New page on Squares waves and amplifier performance
 

"David Looser is an absolute IDIOT "

It may have been "unheard of", but it existed all the same. Any amplifier,
with or without feedback, can exhibit slew-rate limiting if the bandwidth
is insufficient to cope with the rate of rise or fall of the input signal.


** This is just completely WRONG !!

Bandwidth and rise times are small signal parameters - but " slew rate
limiting " is a LARGE signal phenomenon.

All audio amplifiers have slew rate limits, usually quoted in V/uS - often
different rates apply for positive and negative going output voltages.

The usual cause is a current source internal to the amp's topology being
driven to its limit in trying to charge and discharge internal capacitances.

The slew rate is not knowable by any small input signal test like rise time
or BW.



..... Phil

New page on Squares waves and amplifier performance
 

"Ian Iveson" wrote

Or do you believe it can? So let's take a weedy AX7 CF driving 10nF, set
up as per a datasheet example, with 220k cathode resistor, and a bias of
1V at 1.2mA. Can you tell me what it's slew rate limit is, expressed as a
single number?

Why does it need to be expressed as a single figure? And what is an AX7?
I've never heard of it and none of my valve data books list it?

snip

However, a signal of less than 1V would not be limited no matter how high
its slew rate, although it might be attenuated by the filter. So it's not
the slew rate that causes the limit, but rather the combination of
amplitude and frequency.


Your conclusion is erroneous.

OK, let's take your example. With a 220k cathode resistor and 1.2mA the
cathode voltage will be 264V, seems excessively high but let that pass. With
no signal the 10n capacitor is charged to that voltage. Now we apply a fast
0.5V negative going pulse to the grid so the negative grid bias on the valve
increases by 0.5V. Now if the gm of the AX7 is about 1.5mA/V the valve
current drops by 1.5 x 0.5 = 0.75mA, so it is now 1.2 - 0.75 = 0.45mA. The
valve has not cut-off, but it's current is no longer capable of supporting
264V across the cathode resistor. The "missing" current is, instead,
supplied by the discharge of the 10n capacitor. As that capacitor discharges
the cathode voltage falls, reducing the negative grid bias, and the valve
current increases until the cathode voltage has fallen by fractionally under
0.5V, by which time the valve current is now capable of supporting a volt
drop of 263.5V across the 220k cathode resistor. The point is, this takes
*time*, time dependant on the time-constant of the 220k resistor and the 10n
load capacitor. With a current of 0.75mA and 10nF the time for an 0.5V drop
is about 6.6 microseconds. In this case that is modified by the fact that
valve current increases as the voltage falls, but this is clearly still
slew-rate limiting.

David.

New page on Squares waves and amplifier performance
 

David Looser wrote:

Or do you believe it can? So let's take a weedy AX7 CF
driving 10nF, set
up as per a datasheet example, with 220k cathode
resistor, and a bias of
1V at 1.2mA. Can you tell me what it's slew rate limit
is, expressed as a
single number?

Why does it need to be expressed as a single figure? And
what is an AX7? I've never heard of it and none of my
valve data books list it?

12AX7 if you like, clever dick, but any ..AX7 would do.

snip

However, a signal of less than 1V would not be limited no
matter how high
its slew rate, although it might be attenuated by the
filter. So it's not
the slew rate that causes the limit, but rather the
combination of
amplitude and frequency.


Your conclusion is erroneous.

Not according to this AX7 cathode follower I have in front
of me.

OK, let's take your example. With a 220k cathode resistor
and 1.2mA the cathode voltage will be 264V, seems
excessively high but let that pass. With no signal the 10n
capacitor is charged to that voltage. Now we apply a fast
0.5V negative going pulse to the grid so the negative grid
bias on the valve increases by 0.5V. Now if the gm of the
AX7 is about 1.5mA/V the valve current drops by 1.5 x 0.5
= 0.75mA, so it is now 1.2 - 0.75 = 0.45mA. The valve has
not cut-off, but it's current is no longer capable of
supporting 264V across the cathode resistor. The "missing"
current is, instead, supplied by the discharge of the 10n
capacitor. As that capacitor discharges the cathode
voltage falls, reducing the negative grid bias, and the
valve current increases until the cathode voltage has
fallen by fractionally under 0.5V, by which time the valve
current is now capable of supporting a volt drop of 263.5V
across the 220k cathode resistor. The point is, this
takes *time*, time dependant on the time-constant of the
220k resistor and the 10n load capacitor. With a current
of 0.75mA and 10nF the time for an 0.5V drop is about 6.6
microseconds. In this case that is modified by the fact
that valve current increases as the voltage falls, but
this is clearly still slew-rate limiting.

Absolute ********. The "time" it takes depends on the output
impedance which you said doesn't matter, which will always
be low, regardless of dV/dt, for a signal smaller than the
bias voltage. It only becomes equal to the cathode resistor
if the valve is turned off.

If we can't agree on the meaning of "slew rate limiting"
then there is really no point in discussing it. Language
requires common meaning, obviously. Looking at several
textbooks and on-line sources, I see that my definition is
in keeping with informed consensus.

And if you don't understand the fundamentals and have no
interest in learning, then there's no point in me trying to
explain. I've done my best, I've checked with books, I've
checked by experiment just to make sure I won't mislead you,
and you've just blethered none-stop nonsense regardless, and
made no effort whatsoever.

You've invented a world of your own here. That's what
happens with smug cliques.

Ian

New page on Squares waves and amplifier performance
 

I'm afraid of wide open spaces, ****-face.

My friend CleverDick said I'm scared of area, arse-hole.

I said no, I panic if the space I'm in is both wide and
long, die-of-testicular-cancer pin-brain.

That's fear of area said CleverDick, idiot.

No, I said. I'm quite comfortable with area as long as it's
long and thin, or short and wide.

That doesn't matter, said CleverDick. A wide open space has
a large area, and you are afraid of wide open spaces, so you
must be scared of area.

Have it you're own way, CleverDick, I said,
die-in-a-putrid-boil-infested-pit-of-snakes-and-worms.

We came to a long corridor. CleverDick said I'd better not
go in there, because of the area.

Ian

New page on Squares waves and amplifier performance
 

"Ian Iveson" wrote in message
...
David Looser wrote:


12AX7 if you like, clever dick, but any ..AX7 would do.

"clever dick" yourself. If you meant 12AX7 why not say 12AX7? How was I
supposed to know what you meant by an "AX7"?


And if you don't understand the fundamentals and have no interest in
learning, then there's no point in me trying to explain. I've done my
best, I've checked with books,

What "fundamentals" do you think I don't understand? ISTM that you are
working from simplified models of CF operation that don't take into account
real-world effects.

I've checked by experiment just to make sure I won't mislead you,
..
How was this "experiment" conducted? what sort of signals did you use, and
how was the output analysed? ISTM that I need to try some experiments of my
own.

and you've just blethered none-stop nonsense regardless, and made no
effort whatsoever

You've invented a world of your own here. That's what happens with smug
cliques.

So you've resorted to personal abuse have you? Is that any way to settle
technical arguments?.


David.

New page on Squares waves and amplifier performance
 

Ian Iveson wrote:
I'm afraid of wide open spaces, ****-face.


Ah, Phil's got a double-barrel name, must be one of the Australian
'aristocracy'. I thought he was particularly erudite..being able to string more
that four words together an that.

--
Bill Coombes

New page on Squares waves and amplifier performance
 

"David Looser" wrote in message
...
"Trevor Wilson" wrote

**Perhaps our experience differs. Whilst the results of such a waveform
are not perfect, they are considerably superior to Red Book CD. I've even
run such tests on high performance (double speed) Compact Cassette
machines and the square wave response is surprisingly good. Of course,
such machines can easily manage -3dB @ 30kHz.

"easily"? at 9.5cm/sec? I don't think so!

**Yes, easily.


If a double-speed CC has been lined up to be -3dB at 30kHz than it has
been wrongly lined up. 30kHz is quite unnecessary for audio, and in the
case of analogue tape can only be obtained at the expense of poorer
*audio* performance.

**It was aligned to factory specs.


In any
case we are talking about expensive studio machines here, the
relevant comparison would be to 96 and 192kHz sampled audio, which
clearly has a far greater bandwidth.

**Strawman noted.

Hardly a "strawaman".

**It's a strawman. because I SPECIFICALLY referred to Red Book CD. Nothing
else. I have never, nor will I ever, argue that higher sampling rates do not
allow superior square wave performance. Any nong knows that. We're
discussing RED BOOK CD.

You were comparing a digital format unfavorabley to
some analogue formats. I was merely pointing out that higher sampling rate
digital systems easily beat the analogue ones you are praising.

**Duh.



ontact area of even the
smallest stylus?

**This one seems to do the trick:

http://www.dynavector.com/etechnical/microstylus.html

And how much does one of those cost?

**Another strawman. We're simply discussing the abilities of various
formats.

For the record: I am not arguing that a cheap Red Book CD cannot outperform
even high end vinyl in most areas. However, in ONE, VERY SPECIFIC area
(namely: high frequency square waves), SOME vinyl reproduction systems are
capable of superior performance.


The cartridge has an upper frequency limit of 100kHz.

To what point? Who needs 100kHz in an audio system?

**Another strawman. We're discussing the square wave reproduction ability of
a SPECIFIC vinyl reproduction system to RED BOOK CD. The audible
significance is not the issue.

I have some 10kHz square wave CRO photos somewhere around the place.
I'll post them in a week or two. I've examined this stylus under a
binocular microscope. I was more than a little suprised to see pretty
much what the photos show. The stylus is a beautiful thing.


It may be "beautiful", but it's pointless (and no doubt very expensive).
How long does it last?

**Again: Strawman.


I question the ability of even that sort of thing to consistently deliver
100kHz. One of the reasons the CD4 'quadraphonic' system failed was the
inabilty of vinyl technology to reliably deliver the subcarrier to the
decoder.

**No. It failed for a host of reasons. The appalling software was one and
the high cost was another. The short life-span of the disks was a relatively
minor issue.


--
Trevor Wilson
www.rageaudio.com.au

New page on Squares waves and amplifier performance
 

"Jim Lesurf" wrote in message
...
In article , Trevor Wilson
wrote:
David Looser wrote:
"Trevor Wilson" wrote



As for vinyl, a typical minimum groove radius is 6cm. That means a
linear velocity of about 210 mm/sec. So the wavelength of a 60kHz
signal is .0035mm. What is the sidewall contact area of even the
smallest stylus?

**This one seems to do the trick:

http://www.dynavector.com/etechnical/microstylus.html

Due to the deformation of the wall, simply having a physical contact area
of 2 microns does not gurantee that is the actual resolution.

The cartridge has an upper frequency limit of 100kHz.

How was that value measured? I can't see it on the above page.

**The usual way, presumably. A calibrated transducer is attached to the
stylus and is driven with a sweep signal.


--
Trevor Wilson
www.rageaudio.com.au

New page on Squares waves and amplifier performance
 

David Looser wrote:


12AX7 if you like, clever dick, but any ..AX7 would do.

"clever dick" yourself. If you meant 12AX7 why not say
12AX7? How was I
supposed to know what you meant by an "AX7"?

I reckoned if you didn't see what I meant, you wouldn't
care. The 12AX7 (or other ..AX7) is the most well known
valve ever, has the largest gain of common triodes, makes
the weediest CF, and is consequently the best well-known
candidate to illustrate the effect we are talking about.

And if you don't understand the fundamentals and have no
interest in
learning, then there's no point in me trying to explain.
I've done my
best, I've checked with books,

What "fundamentals" do you think I don't understand? ISTM
that you are
working from simplified models of CF operation that don't
take into account
real-world effects.

I've checked by experiment just to make sure I won't
mislead you,
.
How was this "experiment" conducted? what sort of signals
did you use, and
how was the output analysed?

Sine waves of various illustrative combinations of amplitude
and frequency. Simulated well enough to not require the real
thing, that I can't be bothered to make for the sake of an
argument and you probably still wouldn't believe me anyway.
I'm not that daft, quite.

Square waves are awkward to use in simulation because
stepwise approximations are, er, slew rate limited by
nature, and a second can take all day for my pea-brained PC.

ISTM that I need to try some experiments of my
own.

Please do. That would be best.

and you've just blethered none-stop nonsense regardless,
and made no
effort whatsoever

You've invented a world of your own here. That's what
happens with smug
cliques.

So you've resorted to personal abuse have you? Is that any
way to settle
technical arguments?.

It's not intended to be abusive. It's an observation and an
indication of exasperation.
Beginning our exchange with "nope" didn't help. I'm dealing
with several ppl at once who appear to often act in concert.
I can understand that is because I am at odds with a local
consensus, so I'm not too upset. On the other hand, if a
consensus becomes smug as a result of being satisfied with
its own company, it can lose contact and develop a language
of its own, and if it's not careful, an imaginary
engineering to go with it. I can see how Phil got so angry.

I could do this...I could post pictures showing a simulated
12AX7 failing, and not failing, to drive 10n, with whatever
graphs you might fancy. At the end of the day, you would see
that it is happy with high frequencies, and behaves as you
would expect a filter with around 650 ohms and 10n. The cap
charges merrily, and discharges equally well, into that 650
ohms. There is nothing at all that you might call slew rate
limiting.

Until the triode clips. The clipping point depends on both
amplitude and frequency (as a result of the feedback, which
is a necessary part of the process), and when it happens the
result is similar to slew-rate limiting, because the cap
quite suddenly sees 220k instead of 650 ohms and as you have
pointed out in the bit of sense you made, can't discharge
quickly enough. Even then, there isn't the characteristic
straight-line ramp of true slew-rate limiting, because there
is no true constant current source because, sigh, there
isn't enough gain in a triode to make one.

Trouble is, I think you would just say that *is* slew-rate
limiting. To me, if you said a circuit is slew rate limited,
I would assume it would fail if I exceeded whatever you told
me the limit is. If I then found it was perfectly happy with
any slew rate as long as I made sure I didn't drive it into
clipping, I would feel misled.

That's just the way it is. You said "nope", but my cathode
follower says yes.

I'm just here to help avert a worldwide panic of
line-driving cathode-follower owners. Just don't overdrive
them and you'll be fine.

Ian

New page on Squares waves and amplifier performance
 

In article , Trevor Wilson
wrote:

"Jim Lesurf" wrote in message
...

http://www.dynavector.com/etechnical/microstylus.html

Due to the deformation of the wall, simply having a physical contact
area of 2 microns does not gurantee that is the actual resolution.

The cartridge has an upper frequency limit of 100kHz.

How was that value measured? I can't see it on the above page.

**The usual way, presumably. A calibrated transducer is attached to the
stylus and is driven with a sweep signal.

Is that "the usual way"? if so, it is news to me. I assumed measurements
would need to be done with a real test disc if you want to assess the
results with factors like stylus shape, wall deformation, etc, taken into
account. Without that, a claim that it "has an upper frequency limit of
100kHz" become essentially meaningless for the situations of actual use -
i.e. playing vinyl discs.

I do recall that Shure used to use 45rpm test discs for some measurements
so as to help them to get test signals with higher radii of modulation
curvature. That is a little bit of a 'cheat' perhaps. But not in the same
class as basing a figure on directly driving the stylus with a transducer.
Presumably one made of harder material than vinyl.

All that said, you did state the above as a matter of fact. Where did you
get the figure from? Now you are just saying "the usual way, presumably"
which gives the impression that you don't actually know how it was
obtained.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html