Phase Splitter
John Brodie
Does a 12at7 based phase splitter need a neg. screen bias? Also, should the
two drive signals be exactly the same, nicely sinusoidal (with sine input)
equaly magnitude, and 180 out of phase.... I think I may have found the
problem with my amp. For some reason, I've got exactly the same voltage on
the grid and cat. of my phase splitter...
any help would be great..
JB
Grover Gardner
are you using?
--
Grover Gardner
gro...@postoffice.att.net
Grover Gardner
great for SE 845 or 211 filaments. These are the "analogue", not switching
type. I have similar 5V supplies which I have used with KR and Svetlana
tubes, with very good results. They have 115/230 primaries, are mounted on
aluminum frames measuring approx. 7 x 5 x 3.5 inches. They can be mounted
under a large chassis or enclosed as an outboard supply. Brand new in box,
with specs, these are very sturdy, beefy supplies. I'm asking $50 for the
pair, and buyer pays shipping. Please reply via private e-mail. Thanks!
--
Grover Gardner
gro...@postoffice.att.net
CERREM
>
>Does a 12at7 based phase splitter need a neg. screen bias? Also, should the
>two drive signals be exactly the same, nicely sinusoidal (with sine input)
>equaly magnitude, and 180 out of phase.... I think I may have found the
>problem with my amp. For some reason, I've got exactly the same voltage on
>the grid and cat. of my phase splitter...
>
>any help would be great..
>
>JB
I need more info about this circuit...give me voltages and resistor values..
Like I said once before but no one seems to believe me...
When you bias a Cathode follower or a Cathodyne phase splitter you bias it NOT
from the grid to cathode...You bias it from plate to grid....
You will get roughly the same voltage at grid and at cathode when you measure
with respect to ground.... BUT if you were to put a volt meter between cathode
and grid you will notice the bias voltage.....
Another misconception regarding this type of phase splitter is that because the
cathode section is lower output Z thus creating an imbalance at high
frequencies since the plate output is higher output Z.... The cathode output
WILL follow what ever the plate does....BECAUSE the same AC current across the
plate resistor is the same AC current across the cathode resistor since they
are in SERIES..... So they will be balanced.... BUT when there is an
imbalance at high frequencies..it is not the phase splitter causing it...it is
usually the bypass cap to ground for the plate circuit that becomes a problem
at those frequencies...while the cathode circuit does not need to be bypassed
since it is already connected to ground on the other side of the cathode
resistor..
Let me know what amp this is as well voltages and values preceeding stage as
well..
CHEERS
CM
The gain of this phase splitter is unity .....
Proteus
>Another misconception regarding this type of phase splitter is that
>because the cathode section is lower output Z thus creating an imbalance
>at high frequencies since the plate output is higher output Z....
>The cathode output WILL follow what ever the plate does....BECAUSE the
>same AC current across the plate resistor is the same AC current across
>the cathode resistor since they are in SERIES..... So they will be
>balanced....
That would be true if ALL the *signal* current flowed through the plate
and cathode resistors. In real life, however, significant signal current
usually flows in the output-stage grid resistors. Because this current is
sourced from different, finite impedances (the plate and the cathode nodes
of the phase splitter), more signal voltage is dropped internal to the
higher impedance (plate) source, and consequently less signal voltage is
available across that output grid resistor than across the grid resistor
connected to the phase-splitter cathode. The imbalance is not usually
greater than 10-20% at mid frequencies, because the grid resistors are
usually respectably higher than the plate node impedance. However, at
higher frequencies, the declining grid impedance of the output tubes due
to input capacitance requires significant signal current to flow in the
output grid circuits. This causes increasing imbalance with frequency
(as well as increasing phase error).
Proteus (the poster formerly known as Anonymous)
John Byrns
<pro...@redneck.efga.org> wrote:
There has been much complaining in rec.audio.tubes, about Andre and how he
is misleading the poor innocents in the news group. It hardly seems
possible that Ande could mislead anyone, given the close scrutiny his
every word gets. But in this post by the anonymous Proteus, we find a
very egregious case of misinformation. For whatever reason, Proteus is
considered an "expert" by most in rec.audio.tubes, and that high status
places on Proteus the responsibility to be very accurate in his
pronouncements, because they are accepted as fact by most readers here.
In the past Proteus has used his status to pass off what are little more
than opinions, as if they were facts, but here we catch Proteus with his
proverbial pants down, repeating once again the Urban Legend that the
split load phase inverter is inherently unbalanced.
Let's look at the argument Proteus makes. He madly waves his hands
attempting to justify his claim that the split load phase inverter is
unbalanced in real life, with an attempt at an argument based on linear
circuits, and yet he doesn't even mention the effects of the grid to
cathode, and grid to anode capacitance of the phase inverter tube.
The fact is that under the assumption of a totally linear circuit, with
matched loads in the cathode and anode circuits, and ignoring the grid to
cathode, and grid to anode capacitance's, as Proteus did, the split load
inverter is perfectly balanced, as it is easy to show with simple math.
The grid resistors matter not a whit, as long as they are balanced, along
with the anode and cathode resistors, coupling capacitors, stray wiring
capacitance, miller capacitance, and etc., and as long as we consider all
these elements to be linear.
Proteus does not make his argument with math, only hand waving, and he is
wrong, as anyone with a smattering of circuit theory can prove for
himself. Now while Proteus talked about the real world, he didn't discuss
the real world factors which can make the split load phase inverter less
than perfect, but still pretty darn good.
In the real world, the effects of the grid to anode, and grid to cathode
capacitance of the phase inverter tube, do cause imbalance, but the
imbalance is generally very small, occurs above the audio band, and in any
case is less than most other relatively simple phase splitters.
If the grids of the tubes following the split load phase inverter are
driven into grid current, then the two phase splitter outputs will each
experience different nonlinear distortion effects, which is presumably not
good.
One might also imagine that the nonlinear miller capacitance of a triode,
varying with gm, as the current varies over the audio cycle, could
contribute to unbalance. I have not investigated this issue, but there
are several reasons to suspect that this is not really a problem, although
I would be happy to discuss it with anyone who is interested.
The bottom line, Proteus notwithstanding, is that when the grid
capacitance's of the split load phase inverter tube are ignored, and the
rest of the circuit, including the output stage grid circuits, is
considered to be a linear circuit, then if the components are perfectly
balanced, the two will also be perfectly balanced. In the real world, the
phase splitter grid capacitance's cause some imbalance at very high
frequencies, but much less than most other common phase splitters. And
the split load inverter does get into trouble when the following stage is
driven into grid current, which is a good reason to use a buffer stage
between the phase inverter outputs, and the output tubes, as Williamson
did.
As Henry would say, this is a challenge to Proteus, to present some
rudimentary mathematical justification for his claim.
Donning my Nomex suit.
Regards,
John Byrns
Andre Jute
I dunno nuffin bout no phase splitters! I wasn't even there when they
dunnit, Sergeant.
Andre
Transformers are the answer. What was the question again?
--
Andre Jute
an...@indigo.ie COMMUNICATION JUTE
--we support pages for music lovers, writers and audiophiles at
http://indigo.ie/~andre/ComJuteF1.html
Dave Davenport
"... when loaded with substantially equal impedances and driven at a
relatively low level, the high-frequency roll-off of this inverter is
about the same as that of a cathode follower; and the roll-off is the
same for both plate and cathode terminals."
An Analysis of the Split-Load Phase Inverter
George Ellis Jones, Jr. Audio Engineering, December, 1951
"... as long as the two load impedances are maintained equal at all
frequencies of interest, no concern need be felt about the differences
in apparent source impedance."
Notes on the Cathodyne Phase-Splitter
Albert Preisman, Audio, April, 1960
If you like math, there is plenty of it in both articles. Preisman
cites an earlier paper where:
"... The writer employed this type of phase-splitter back in the days
when cathode-ray oscilloscopes had a bandwidth of 3 mc at most, and
could maintain a flat response with such a circuit out to 8 mc."
A wide-range video amplifier for a cathode ray osciloscope
A. Preisman, RCA Review, April, 1939
Proteus
>here we catch Proteus with his
>proverbial pants down, repeating once again the Urban Legend that the
>split load phase inverter is inherently unbalanced.
Hey John, guess what? You're right about my recent post.
I'm responding because I'm awed and amused by the religious fervor
with which you pounced. No need to shout, John -- I'm perfectly
willing to admit and retract when I'm wrong. I had an incorrect
mesh diagram in mind, which I've never seriously questioned. It's
no excuse for posting without thinking seriously, but then I'm not
perfect. Besides, it would have deprived you of the opportunity to
unload your little rant, which seems very important to you.
>He madly waves his hands
>and yet he doesn't even mention the effects of the grid to cathode,
>and grid to anode capacitance of the phase inverter tube.
I assumed it was clear that my (mistaken) claim was that the splitter
is unbalanced, even to a first approximation. In fact, you seem to
have understood that. So excuse me if I didn't cloud the waters with
higher-order effects that were irrelevant to my (again, mistaken) claim.
>Proteus does not make his argument with math, only hand waving
The math point is actually one that deserves discussing. There are
a handful of engineering-math adepts in the group, but it seems to me
that the real value of the group is to a much larger readership who
aren't particularly interested in the math but who would like a
better intuitive understanding of the circuits. There's no reason
not to post the math when the thread warrants it, but in my mind the
most valuable posts are the ones that make what the math shows to be the
case understandable to someone who doesn't care (or know) about the math.
This is the reason, for example, that I have not yet posted what I have
been working on for quite a while -- a useful intuitive explanation of
phase stability and compensation. I haven't figured out how to make it
intuitive without the math. Those who know the math don't need me to
explain it, and what I can say about it at this point wouldn't do someone
who doesn't know the math any good. I'm confident that something can be
said that fits the bill, but I haven't thought of it yet. If I may say so,
John, your efforts would be much better directed towards something like
this than gleefully pointing out in a thousand words or more that someone
else made a mistake.
Proteus
I quoted CM re: split-load phase inverters:
>>Another misconception regarding this type of phase splitter is that
>>because the cathode section is lower output Z thus creating an imbalance
>>at high frequencies since the plate output is higher output Z....
>>The cathode output WILL follow what ever the plate does....BECAUSE the
>>same AC current across the plate resistor is the same AC current across
>>the cathode resistor since they are in SERIES..... So they will be
>>balanced....
Then I put my foot in my mouth, as John Byrns pointed out on another
thread.
The reason that what CM said is right and what I said is wrong is exactly
the one that CM gave -- the entire cathode load is in series with the
entire plate load, and as long as they are identical the same signal
current flows in both. Consequently, the same signal voltage is dropped
across both loads.
My one remaining quibble is that the plate circuit really follows the
cathode circuit, not vice versa.
My apologies for posting without thinking this time.
Mats Wiklander
--
Mats Wiklander
The Society Of High Tension Audio
mats.wi...@nospam.cml.com (remove nospam for replies)
Henry Pasternack
John Byrns wrote in message ...
>There has been much complaining in rec.audio.tubes, about Andre and
>we catch Proteus with his proverbial pants down, repeating once again
>the Urban Legend that the split load phase inverter is inherently
unbalanced.
Proteus is forgetting that the finite plate resistance of the tube
causes the impedance looking into the cathode to rise when there
is a resistance in series with the plate. Therefore, both the plate
and cathode nodes have equal output impedance in this circuit.
This is easier to visualize when you use the current source model
of the vacuum tube.
Poor Proteus! You do him a real injustice in your enthusiasm
to prop up Andre.
-Henry
Andy Moss
> Proteus is forgetting that the finite plate resistance of the tube
> causes the impedance looking into the cathode to rise when there
> is a resistance in series with the plate. Therefore, both the plate
> and cathode nodes have equal output impedance in this circuit.
> This is easier to visualize when you use the current source model
> of the vacuum tube.
Henry, how does Miller capacitance come into this? Can we necessarily
claim there is none due to the gain being nil?
Henry Pasternack
Proteus wrote in message <1999060906312...@anon.efga.org>...
>... [I]t seems to me that the real value of the group is to a
>much larger readership who aren't particularly interested in
>the math but who would like a better intuitive understanding
>of the circuits... [T]he most valuable posts are the ones that
>make what the math shows to be the case understandable
>to someone who doesn't care (or know) about the math.
>
>This is the reason, for example, that I have not yet posted
>what I have been working on for quite a while -- a useful
>intuitive explanation of phase stability and compensation.
>I haven't figured out how to make it intuitive without the math.
Hey, I understand. Sometimes an insight seems so clear
and obvious that anyone could understand it if only he would
sit still and listen for a minute or two. And it's so exciting to
have this insight, all you want to do is share it with others.
I think the problem is that these insights depend on a rich
fabric of facts, relationships, and experience that are so
ingrained one tends to forget about them. Lacking this
background, the non-specialist reader struggles with minor
details and misses or misinterprets major points. Confusion
leads to debate and debate leads to acrimony.
It's a pity you choose to remain anonymous, Anonymous.
I would be interested in getting to know you better.
-Henry
Paul Gustafson
I'm no math genius but I would be very interested in your thoughts on
this subject. As far as I can tell this has quite a bit to do with the
high (and low) frequency stability of an amp (that uses nfb). I build
mainly guitar amps using local feedback at the output (sparingly) and
degenerative feedback at the voltage amp stages, but I noticed that I
get quite a bit of phase shift through the tone control stack. These
kinds of things I wish I knew more about. Any info I can assimilate
about this stuff is good for me.
Thanks for your time,
Paul G.
Proteus wrote:
> This is the reason, for example, that I have not yet posted what I have
> been working on for quite a while -- a useful intuitive explanation of
> phase stability and compensation. I haven't figured out how to make it
> intuitive without the math. Those who know the math don't need me to
> explain it, and what I can say about it at this point wouldn't do someone
> who doesn't know the math any good. I'm confident that something can be
> said that fits the bill, but I haven't thought of it yet. If I may say so,
> John, your efforts would be much better directed towards something like
> this than gleefully pointing out in a thousand words or more that someone
> else made a mistake.
>
Henry Pasternack
Andy Moss wrote in message <375E7E...@ibm.net>...
>Henry, how does Miller capacitance come into this? Can we necessarily
>claim there is none due to the gain being nil?
The gain from grid to plate is on the order of -0.9 or so, depending on
circuit specifics. So you expect to see a corresponding increase in the
apparent grid-plate capacitance due to the Miller effect. The gain from
grid to cathode is about +0.9, so the effect of grid-cathode capacitance
is actually diminished by feedback. This is off the top of my head; I
didn't get much sleep last night.
-Henry
ELECTAUDIO
source so at the high freq will be unequal. the plate side will rolloff first.
Jack
John Byrns
elect...@aol.com (ELECTAUDIO) wrote:
Hi Jack,
Can you provide any justification for the claim that "the plate side will
rolloff first"? In real life the split load inverter goes out further
than most other simple inverters before becoming unbalanced. The supposed
unbalance of the split load inverter is just an Urban legend that has
never been justified by anyone anywhere that I have ever been able to
find. The differing impedances of the palte and cathode do not cause
imbalance.
Regards,
John Byrns
Ken Gilbert
> The differing impedances of the palte and cathode do not cause
> imbalance.
John, I know that you know of course that a different output Z _MUST_
cause a different frequency response into a given reactive load. If
the plate and cathode have different impedances, which they do, then
they absolutely MUST have different capabilities as drivers of
capacitive loads. Therefore the imbalance is a real effect.
BUT
I do wholeheartedly agree with you that this effect is probably nowhere
near audible, so for OUR intents and purposes the cathodyne is
"balanced."
--
Ken Gilbert
Tube Guitar Amp Design/Repair Technician
The Guitarist's Choice http://www.tgcguitar.com
http://www.geocities.com/SunsetStrip/Garage/5701
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
Wim Kotterman
Ken Gilbert wrote in message <7k5lo1$mjp$1...@nnrp1.deja.com>...
>
>If the plate and cathode have different impedances, which they do, then
>they absolutely MUST have different capabilities as drivers of
>capacitive loads. Therefore the imbalance is a real effect.
>
I recommend to read the discussion in the section "Letters" of EW&WW of
January/March 1996, in which Morgan Jones shows the photographs of a
oscilloscope depicting output waveforms of a cathodyne phase-splitter.
Loaded with 100 pF in both legs it delivers (at 500 kHz !!!!) square waves
that still look like square waves and are identical (after inverting one, of
course, and judged by the eye). Output impedances are equal at about 600
Ohms conform theory (ref. the posting of Dave Davenport of Wednesday). The
tube was an ECC83.
In another posting I wrote that I accept the facts but I don't understand
them. Who does?
Greetings,
Wim Kotterman.
John Byrns
<ride...@ride.ri.net> wrote:
> > The differing impedances of the palte and cathode do not cause
> > imbalance.
>
> John, I know that you know of course that a different output Z _MUST_
> cause a different frequency response into a given reactive load. If
> the plate and cathode have different impedances, which they do, then
> they absolutely MUST have different capabilities as drivers of
> capacitive loads. Therefore the imbalance is a real effect.
Hi Ken,
No, with respect to the split load phase inverter, I do not "know of
course that a different output Z _MUST_ cause a different frequency
response into a given reactive load"! That belief is the fallacy that
gives this old Urban Legend life. I am tempted to discuss the theories an
old associate of mine had about the different "levels of consciousness"
you go through as you begin to gain understanding of something, but I will
spare you all that.
Since you claim that it is the different output Z's that must cause an
output imbalance, it should be acceptable to stick with a simple model,
where the loads on the plate and cathode are equal and linear, there are
no capacitance's associated with the grid of the phase splitter tube, and
the tube is modeled as a simple fixed linear transconductance in parallel
with a simple fixed linear resistance. If you accept that, then you must
come to the conclusion that even though the output Z's are different at
the plate and cathode, the signals there are perfectly balanced. This
follows directly from the fact that the same identical current must flow
in both the plate and cathode circuits, and nonlinear effects can't cause
trouble because we have defined all the circuit elements as being linear.
Can you explain how the AC currents in the two loads can be different,
assuming balanced linear loads?
Now in real life, there are issues to deal with, which will cause slight
imbalances, such as the effects of the grid to plate and grid to cathode
capacitance's, when the grid is driven by a finite source impedance.
Also, in the case of a triode output stage, the Miller capacitance at the
grid will vary with the gm of the tube, which will vary with the signal on
the grid. This nonlinear Miller capacitance could theoretically cause a
large signal high frequency imbalance, although I haven't investigated
this to determine if it is likely to amount to much in practice.
There may be other factors to consider, and I don't claim to fully
understand all the issues, but two things seem pretty clear. 1.) The
split load phase inverter is not unbalanced at high frequencies as a
result of the differing output impedance's of the cathode and the plate.
2.) The split load phase inverter has the best balance, at all
frequencies, of any simple phase inverter, that is those using 3 or fewer
active tube sections.
Regards,
John Byrns
Kurt Strain
>>they absolutely MUST have different capabilities as drivers of
>>capacitive loads. Therefore the imbalance is a real effect.
>>
>
>January/March 1996, in which Morgan Jones shows the photographs of a
>oscilloscope depicting output waveforms of a cathodyne phase-splitter.
>Loaded with 100 pF in both legs it delivers (at 500 kHz !!!!) square waves
>that still look like square waves and are identical (after inverting one,
of
>course, and judged by the eye). Output impedances are equal at about 600
>Ohms conform theory (ref. the posting of Dave Davenport of Wednesday). The
>tube was an ECC83.
>In another posting I wrote that I accept the facts but I don't understand
>them. Who does?
>
I did a little research to see if I can confirm this. Sorry, I couldn't. I
think
John Byrns is more correct on this issue, and I wrote something that
wasn't a complete picture in another thread, so I want to rectify that
situation.
Going back to the trusty RDH4, page 330 (be patient before saying
there's more known since then):
"The effective output resistance is different for the two output channels,
since P operates with current feedback and K with voltage feedback.
Channel P: rp' = (u - 1)Rk where Rk = Rl (30)
(from equation 25a, Sect. 1)
Channel K: rp' = (rp + Rl)/(u + 1)
(31)
(from equation 3)
but this does not affect the balance at either low or high frequencies
when the total impedance of channel P is equal to that of channel K.
The same signal plate current which flows through one impedance Zp
also flows through the other impedance Zk, and if Zp = Zk then the
two output voltages are equal. The ratio of the output voltages is Rl : Rk
at low frequencies and Zp : Zk at any high frequency. The capacitive
component of Zp is the sum of the input capacitance of the following
stage, wiring and stray capacitances, and the output capacitance of P
channel (eqn. 33), and similarly with Zk (eqn. 34). The output capacitances
of the phase splitter normally differ by only 1 or 2 uuF [ pF. ], having
a negligible effect on the balance at 10000 c/s - see also pages 522-523.
...
Output capacitance
P channel: Co' = 2Cpk + Cgp(1 + 1/A') (33)
K channel: Co' = 2Cpk + Cgk(1/A' -1) + Chk (34)
"
So there is not a complete balance, but in fact there turns out to be good
balance even into the highs. How exactly is that happening with different
output impedances? Well, I'm going to make a hypothesis on that.
Like I said before, the plate current is sourced by Vk/Rk in the constant
current source analysis. For AC, call it Vk/Zk. Due to capacitance at
the K channel, Zk is falling off at the same rate that Zp is falling off.
So
the current source is upping its current Vk/Zk as Zk falls, just enough
to offset the Zp drop. They track one another. So even though the
output impedance of the plate side is higher than the cathode side,
the current source gets bigger to the plate source that compensates
for the gain loss. A neat feature, something we forgot to see happening.
So I believe that the frequency response tracks well for the
plate and cathode end, but that does not happen because the output
impedances are identical, there's instead a little boosting effect going
on that makes it appear to have the same output impedances into
similar capacitive loads.
Does this make sense?
Kurt
Andre Jute
Of course, the lateral thinker's paranoid solution to getting rid of the
imbalance, which the splitter is just waiting to inject (the minute you
switch the test equipment off--of course!), is to lose the splitter
altogether. Go SE*, young man!
Andre
*S(outh)E(ast)
Wim Kotterman
Kurt Strain wrote in message <7k77nu$jto$1...@news.ap.net>...
>
>So I believe that the frequency response tracks well for the
>plate and cathode end, but that does not happen because the output
>impedances are identical, there's instead a little boosting effect going
>on that makes it appear to have the same output impedances into
>similar capacitive loads.
>
>Does this make sense?
>
Although I asserted quite firmly otherwise in my postings, I'm afraid it
does:-).
Yesterday evening after posting I still had the feeling that that low
impedance at the plate still had something magical to me. Normally such
magic turns out to have a quite reasonible explanation. Then it appeared to
me that Jones did not measure output impedance but just showed that plate
output is following cathode output up to high frequencies under identical
load conditions. And that the current through the tube can't be constant
with increasing frequency but should increase too.
The same mistake, but now by me, with the references given by Dave
Davenport, they speak of high-frequency roll-off being the same. They didn't
mention identical output impedances, that's what I made of it.
So, nothing magical, a current source on top of a cathode follower is to me
again an accurate description.
Could you please expand a little on that four letter word ( RDH4 :-) as I
don't know it.
Greetings,
Wim Kotterman.
Henry Pasternack
> Proteus is forgetting that the finite plate resistance of the tube
>causes the impedance looking into the cathode to rise when there
>is a resistance in series with the plate. Therefore, both the plate
>and cathode nodes have equal output impedance in this circuit.
This is totally wrong. Ignore it.
Kurt Strain wrote in message <7k77nu$jto$1...@news.ap.net>...
>So there is not a complete balance, but in fact there turns out to be good
>balance even into the highs. How exactly is that happening with different
>output impedances? ...I believe that the frequency response tracks well
>for the plate and cathode end, but that does not happen because the
>output impedances are identical, there's instead a little boosting effect
>going on that makes it appear to have the same output impedances into
>similar capacitive loads.
>
>Does this make sense?
Here is the answer to the puzzle. It doesn't have to do with
capacitance.
If you vary the load resistors individually, you will find that the
impedance
at the plate is relatively high and the impedance at the cathode is
relatively
low. If you vary both resistors at the same time and by the same amount,
both cathode and plate nodes will appear to have low output impedances.
This would seem to be a paradox.
The reason is that the value of the cathode resistor is reflected in the
impedance looking into the plate. And the value of the plate resistor is
reflected in the impedance looking into the cathode. But going from
cathode to plate, any change in resistance is multiplied by a factor of mu,
whereas going from plate to cathode the resistance change is divided by
a factor of mu. I think it's actually a factor of (1 + mu), but this is
about
the same as mu when mu >> 10.
So, if the load resistors decrease, there is a substantial lowering of
the impedance looking into the plate due to the reduction in the value
of the cathode resistor. This largely offsets any drop in the plate signal
amplitude due to heavier loading. The same is true looking into the
cathode, except that the change in resistance is much smaller because
impedances are divided, not multiplied, going from plate to cathode. As
well, the cathode is already a low-impedance node and isn't bothered
much by changes in load resistance, as long as Rk >> rk.
Another way of looking at this is that making the cathode resistor
smaller causes the plate current to increase. This mostly cancels out
the variation in plate signal level due to the change in the plate resistor.
On the other hand, changing the plate resistor doesn't have much effect
on the plate current, nor consequently on the cathode signal amplitude.
Finally, when making measurements it's important to vary only one
parameter at a time. It's not valid to measure plate output impedance
while changing the value of the cathode resistor, or vice-versa. This
is the origin of the output impedance paradox. If you change the load
resistors individually, the expected results are obtained.
By the way, in my Williamson-style amps, there is a substantial
high-frequency imbalance, but it happens well above the audio band.
The signal on the cathode, as I recall, goes out to 1MHz or more,
whereas the signal on the plate rolls off much earlier. I don't remember
the exact numbers.
-Henry
Kurt Strain
: Could you please expand a little on that four letter word ( RDH4 :-) as I
: don't know it.
Sorry, that stands for Radiotron Designer's Handbook, 4th revision.
It's a pretty trusty and comprehensive source of information on tube
circuits.
Kurt
Ron Selberg
> high-frequency imbalance, but it happens well above the audio band.
> The signal on the cathode, as I recall, goes out to 1MHz or more,
> whereas the signal on the plate rolls off much earlier. I don't remember
> the exact numbers.
I've measured it. It's a bit tricky, you have to measure both
simultaniously, with the same kind of scope probe on each. In order to
get the real results, you have to use very low capacitance probes. I
too forget the numbers, but in reality , both the plate and the
cathode's response went well in excess of 300khz, and not until after
that did the balance begin to change.
It's a good splitter.
I think it's also important to remember that more than a few amps used
it to drive the output tubes as well, and in this case it doesn't work
so hot. It's way better to have a stage between it and the output tubes.
Ron
Henry Pasternack
Andre Jute wrote in message
<1dthdcz.kqh...@ts01-005.bantry.indigo.ie>...
>Of course, the lateral thinker's paranoid solution to getting rid of the
>imbalance, which the splitter is just waiting to inject (the minute you
>switch the test equipment off--of course!), is to lose the splitter
>altogether. Go SE*, young man!
It may be lateral thinking, and it may be a paranoid solution, but
it's also completely wrong. Single-ended amplifiers are, by definition,
totally unbalanced. If imbalance is a bad thing, then single-ended is
the archetype of badness.
I wouldn't expect a Jute to understand the subtleties, however.
-Henry
John Byrns
<nob...@nobody.com> wrote:
> By the way, in my Williamson-style amps, there is a substantial
> high-frequency imbalance, but it happens well above the audio band.
> The signal on the cathode, as I recall, goes out to 1MHz or more,
> whereas the signal on the plate rolls off much earlier. I don't remember
> the exact numbers.
Hi Henry,
I haven't read through your explanation in enough detail to be able to say
if I think it holds water or not, but that is beside the point. Even if
this explanation is correct, it is way to complicated, and obscure. The
problem is really much simpler to resolve, and without all the complicated
hand waving. All one need do is notice that with the exception of any
small current flowing in the grid circuit, the current that flows into the
plate, must all flow out the cathode. In other words, the signal currents
in the plate and cathode circuits must be identical, and if the load
impedance's are identical, the two output phases will also be identical
mirror images of each other.
I can think of only three reasonable ways to evaluate the balance of the
split load phase inverter, and considering the output impedance's of the
plate and cathode is not one of them. The first, discussed above, is to
note that the signal currents in the plate and cathode circuits must be
identical, because there is nowhere else for the current to go. The
second way is to connect identical pieces of measuring apparatus to both
the plate and cathode, and measure the response at both outputs. The
third way is to write the loop equations for the circuit, and
mathematically solve for the frequency response at both the plate and
cathode. I mentioned these approaches in an earlier posting, a month or
so ago.
All this complicated talk of the variations in the plate, and cathode,
source impedance's is way too confusing, at least for me.
Regards,
John Byrns
John Byrns
<nob...@nobody.com> wrote:
Hi Henry,
I was tempted to post a very similar retort about a month ago, in response
to another poster was lamenting over the "problem" of phase inverter
imbalance, and cited this as one of the advantages of SE operation, but I
was able to resist the temptation.
I do think you are being a little too absolutist, in saying the idea is
"completely" wrong, it may be mostly wrong, but an unqualified
"completely" seems to be overstating the case.
Seems like you are ignoring the subtleties yourself, with this sort sledge
hammer approach to the subject.
Regards,
John Byrns
Eric Vannerson
this circuit are here, but no one has quite put them all together.
Consider this:
At high frequencies the impedance connected to the cathode decreases, but
the signal voltage at the cathode is not much affected because of the large
amount of negative feedback (All that happens is that the signal from grid
to cathode increases slightly).
Since, as was pointed out, the currents in plate and cathode currents are
essentially the same, and the impedance in the plate circuit is the same as
that at the cathode, the plate voltage doesn't change much with frequency
either.
It works for me.
John Byrns wrote in message ...
Andre Jute
> In article <7k8nm1$1gmc$1...@rtpnews.raleigh.ibm.com>, "Henry Pasternack"
> <nob...@nobody.com> wrote:
>
> > Andre Jute wrote in message
> > <1dthdcz.kqh...@ts01-005.bantry.indigo.ie>...
> > >Of course, the lateral thinker's paranoid solution to getting rid of the
> > >imbalance, which the splitter is just waiting to inject (the minute you
> > >switch the test equipment off--of course!), is to lose the splitter
> > >altogether. Go SE*, young man!
> >
> >
> > It may be lateral thinking, and it may be a paranoid solution, but
> > it's also completely wrong. Single-ended amplifiers are, by definition,
> > totally unbalanced. If imbalance is a bad thing, then single-ended is
> > the archetype of badness.
> >
> > I wouldn't expect a Jute to understand the subtleties, however.
Literary allusion:
>Go SE*, young man!
In my Bartlett, nicely inscribed by the editor, it is 554:3
"Go west, young man." --- John Babsone Lane Soule (1815-1891), article
in Terra Haute (Indiana) Express, 1851
(...not Horace Greeley, as the uncouth and careless often claim--Greeley
himself republished Soule's article to show from whence he received his
inspiration)
With the pun explained so simply that even a Stanfard engineering
graduate should get it:
>*S(outh)E(ast)
which is a point of the compass, finely balanced rather than either
fully balanced or unbalanced, totally without moral judgement except by
those who from bile confuse the compass rose with floristry.
> Hi Henry,
>
> I was tempted to post a very similar retort about a month ago, in response
> to another poster was lamenting over the "problem" of phase inverter
> imbalance, and cited this as one of the advantages of SE operation, but I
> was able to resist the temptation.
>
> I do think you are being a little too absolutist, in saying the idea is
> "completely" wrong, it may be mostly wrong, but an unqualified
> "completely" seems to be overstating the case.
If it weren't a joke, I'd be tempted to make a case, John, with my
tongue making a huge bulge in my cheek, something along the lines of: A
known total lack of balance is preferable to an uncontrollable
frequency-dependent lack of balance.
But the wrong people will take me seriously--and I just started work on
a totally balanced linestage...
>
> Seems like you are ignoring the subtleties yourself, with this sort sledge
> hammer approach to the subject.
>
>
> Regards,
>
> John Byrns
Poor fellow musta bought his sense of humour the same place he bought
his brains---in the fire sale of the Dim Sim Memory Company.*
Andre
Transformers are the answer. What was the question again?
*Motto: "Our DIMMs are D-d-d-dimmer!"
Henry Pasternack
John Byrns wrote in message ...
>if I think it holds water or not, but that is beside the point. Even if
>this explanation is correct, it is way to complicated, and obscure.
Before dismissing my comments out of hand, you really ought to
take the time to read and understand them. If you did, you'd realize
I was answering a different question. Namely, how can it be that
the outputs of the split-load phase splitter have different impedances,
yet the signal amplitudes track one another as the load resistances
vary equally? The answer to this question does indeed benefit from
looking at the reflection of the cathode resistance in the plate circuit.
In fact, the notion of the tube as impedance transformer is elegant
and fundamental. An interesting observation that follows is that for
most practical circuit values, the plate output impedance is dominated
by the value of the external cathode resistor. It then can be shown
in terms of a simple voltage divider why the gain from grid to plate is
necessarily close to unity. In fact, I could see spending a couple of
days in a classroom exploring the implications of material you so
casually brush aside as "hand-waving".
I post thoughtful and original comments in the hope of getting
meaningful, even appreciative, responses. You'll forgive me, I trust,
if I don't thank you for your reply.
-Henry
Henry Pasternack
1) The phony Jute killfile:
On June 8, Andre Jute wrote about me:
>Note that you're not bothering me. You're permanently
>in my killfile. I shan't flatter you again by noticing you
>even at second-hand in someone else's post. You're
>an unperson.
But now:
>With the pun explained so simply that even a Stanfard
>engineering graduate should get it...
2) The literary intellectual snow-job:
>In my Bartlett, nicely inscribed by the editor, it is 554:3
>
>"Go west, young man." --- John Babsone Lane Soule
>(1815-1891), article in Terra Haute (Indiana) Express,
>1851
>
>(...not Horace Greeley, as the uncouth and careless
>often claim--Greeley himself republished Soule's article
>to show from whence he received his inspiration)
3) The bogus "It was a joke" defense:
>If it weren't a joke, I'd be tempted to make a case,
>John, with my tongue making a huge bulge in my
>cheek, something along the lines of...
4) The "Change what I said and hope nobody notices" ploy:
The original bogus comment, which even John Byrns called
"mostly wrong":
>Of course, the lateral thinker's paranoid solution
>to getting rid of the imbalance, which the splitter
>is just waiting to inject... is to lose the splitter
>altogether.
And the new, incorrect formulation:
>A known total lack of balance is preferable to an
>uncontrollable frequency-dependent lack of balance.
5) The "Imaginary project" deception:
>But the wrong people will take me seriously--and I
>just started work on a totally balanced linestage...
Congratulations, Andre! Your recent posting scores 100% on
the Jute Bogosity Scale.
-Henry
Ron Bales
Sorry Henry, but you score far too generously - we've missed a couple
of standard tactics:
The "Oh, of course we learned about that back when we were in short
pants, but called it by another name. The name you use for it is so
unintuitive I couldn't recognize it." This is notably lacking.
AND
He didn't drag in a single artist, poet, F1 team owner or other
luminary.
Otherwise you have correctly identified one of the most completely
Juted pieces ever published. It probably sets a record for its
brevity and thoroughness.
ROn
Henry Pasternack
Ron Bales wrote in message <37690dc4...@news.gte.net>...
>Sorry Henry, but you score far too generously...
Andre is an idiot. He's such an idiot, he doesn't know he's an
idiot. How do you deal with an idiot who's too stupid to follow
any argument proving what an idiot he is? I guess you have to
ignore him. Phooie. What a waste of time this usenet is.
-Henry
Kurt Strain
: Wim Kotterman wrote in message <3766b...@pandora.research.kpn.com>...
: >
: >Ken Gilbert wrote in message <7k5lo1$mjp$1...@nnrp1.deja.com>...
: >>
: >>If the plate and cathode have different impedances, which they do, then
: >>they absolutely MUST have different capabilities as drivers of
: >>capacitive loads. Therefore the imbalance is a real effect.
: >>
: >
: >I recommend to read the discussion in the section "Letters" of EW&WW of
: >January/March 1996, in which Morgan Jones shows the photographs of a
: >oscilloscope depicting output waveforms of a cathodyne phase-splitter.
: >Loaded with 100 pF in both legs it delivers (at 500 kHz !!!!) square waves
: >that still look like square waves and are identical (after inverting one,
: of
: >course, and judged by the eye).
I just looked at Morgan Jones' book "Valve Amplifiers" and he says on page
214 concerning this phase splitter:
"Various observers have commented that the output resistances are not
equal, and that when the stage drives a real load, its balance will therfore
be compromised, although none of them has offered any analysis to show
what these resistances might be, or what remedial action should be taken."
The thing is, Langford-Smith already had it in his Radiotron Designer's
Handbook, as I referred to it. And it is clear that no "remedial action"
should be taken. So here's the interesting part - page 233 shows an
amp with a "build-out resistor" on the cathode side of this phase splitter
supposedly to address the "problem" of impedance imbalance. He intentionally
adds impedance on the cathode side to make it close to the plate side.
The text on page 232 says "That this circuit is quite similar to the GEC 912
Plus demonstrates that there is little new under the sun, but the design
rationale is new, and to the author's knowledge, the cathode build-out
resistor in the phase splitter is unique."
Well, yeah, it's a unique feature to add something that screws up a
properly balanced phase splitter. Now the criteria well known by the
time he published this, that it remains balanced so long as the loads are
identical, is destroyed. He made an error thinking he was correcting for
some fault that's not there. This book was first published in 1995,
and so it looks like he learned of his mistake later as you indicate above.
So you might want to make a note of it in your copy of "Valve Amplifiers"
that the "build-out resistor" is something to keep out - in fact it will
cause an imbalance. He never noticed I'm sure, it adds a bit more 2nd
order distortion, something he applies feedback to correct for anyway.
Just thought I'd pass that along. Mistakes like this happen to the best
of them.
Kurt
Kurt Strain
: Ron Bales wrote in message <37690dc4...@news.gte.net>...
It's all a waste of time. You're born , you die, and the rest in between is
just filler.
Really, Henry. Your good writing and good explanations are not all unnoticed.
It's just that good writing seldom gets a response out of people. People
here want to tell others more than they want to learn from others, the ones
that post a lot, that is. The lurkers are still reading and learning.
Most reasonable people can tell the truth from the trash, and you want
your audience to be the reasonable people. Writing about Andre is publically
writing just to Andre. Few others care about this stuff. I know I don't.
Writing about Andre is not where the reasonable people go to learn.
Why do you care what opinion a man whose opinion you don't respect is?
I see another flame about or from Andre and I just skip the message entirely.
I read your statements concerning a technical argument and I read it
completely. I don't usually respond - I usually can't disagree with it, and
there's often nothing I can add to it. So it often stands alone. You know
you've made a good point when there's no response. John may want the simplest
possible form of explanation, but others are willing to hear other points
of view.
I think Usenet usually is a waste of time. But it also can be a
chance to hone your skill at presenting a good argument, and teach. And
you've done both waste time and contributed well. It's up to you, but
you just can't expect much reward from a freely given public speech, except
to know that perhaps some people got your message.
Kurt
Andre Jute
hundreds of flames *about* me from Pasternack and his equally creepy
fellow stalkers in the Magnequest Mob, pouring their vitriol over
everyone on RAT except me--I just don't read them. To me they're
irrelevant.
Andre
Kurt Strain <kst...@sr.hp.com> wrote:
John Byrns
<nob...@nobody.com> wrote:
Hi Henry,
Sounds to me like you are answering exactly the same question. Perhaps
"hand-waving" was not exactly the best way to describe your explanation.
The point is that your explanation is too complicated for most people here
to understand, especially considering there is a perfectly valid
explanation that is easy to understand, if you can get source impedance's
out of your mind. I am sure your explanation is very interesting to the
more technically advanced, but for those with more limited math skills,
the fact that the plate and cathode currents must be essentially
identical, offers a more readily understood explanation to the phenomenon,
that doesn't require any great insight on the part of the reader. Once
the reader begins to get a grip on the situation, then your explanation
can offer some additional insight, but it seems about as clear as mud, and
can't be considered as a simple, or intuitive, explanation for the
beginner. KISS
Regards,
John Byrns
bobcx
Kurt Strain <kst...@sr.hp.com> wrote in message
news:7kbpbe$1...@canyon.sr.hp.com...
>
> The lurkers are still reading and learning.
> Most reasonable people can tell the truth from the trash, and you want
> your audience to be the reasonable people. Writing about Andre is
publically
> writing just to Andre. Few others care about this stuff.
Mr Jute's only purpose here is to abuse and victimize decent people, and con
manufactures and publishers. He takes advantage of those who do not know
better.
What about the new people? The people who do not "KNOW" him. His tactic is
to leave for a while to go "fishing" then come back wishing for a fresh
audience to run his con on. It is sad that people don't care, what if it was
one them that was being abused, lied about, coned, then the would care.
>I know I don't.
> Writing about Andre is not where the reasonable people go to learn.
> Why do you care what opinion a man whose opinion you don't respect is?
I think we all know what can happen when people like him are left unchecked.
> I see another flame about or from Andre and I just skip the message
entirely.
When one tells the truth about him it is not a "flame" it is a public
service message.
BC
Kurt Strain
: hundreds of flames *about* me from Pasternack and his equally creepy
: fellow stalkers in the Magnequest Mob, pouring their vitriol over
: everyone on RAT except me--I just don't read them. To me they're
: irrelevant.
Name withheld by request. But this does appear to be a flame to me, a
simple casual observance, despite the claim that it's not.
Kurt
bobcx
Andre Jute <an...@indigo.ie> wrote in message:
How not to become a Timmie!
(Besides my novels, I write how-to books for other artists,
including standard references like Writing a Thriller which
Ruth Rendell described in The Times as 'a private godsend',
and about which others of you have been kind enough to write
me grateful notes. A few years ago I chaired an invitation only
meeting on 'How to protect yourself against fans and wannabes
over twelve hundred writers and illustrators sought invitations
and a scheduled single morning session was extended for
three full days and evenings, causing me to miss all the other
conferences. This is the draft of a chapter from my text book,
'How to be a Successful Wannabe', for which I got the idea
from recent correspondence on this board; a publisher loves
the idea, contracts are here, so here we go.
BE 'ANARTIST' WITHOUT EVER CREATING ANYTHING EXCEPT
DENUNCIATIONS OF YOUR LETTERS
I know one fellow who really wants to be an author Well,
actually, he'd like to be an artist of any kind. It doesn't
matter what sort. Understand, he doesn't have any creative
urge, or even ideas that itch for expression. what he really
wants to be is to be 'a somebody'. He's got this idea in his
head, perhaps because there were artists in his family,
perhaps because he sees pop intellectuals being interviewed
on the television, that artists are accorded respect beyond
that given to, say, council employees or accountants.
He wants to be an artist so badly, he has run the words
together, 'anauthor' .
Let's call him Timmie. Of course he has no talent. But that
is no inpediment to being 'anauthor' . Nor does he have the
discipline to write anything extended. But don't let that worry
you either. This chapter tells how even a Timmie can define
himself as an intellectual without the need ever to do any
intellectual work. First, get some kind of a career on the
fringes of the arts. Desktop publishing is good because you
can call yourself a graphic designer without bothering to be
trained or otherwise inform yourself. Public relations, especially
for television, is also good- Next, join some group where you
are likely to run into real intellectuals, preferably with achievements
to their name. Published books are good, as are plays, even musicals;
painters are okay but musicians insist on technical jargon that is a
bore to learn. Give architects a miss; most of them are unemployed
and unemployable. If you are short of ideas, check out Chapter 3
"Hanging Out Right, er,Left--PC Places and People"
Having chosen your forum, ingratiate yourself- Be a clown, tell jokes.
Don't for, god's sake, actually attempt any serious work; if and people
can use it to judge the quality of your mind, your career as an intellectual
will be finished before you have even started. Let it be known you
are working on a magnum opus but are too modest to let anyone see any
part of it. You will of course imply at every opportunity that one day the
world will be stunned by your brilliance.
After a while you will be a fixture. You will not have to prove that you
belong. Now, pick on some prominent fellow with real achievements.
Suck up to him with flattery before you launch your attack. Be absolutely
certain that there is no cause for the attack except your ambition; people
will be too embarrassed to point out you are acting from envy alone.
Above all you don't want to be involved in a real argument. The best sort
to pick on is someone too busy to fight back, preferably someone with
strong, consistent opinions- Dont for the time being attack the opinions,
instead pick on some personal aspect of the great man's feet of clay.
Does he lack modesty? Condemn him as a braggart. If he is modest,
attack him for hypocrisy. If he wears glasses, attack him for
shortsightedness. That is a good start.
Now demonstrate how ruthless you are by attacking without
provocation a member of his family, preferably a child. That should
frighten him badly enough to shut him up while you go your merry way.
If he is over thirty, attack him as 'an old tart' Now switch the attack.
If he has strong opinions, attack him for stubbornness. If he is always
reasonable, attack him for being indecisive.
If in selfdefense he makes an appeal to intellectual honesty in discourse,
immediately and repeatedly accuse him of being a snob.
You can pretend to find fault with his work. Read up some past criticism
bun don't get bogged down in serious argument. If he has solid achievement
it has survived criticism and you as a wannabe are by definition incapable
of bringing serious criticism of substantial work. The best compromise is
to find a set of criticisms so lightweight that the obvious lack of
enterprise is in itself an insult.
Tell a lot of brazen lies brazenly. Do not deign to argue the merits of
these lies when challenged. Simply think up some bigger lies and make new
charges. Or repeat old ones in slightly different words. Take a high moral
tone; claim to be making the charges as a public duty. Study the career of
Joseph Goebbels, Hitler's propaganda minister; Goebbels was the greatest
PR man of the century.
Make everything you say a personal insult. Nothing is more dangerous than
the facts. The minute you allow the facts to be discussed seriously, it will
become clear that you are in a card game where you haven't price of a
stake.
Now, of course sometimes you will pick badly and some intellectual
heavyweight will emerge languidly to maul your tender sensitivities. In
that case smile with boyish charm and say it was all a bit of a joke,
wasn't it. If he doesn't believe you, accuse him of lacking a sense of
humour.
Retire as gracefully as you can manage. Write a grovelling private letter
saying you were under stress because your mother is sick and you of
course apologize abjectly. Appeal to his decency to let it end right there.
That way you can survive to betray someone else upon another day.
Don't make the stupid mistake of attacking someone who has rolled
over you once a second time. Lie in wait for a softer touch. Sooner
or later you will find a weakling and you will by lying about him be able
to destroy his confidence. Then your name will be made. You will be
the man who destroyed X!s reputation. You will never have to do anything
so brow wrinkling and perspiration-breaking as creating something original.
For the rest of your life you can do what you are best at, pose as
something you are not.
CASE STUDY
Now let us take Timmie and spot the mistakes he made on the way to
perfection because he didn't buy my book and follow my rules precisely.
First, he picked on a fellow who had already demonstrated that he would
wipe his backside with impertinent snots. Next, the fellow Timmie picked on
has had so many careers, he cannot remember all his achievements
(he offers visitors an expanding rule to measure his hardcover first
editions
in shelf-feet, for instance) ; even a serious critical attack on some part
of
his work simply leaves him fall back on the prestige of the rest while he
brings the less than brilliant part up to scratch. This was insensitive of
Timmie. Wannabe artists must at least pretend to sensitivity, or if they
lack it be lucky enough not to be caught out this badly.
Timmie should have had the intelligence to pick on someone whose single
achievement or few achievements are treasured like children. It also helps
no pick on someone who doesn't already possess monumental self confidence.
Many artists of perfectly good achievement have poor self-image and they
are easier to reduce to snivelling wrecks.
Bad signs in your choice include high academic honours, high-level business
experience and competence; political experience; military experience;
sporting achievements at national or international level particularly in
contact sports but most especially in sports where people die like
auto-racing or competitive transocean yachting.
Don't let ambition lead you to fucking with someone utterly out of your
league, as it did to Timmie. You're not going to believe this: Timmie
picked on a guy to whom a government erected a statue in his own lifetime,
the most monumental confidence-builder imaginable. If by now you're
thinking Timmie is so stupid I must have made him up, believe me, I didn't.
he is real- And there are a multitude more like him out there. He's a
lowest. common denominator case.
Then Timmie picked on this heavyweight bruiser not once, but again after
he was warned off, three times in all. This is seriously stupid. You have
to finish the job the first time or give it up for good. Worse, he tried
his luck the last time after it had been cogently pointed out, and never
contested, that Timmie was acting purely from envy. Not too bright, our
Timmie, as you have already observed. But this is only where the nightmare
starts.
Next Timmie made the serious mistake of protesting too much. It became
quite clear that he was not acting from high-minded public duty but from a
desire to be precisely like the great man, or whatever his misconception of
this fellow was. He admitted as much through his attempts to vary his
attack. This is dumb. You must not try for literary excellence -if you had
any, you wouldn't be following this route--but ram the same simple message
home again and again.
Timmie made the appalling mistake of conceding that this fellow's
achievements were substantial and unassailable. Wrong! wrong
Wrong! A cleverer wannabe simply concludes that the works are in fact
assailable by someone of real talent but keeps quiet about it.
One trick all wannabes should learn before they try their luck is when to
shut up- Timmie was so impressed with his clever gush of meaningless
words he did not know there was a point where they would stop being
meaningless and start hurting him.
Timmie's list of criticisms was so slight they could not pass even as an
insult; it was too clear to everyone that he had tried hard and failed.
He made the worse mistake of panting eagerly to be thrown another titbit
from the table of the great man's creativity when the correct approach
would be avidly to devour in secret everything the great man made,
consumption to be announced only when serious shortcomings were
discovered. discreet secrecy otherwise to be maintained with a lofty
disinterest expressed publicly at every opportunity.
Timmie is of course a good argument for permitting genetic scanning
and abortion of fetuses too low on the food chain to be anything but
a burden on society, but even Timmie could have succeeded if only he
had followed rny guidelines carefully.
Don't be a smartass and end up like Timmie, a well-known arsehole
and butt of cocktail party sniggers, the original for the sneering
admonition
'Don't be such a Timmie. He used not to be invited because of his pushy
personality, now he is not invited because everyone knows he's a no-talent;
an arsehole and a three-time loser.
If you were smart enough to modify the rules you wouldn't need this book.
Trust me.
Andre Jute
Andre Jute
Thanks, Kurt. As long as you understand that I have nothing whatsoever
to do with this scum's silly flame war except being its intended victim.
I'm on your side. I want these losers to stop trashing RAT as an
expression of their rage and frustration that their unprincipled
18-month campaign to destroy my reputation and livelihood has failed to
intimidate me.
By all means use any influence you may have with these anti-social
elements to persuade them to behave like human beings, but please leave
my name out of it.
Andre
An innocent bystander
henry_pa...@my-deja.com
> Sounds to me like you are answering exactly the same question.
No, it's not the same question.
> Perhaps "hand-waving" was not exactly the best way to describe
> your explanation.
Actually it was insulting and inaccurate.
> The point is that your explanation is too complicated for most
> people here to understand, especially considering there is a
> perfectly valid explanation that is easy to understand, if you
> can get source impedance's out of your mind.
First of all, would you please learn that you don't use an
apostrophe to make words like "impedance" plural?
The fact that equal currents flowing through equal impedances
cause equal voltage drops is simple and easy to understand. It
is the best answer to the question, "Is the split-load phase
splitter balanced?"
I already told you I was answering a different question: "If
the outputs remain balanced as the loads vary equally, how can
it be that the output impedances are different?"
> I am sure your explanation is very interesting to the more
> technically advanced, but for those with more limited math
> skills, the fact that the plate and cathode currents must be
> essentially identical, offers a more readily understood
> explanation to the phenomenon, that doesn't require any great
> insight on the part of the reader. Once the reader begins to
> get a grip on the situation, then your explanation can offer
> some additional insight, but it seems about as clear as mud,
> and can't be considered as a simple, or intuitive, explanation
> for the beginner. KISS
This is an incredibly condescending and presumptuous statement.
If I read you correctly, you have put yourself in the position of
deciding how much technical detail is appropriate for this newsgroup.
Listen, John. The fact that you do not understand what I have
written does not give you the right to dismiss it. You have the
right to post polite questions or criticism. You have the right
to ignore my article. But you do not have the right to scold me
simply because I have posted an idea that doesn't fit the narrow
bounds of your experience or imagination.
If you want to make yourself look intelligent instead of
merely dumb, figure out what I was saying and restate it more
clearly. The question is valid, the concept is elegant, and
the principle is useful and relevant. Don't blame me for the
limitations of your own intelligence.
Better yet, piss off. And you wondered aloud why I don't
bother to make a "positive contribution" to the newsgroup
anymore.
Damn! What a bummer.
-Henry
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
Ken Gilbert
an...@indigo.ie (Andre Jute) wrote:
> Leave my name out of it, Kurt. There are no flames *from* me, only
> hundreds of flames *about* me from Pasternack and his equally creepy
> fellow stalkers in the Magnequest Mob, pouring their vitriol over
> everyone on RAT except me--I just don't read them. To me they're
> irrelevant.
You're out of your fucking gourd, man--Absolutely, positively on crack.
Must get pretty lonely sometimes, what with the world being against you
and all.
--
Ken Gilbert
Tube Guitar Amp Design/Repair Technician
The Guitarist's Choice http://www.tgcguitar.com
http://www.geocities.com/SunsetStrip/Garage/5701
Andre Jute
> In article <1dtki1o.1hr...@ts01-007.bantry.indigo.ie>,
> an...@indigo.ie (Andre Jute) wrote:
> > Leave my name out of it, Kurt. There are no flames *from* me, only
> > hundreds of flames *about* me from Pasternack and his equally creepy
> > fellow stalkers in the Magnequest Mob, pouring their vitriol over
> > everyone on RAT except me--I just don't read them. To me they're
> > irrelevant.
>
> You're out of your fucking gourd, man--Absolutely, positively on crack.
>
> Must get pretty lonely sometimes, what with the world being against you
> and all.
>
> --
> Ken Gilbert
You're foaming at the mouth again, Kennie, and sputtering your inchoate
rage all over us.
Wipe your chin before you leave.
Andre
BobCZ
> Ken Gilbert <ride...@ride.ri.net> wrote:
>
> > In article <1dtki1o.1hr...@ts01-007.bantry.indigo.ie>,
> > an...@indigo.ie (Andre Jute) wrote:
> > > hundreds of flames *from* me from Andre Jute and my equally creepy
> > > fellow stalker in the Jute Mob, John Byrns, pouring our vitriol over
> > > everyone on RAT.--I just don't care I just lie, and lie to all. To me
I am
> > > irrelevant.
> >
> > You're out of your fucking gourd, man--Absolutely, positively on crack.
> >
> > Must get pretty lonely sometimes, what with the world being against you
> > and all.
> >
> > --
> > Ken Gilbert
>
my inchoate rage all over the decent honest people of the group.
>
>I wipe my chin with my sleeve before I leave.
>
> Andre
> --
> Andre Jute
> an...@dingo.ie CON MAN JUTE
> --we support NOTHING
John Byrns
Hi Eric,
In addition to the natural balance of the split load inverter, which holds
up well beyond the audio band, you have pointed out an additional feature
of the split load inverter. In addition to the plate tracking the cathode
very well, the response at both cathode and plate is quite flat, since as
you say the cathode follows the grid closely. Good observation.
Regards,
John Byrns
In article <7k9mk4$oc7$1...@holly.prod.itd.earthlink.net>, "Eric Vannerson"
<eric...@earthlink.net> wrote:
> It seems like all the elements of a clear explanation of good balance in
> this circuit are here, but no one has quite put them all together.
>
> Consider this:
>
> At high frequencies the impedance connected to the cathode decreases, but
> the signal voltage at the cathode is not much affected because of the large
> amount of negative feedback (All that happens is that the signal from grid
> to cathode increases slightly).
>
> Since, as was pointed out, the currents in plate and cathode currents are
> essentially the same, and the impedance in the plate circuit is the same as
> that at the cathode, the plate voltage doesn't change much with frequency
> either.
>
> It works for me.
>
> John Byrns wrote in message ...
mongrel
>
snip
> Does a 12at7 based phase splitter need a neg. screen bias?
snip
ELECTAUDIO wrote:
> The 12at7 has no screen!!
snip
Probably, he meant to say grid.
See, he restates the question before asking again:
snip
> I think I may have found the
> problem with my amp. For some reason, I've got exactly the same voltage on
> the grid and cat. of my phase splitter...
>
> any help would be great..
snip
It is unlikely that he thinks his 12at7 has a screen, and that it needs a
"negative bias".
mongrel
About the split-load phase inverter,
ELECTAUDIO wrote:
snip
> the sourceZ from the cathode is lower than the plate
> source so at the high freq will be unequal. the plate side will rolloff first.
snip
This comment led to an interesting thread.
LATER POSTS discussed (primarily) frequency response.
But, I wonder: were they OVERLOOKING, or were they simply NOT PRESENTLY
DISCUSSING, much bigger effects, which are functions of AMPLITUDE.
#1:
They mentioned the fact that, except for the small grid current of the driver,
the current in the plate load and cathode load must be the same. This is
true for the DC current through the tube. But AC current can flow through the
coupling capacitors on the grids of the power tubes. And this current becomes
large as the power tube grids begin to conduct. In the case of 6V6s, this happens
very early!
When the same thing happens on the DRIVER side of these capacitors, you
suddenly have (AC) current sinks and sources whose current does NOT pass
through the other load on the driver. And when this happens, the cathode side
of the driver is much better able to supply the power required.
#2:
Were they ignoring (just for the sake of argument) the curvature in the plate
characteristic of the driver? Ignoring other effects, such as #1 above, this
curvature causes the two outputs to be (symmetrically) distorted. And ideally,
if the power tubes were identical and running in class A, this symmetrical driver
distortion would be cancelled in the output transformer. In the case of my amp,
and many others, the power stage is not class A, and this particular driver
distortion will not cancel. (This is also a function of amplitude.)
Numbers 1 and 2, and probably others, are why the phase "splitting" function
should be done early, where the amplitude is very low. My compliments to the
poster elsewhere who mentioned that he prefers to do some voltage amplification
between the driver and power tubes.
If you were ONLY discussing frequency response, then never mind. It is good at
that.
LGeoCole
If the cathode impedance is lower than the plate impedance, there must be some
result. At low frequencies, these impedances appear in series with the reactive
coupling caps to the next stage. The result here would be that the cathode side
would have a flat response to a lower frequency. This could be accomplished by
the cathode side stealing current, positive and negative, from the plate side.
The average current in the load resistors would be the same, but the
instantaneous current would not.
The unequal impedances are from the tube only. The load resistors appear in
parallel with them and reduce their effect. The lower the load resistor values,
the greater the reduction. The grid resistors of the driven stage appear in
series with the splitter output impedances. The higher the values there, the
lower the effect of the differing impedances.
Again, the wisdom of the Williamson shows. A low Mu splitter with its low value
load resistors and an output driver stage with high value grid resistors works
well.
At high frequencies, miscellaneous parasitic circuit and conponent capacities
will mess up balance. That phenomenon is not greatly due to impedance
imbalance, but more to capacity imbalance.
Another note: There seems to be a fuzziness in the understanding of "Mu" by
some people. It is the maximum Vout versus Vin that a particular tube can
deliver at a specific plate current. Real life stage voltage gain will always
be less than Mu because of circuit loading (load resistor, driven grid
resistor, whatever).
George C.
mongrel
LGeoCole wrote:
> Maybe I can roil the waters somewhat, keep the ball rolling.
>
> If the cathode impedance is lower than the plate impedance, there must be some
> result. At low frequencies, these impedances appear in series with the reactive
> coupling caps to the next stage. The result here would be that the cathode side
> would have a flat response to a lower frequency. This could be accomplished by
> the cathode side stealing current, positive and negative, from the plate side.
One of my points was that the cathode can "current steal" from its coupling
capacitor.
For the same to happen on the plate side, this current would have to pass through
the
plate resistance: a serious asymmetry.
John Byrns
<mon...@bellsouth.net> wrote:
> This is addressed to all interested contributors to this thread.
>
> About the split-load phase inverter,
> ELECTAUDIO wrote:
>
> snip
>
> > the sourceZ from the cathode is lower than the plate
> > source so at the high freq will be unequal. the plate side will
rolloff first.
>
> snip
>
> This comment led to an interesting thread.
> LATER POSTS discussed (primarily) frequency response.
> But, I wonder: were they OVERLOOKING, or were they simply NOT PRESENTLY
> DISCUSSING, much bigger effects, which are functions of AMPLITUDE.
The discussion was primarily about balance vs. frequency. Earlier in the
thread, I did point out several times that the split load inverter doesn't
suffer grid current on the following stage well, and it is best to
intopose some sort of driver stage between.
> #1:
> They mentioned the fact that, except for the small grid current of the driver,
> the current in the plate load and cathode load must be the same. This is
> true for the DC current through the tube. But AC current can flow through the
> coupling capacitors on the grids of the power tubes. And this current becomes
> large as the power tube grids begin to conduct. In the case of 6V6s,
this happens
> very early!
> When the same thing happens on the DRIVER side of these capacitors, you
> suddenly have (AC) current sinks and sources whose current does NOT pass
> through the other load on the driver. And when this happens, the cathode side
> of the driver is much better able to supply the power required.
Actually it's even worse than that.
> #2:
> Were they ignoring (just for the sake of argument) the curvature in the plate
> characteristic of the driver? Ignoring other effects, such as #1 above, this
> curvature causes the two outputs to be (symmetrically) distorted. And ideally,
>
> if the power tubes were identical and running in class A, this
symmetrical driver
> distortion would be cancelled in the output transformer. In the case of
my amp,
> and many others, the power stage is not class A, and this particular driver
> distortion will not cancel. (This is also a function of amplitude.)
What do you mean by the "driver"? Are you talking about the phase
splitter tube, or another stage between the phse splitter and outputs? If
the latter, wouldn't this problem affect most other types of phase
spliters too?
> Numbers 1 and 2, and probably others, are why the phase "splitting" function
> should be done early, where the amplitude is very low. My compliments to the
> poster elsewhere who mentioned that he prefers to do some voltage
amplification
> between the driver and power tubes.
>
> If you were ONLY discussing frequency response, then never mind. It is good at
> that.
As I said, balance vs. frequency, which is different than frequency was
the main topic, but other issues, like grid current in the following stage
were mentioned several times.
Regards,
John Byrns
mongrel
My understanding is mostly qualitative. This was my first split-load
"splitter-driver".
I learned from this project that what I was aiming for would be better done by first
splitting,
and then driving BOTH sides with cathode followers (or other form of driver, as you
said)
>
> > #2:
> > Were they ignoring (just for the sake of argument) the curvature in the plate
> > characteristic of the driver? Ignoring other effects, such as #1 above, this
> > curvature causes the two outputs to be (symmetrically) distorted. And ideally,
> >
> > if the power tubes were identical and running in class A, this
> symmetrical driver
> > distortion would be cancelled in the output transformer. In the case of
> my amp,
> > and many others, the power stage is not class A, and this particular driver
> > distortion will not cancel. (This is also a function of amplitude.)
>
> What do you mean by the "driver"? Are you talking about the phase
> splitter tube, or another stage between the phse splitter and outputs?
I was referring to the phase splitter.
> If
> the latter, wouldn't this problem affect most other types of phase
> spliters too?
>
> > Numbers 1 and 2, and probably others, are why the phase "splitting" function
> > should be done early, where the amplitude is very low. My compliments to the
> > poster elsewhere who mentioned that he prefers to do some voltage
> amplification
> > between the driver and power tubes.
> >
> > If you were ONLY discussing frequency response, then never mind. It is good at
> > that.
>
> As I said, balance vs. frequency, which is different than frequency was
> the main topic, but other issues, like grid current in the following stage
> were mentioned several times.
>
> Regards,
>
> John Byrns
Thanks for your reply.
Because of my recent hard work on this subject, I wanted it to be clear
that statements like this:
snip
Can you provide any justification for the claim that "the plate side will
rolloff first"? In real life the split load inverter goes out further
than most other simple inverters before becoming unbalanced. The supposed
unbalance of the split load inverter is just an Urban legend that has
never been justified by anyone anywhere that I have ever been able to
find. The differing impedances of the palte and cathode do not cause
imbalance.
snip
apply to its frequency response, but not to its ability to supply power.
And I hope everyone noticed comments like these:
snip
It's a good splitter.
I think it's also important to remember that more than a few amps used
it to drive the output tubes as well, and in this case it doesn't work
so hot. It's way better to have a stage between it and the output tubes.
Ron
snip
Thanks.
mongrel
mongrel
> In article <376EA298...@bellsouth.net>, mongrel
> <mon...@bellsouth.net> wrote:
>
> > This is addressed to all interested contributors to this thread.
> >
> > About the split-load phase inverter,
> > ELECTAUDIO wrote:
> >
> > snip
> >
> > > the sourceZ from the cathode is lower than the plate
> > > source so at the high freq will be unequal. the plate side will
> rolloff first.
> >
> > snip
> >
> > This comment led to an interesting thread.
> > LATER POSTS discussed (primarily) frequency response.
> > But, I wonder: were they OVERLOOKING, or were they simply NOT PRESENTLY
> > DISCUSSING, much bigger effects, which are functions of AMPLITUDE.
>
> The discussion was primarily about balance vs. frequency. Earlier in the
> thread, I did point out several times that the split load inverter doesn't
> suffer grid current on the following stage well, and it is best to
> intopose some sort of driver stage between.
Perhaps I overlooked some of your contributions. I just revisited the group after
a long absence.
>
> > #1:
> > They mentioned the fact that, except for the small grid current of the driver,
> > the current in the plate load and cathode load must be the same. This is
> > true for the DC current through the tube. But AC current can flow through the
> > coupling capacitors on the grids of the power tubes. And this current becomes
> > large as the power tube grids begin to conduct. In the case of 6V6s,
> this happens
> > very early!
> > When the same thing happens on the DRIVER side of these capacitors, you
> > suddenly have (AC) current sinks and sources whose current does NOT pass
> > through the other load on the driver. And when this happens, the cathode side
> > of the driver is much better able to supply the power required.
>
> Actually it's even worse than that.
>
> > #2:
> > Were they ignoring (just for the sake of argument) the curvature in the plate
> > characteristic of the driver? Ignoring other effects, such as #1 above, this
> > curvature causes the two outputs to be (symmetrically) distorted. And ideally,
> >
> > if the power tubes were identical and running in class A, this
> symmetrical driver
> > distortion would be cancelled in the output transformer. In the case of
> my amp,
> > and many others, the power stage is not class A, and this particular driver
> > distortion will not cancel. (This is also a function of amplitude.)
>
> What do you mean by the "driver"? Are you talking about the phase
> splitter tube, or another stage between the phse splitter and outputs? If
> the latter, wouldn't this problem affect most other types of phase
> spliters too?
Yes, this applies to whatever is driving the usual push-pull output. I'm a
guitar-amp type. (There goes my credibility in this ng.) But I'm sick of soft driver
stages. In fact, I'm a little bit sick of guitar amp distortion. (Don't tell
anybody.)
A cute little Silvertone amp has inspired me to clean up my act. Its performance
as a hi-fi amp spoiled me.
When guitarists speak of "clean tone", they are usually referring to sounds that
are still very distorted, in special ways. Even audiophiles are really in love with
small amounts of their favorite forms of distortion.
I'm visiting this newsgroup in search of "Less".
"Moderation in all things", my mother often said.
> > Numbers 1 and 2, and probably others, are why the phase "splitting" function
> > should be done early, where the amplitude is very low. My compliments to the
> > poster elsewhere who mentioned that he prefers to do some voltage
> amplification
> > between the driver and power tubes.
> >
> > If you were ONLY discussing frequency response, then never mind. It is good at
> > that.
>
> As I said, balance vs. frequency, which is different than frequency was
> the main topic, but other issues, like grid current in the following stage
> were mentioned several times.
For the sake of anyone studying the matter, I'm glad this is clear.
>
> Regards,
>
> John Byrns
John Byrns
<mon...@bellsouth.net> wrote:
[Snip]
> > > #1:
> > > They mentioned the fact that, except for the small grid current of
the driver,
> > > the current in the plate load and cathode load must be the same. This is
> > > true for the DC current through the tube. But AC current can flow
through the
> > > coupling capacitors on the grids of the power tubes. And this
current becomes
> > > large as the power tube grids begin to conduct. In the case of 6V6s,
> > this happens
> > > very early!
> > > When the same thing happens on the DRIVER side of these capacitors, you
> > > suddenly have (AC) current sinks and sources whose current does NOT pass
> > > through the other load on the driver. And when this happens, the
cathode side
> > > of the driver is much better able to supply the power required.
> >
> > Actually it's even worse than that.
>
> My understanding is mostly qualitative. This was my first split-load
> "splitter-driver".
> I learned from this project that what I was aiming for would be better
done by first
> splitting,
> and then driving BOTH sides with cathode followers (or other form of
driver, as you
> said)
Hi Again,
I guess for benefit of everyone, I should have explained why I said "it's
even worse than that". When the grid of the following tube on the cathode
side is driven into grid current, a low impedance is presented to the
cathode for the period that the grid current flows. This low impedance on
the cathode of the phase splitter increases the gain at the plate way
above unity, creating high amplitude negative going spikes on the plate,
and the grid driven by the plate. This results in even greater distortion
than the grid current would have caused in the first place. The effect is
not symmetrical, when the grid connected to the plate is driven into grid
current, the gain at the cathode only increases slightly, since it is like
a cathode follower, and can't go above one.
Regards,
John Byrns
David A. Basford
mongrel wrote:
>
I'm a
> guitar-amp type. (There goes my credibility in this ng.)
You're not the only one here. While my interest in vacuum tubes covers
many bases, I am highly interested in guitar amp design. It's fun
because it involves more trial and error than hi-fi for the simple
reason that distortion is often intentional.
> But I'm sick of soft driver stages. In fact, I'm a little bit sick of > guitar amp distortion. (Don't tell
> anybody.)
> A cute little Silvertone amp has inspired me to clean up my act. Its performance
> as a hi-fi amp spoiled me.
> When guitarists speak of "clean tone", they are usually referring to sounds that
> are still very distorted, in special ways. Even audiophiles are really in love with
> small amounts of their favorite forms of distortion.
> I'm visiting this newsgroup in search of "Less".
> "Moderation in all things", my mother often said.
>
I just removed the old Schmitt inverter that was in my homebrew guitar
amp. The Schmitt was mushy sounding. I even tried using a 12AT7 instead
of a 12AX7. This improved it a little, but it was still lacking clarity.
Now I have installed a phase splitter followed by a differential
amplifier (a la Williamson) using two 12AU7's. Wow! What a difference. I
can hear the individual notes now. I'm of the view that a guitar amp
should allow the natural sound of the guitar to shine through. I do like
distortion though; it only needs to add edge to the sound and not any
color. More tweaking is eminent, but I'm getting very close.
--
-=-=-=-=-=-=-=-=-=-=-=-=-
David A. Basford
therm...@mindspring.com
-=-=-=-=-=-=-=-=-=-=-=-=-
Andre Jute
> Even audiophiles are really in love with
> small amounts of their favorite forms of distortion.
Never a truer word.
But the phase splitter discussion was specifically aimed at AC balance,
which is in itself a form a distortion reduction.
Of course, an audiophile's "favourite" form of distortion is usually
merely what he is used to. Of those audiophiles who have never heard a
class A1 PP amp, or an SE amp with suitable speakers, we should really
say that they are in love with *familiar* distortion.
A motoring analogy: every family car is set up with the front suspension
biased towards understeer because that is safest. The same applies to
the largest majority of so-called "sports" cars too. (A car set up to
handle neutrally under most conditions, at the limit has whiplash
oversteer, like the Porsche RS of fond memory.) Anyone who had driven
only family saloons or posture-cars might easily say he preferred the
soggy handling (handling is what happens at the limit, when roadholding
is exceeded). It is safe, uninspired, familiar--and the inexperienced do
not know that it is dull, that there is a horizon of experience and
delight beyond that point. However, if placed only briefly in a best
quality sports car an inexperienced person might conclude that it is
intrinsically dangerous (as the E-type was because of its narrow rear
track) or that one requires too much concentration and skill to drive it
successfully (as any Porsche short of the 4WD versions do). Genuinely
fast cars are an acquired taste.
So is malt whisky, great art, and pure class A1 PP and SE amps.
It all depends on where you start from.
bobcx
Andre Jute <aph...@indigo.ie> wrote in message
845.cork.in.his.butt.eeeeeee...
>
>
> Audiophiles are sick of my lies with
> large amounts of my ridiculous phony distortions.
>
>I have never spoken a true word.
>
> But the phase splitter discussion was specifically aimed at honest people,
> but I got involved which is in itself a form a distortion.
>
> Of course, my "favorite" forms of distortion are usually my "Action Man"
tails.
> Of those audiophiles who have never herd of me I am Andre Jute, con-man,
liar , and thief. I am infamous. Please read my story below and marvel at my
greatness!
****************************************************************************
***********
Subject: Re: The Quantum Mechanics of SRPP--or, Is SRPP a low entropy region
in tube audio?
From: an...@indigo.ie (=?ISO-8859-1?Q?Andr=E9_Jute?=)
Date: 1998/06/05
Newsgroups: rec.audio.tubes
I have this Russian military aeroplane designer doing
the aerodynamics on a hunting car I'm designing for an Arab.
The rich are different from the rest of us, XXXXXXX. Waste is not
a parameter. I doubt this particular hunting car will ever go off the
tarmac except when on delivery I demonstrate that it will go faster
crosscountry than anything else. I make it capable of going crosscountry
with the best of them, because that is what I am paid to do, but also
luxurious and fast, because I like repeat orders.
>
I would refuse a commission to design a hunting car for anyone I thought
would actually hit any animals he shot at. because of my self-evident lack
of
culture. Now, if the fellow was shooting at peasants, I would give him a
discount for his contribution towards population control...
>) doesn't involve driving huge distances or going particularly
> fast. So good aerodynamics in a vehicle is really pointless because the
> speeds and distances are relatively small. However functionality is a
> must. So what are you doing in a wind tunnel?
Oh, it's functional all right, as a hunting vehicle, with a huge sliding
roof and a shooting chair with more elevating and swivelling and tilting
controls than an entire Hatteras 60. But it is also functional as a
tarmac express, 155mph worth of express, I'm aiming for. That is what I
am doing in a wind tunnel. The main thing is to get a bellypan which
smoothes the air through the necessary ground clearance and channels it
out to the advantage of downforce, traction and control.
In fact, apart from the pure frontal area, a hunting car or any estate
is an easy solution to a fast car problem because the side elevation at
the rear gives you a good aerodynamic centre of pressure the moment you
draw the lines, plus uninterrupted air adhesion all the way back to a
natural Kamm cutoff, and if you make the sunroof rails external they
serve the double function of aerodynamic guidons.
****************************************************************************
***
So is bull whisky a great drink or not? As I smack my lips and smile, and
then wipe my chin. It all depends on which end of the bull you start from.
>
> Andre
> --
> Andre Jute
> a...@digo.ie CON MAN JUTE
> --we support pages for liars, thieves and con-men at
> http://ndigo.ie/~andre/ConJuteFU.html
mongrel
> mongrel wrote:
> >
> I'm a
> > guitar-amp type. (There goes my credibility in this ng.)
>
> You're not the only one here. While my interest in vacuum tubes covers
> many bases, I am highly interested in guitar amp design. It's fun
> because it involves more trial and error than hi-fi for the simple
> reason that distortion is often intentional.
Yes. It is very interesting.
Here is something I didn't understand when I first started, years ago. Guitar amp
design is not about "optimization". Of course, it requires a suble blending of compromises.
((In fact, all of electrical engineering and all of life is more about compromise and
opinion, than optimization.))
I had trouble "finishing" amps, because I wanted to "optimize" them before declaring
them finished. I then continuously oscillated between component values here and
there, depending on my mood. I would advise one to get A desired, groovy tone,
record the schematic, put the chassis IN THE CABINET and enjoy it, keeping it as
a reference, then build another, different one! (There is another good reason: the
memory is not a good enough reference.)
> > But I'm sick of soft driver stages. In fact, I'm a little bit sick of > guitar amp distortion. (Don't tell
> > anybody.)
> > A cute little Silvertone amp has inspired me to clean up my act. Its performance
> > as a hi-fi amp spoiled me.
> > When guitarists speak of "clean tone", they are usually referring to sounds that
> > are still very distorted, in special ways. Even audiophiles are really in love with
> > small amounts of their favorite forms of distortion.
> > I'm visiting this newsgroup in search of "Less".
> > "Moderation in all things", my mother often said.
> >
>
> I just removed the old Schmitt inverter that was in my homebrew guitar
> amp. The Schmitt was mushy sounding. I even tried using a 12AT7 instead
> of a 12AX7. This improved it a little, but it was still lacking clarity.
>
> Now I have installed a phase splitter followed by a differential
> amplifier (a la Williamson) using two 12AU7's. Wow! What a difference. I
> can hear the individual notes now. I'm of the view that a guitar amp
> should allow the natural sound of the guitar to shine through. I do like
> distortion though; it only needs to add edge to the sound and not any
> color. More tweaking is eminent, but I'm getting very close.
Very good. Good luck.