Patrick Turner wrote:
Ian Bell wrote:
Ian Iveson wrote:
Ian Bell wrote:
I have just about completed my distortion tests of 6SN7,
7N7 and 6CG7 mu 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 (Matsu****a), 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 we
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.
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.
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.
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.
* 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.
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%.
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.
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.
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
µ-follower. It also measured well.
You say they 'measured well'; care to put some actual figures to that
statement?
Cheers
ian
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...hono-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.....
cheers, and happy new year, Ian
An A Happy New Year to you to.
Cheers
ian