2 12AU7 driving a 50k gain pot. But the anode is a CCS. If
there was a dc load R the value would have been about 33k. With the 50k
pot the load would have been about 20k, and only 2Ra at 4mAdc. But I
have RL at 5Ra approx, and THD is low. It'd even be lower if the pot had
been 100k, which may have been even better, as I have in my later Nemo
line stage circuit at
http://www.turneraudio.com.au/preamp-nemo-line+psu-2005.html
> Okay, I don't want to sound like a cost accountant, but someone must
> be realistic.
Sometimes it costs more to employ a bean counter than using a few more
beans :-)
Patrick Turner.
Patrick Turner
Say you set up an SE 300B amp which can make 8W into 5 ohms, while
dissipating no more than 30W at idle. Then you may find that maximum
clipping power will decline with loads below 5 ohms and above 5 ohms,
but it is an ideal way to set up a 300B to deal with speakers that are
nominally 8 ohms, so tht when indeed the load is 8 ohms the PO might be
much less than 8 ohms but fidelity and DF is higher.
In fact, the 300B set up as I suggest will indeed put up with speakers
nominally 4 ohms if the levels expected are low because one great
feature of triodes is that they are fairly load tolerant. But where the
host of the party is a triode, don't abuse his hospitality by inviting
too many guests who will draw excessively from the limited barrel of
amps available.
If very few guests turn up and the load is deemed to be 16 ohms then
decorum and convivialty will prevail and without the distortions of bad
behaviours.
The secret to partying well on one 300B for each channel is to have
sensitive speakers!
The more sensitive the speaker is, the less important is its impedance,
and the sweeter the sound if you can make a sensitive speaker work
properly.
Patrick Turner.
Ian Iveson
Thanks very much, Ian, especially for the model. Sorry for
the delay: I've been fully occupied with the crisis in
Palestine.
When time permits I'll compare the two models and try to
produce a valid composite. In passing, that seems like an
awful lot of decimal places!
Cheers, Ian
in message news:gjqlfg$ug0$1...@localhost.localdomain...