| Author |
Message |
Patrick Turner
Guest
|
 Posted:
Thu Dec 30, 2004 7:38 pm Post subject:
Re: More Turneroid Bull |
|
|
Phil Allison wrote:
| Quote: | "Patrick Turner"
Phil Allison wrote:
Then imagine a TT is in use.
** The Turneroid will never get near it.
What happens when a stylus is dropped onto a disk ??
Could there be a subsonic transient ?
Only pumps the cone of a woofer.
The ESL57s input tranny saturates at a few volts at a few Hz.
??????
There's not a lot more i wish to conribute to the discussion on Quad ESL
speakers because we don't have an LCR model to refer to.
** All you need to know has been provided.
so the 5 Hz of impedance you have calculated above must be about right.
** You bet your sweet bippy ........
There would be SS amps which would blow a fuse/shit themsleves/trip the
protection circuits if one were to drop a stylus carelessly, or suffer a
turn on
thump from a preamp.
** Blow a fuse = annoying.
Trip the overcurrent crow bar = mildly annoying.
Sand bits shit themselves = very bloody annoying !!
( Unless you are a tech type and can swap a few devices quickly )
Such reactions don't occur in much gear, one just gets a
momenatry speaker wobble, and a bang, no worse than a large drum beat.
A solitary short lived transient through BJTs or tubes isn't usually
enough
to kill them.
** Think of 2N3055s driven by BD139/140s with say 33 volts and 9 amps for
about 100 mS, repeated a few times rapidly.
Sayonara baby.
|
Perhaps so.
I get a steady stream of SS amps which come to grief at christmas parties.
In a fortnight, the next mob of blown party amps are due.
First thing young Freddie does is boost the bass full, and press the loudness
button on.
Then on goes the techno-rap et all and poof goes the amp.
Often its because of touching frayed speaker leads, and the amp manages at
50mV audio levels parents use, but not at 10 vrms levels junior uses at a
party.
| Quote: |
Nobody I knows plays a tune called "Drop the Stylus Gladly",
composed by Troggo McWrecker, hi-fi system buster extraordinaire,
and played through a high power SS amp connected to ESL57.
** The advent of CD players has made it possible to use even fragile amps
with ESL57s.
With just ordinary care, sub sonic thuds are a thing of the past.
BTW
Nothing stops a simple RC combination ( say 2.2 ohms // 470 uF) being
installed in series.
|
Now you will have all the audiophiles screaming that raising the amp Ro
to 2 ohms at about 50 Hz will ruin the sound.....
2.2 ohms and 470 uF have a reactive impedance of 1.54 ohms at 154 Hz.
| Quote: |
Help heaps with any DC offsets and nasty sub sonics.
|
Perhaps.
I just have just about finished the SS amp with a ten-pack of MJL21194/93
outputs.
I assume the instantaneous current ability is about 75 amps peak.
I do hope that during normal use that if there is a problem,
the rail/mains fuses will blow before the output devices.
But I do have the all important DC detector, and a volt of offset for longer
than a second
trips a relay which turns off the main power tranny.
I have ginormous heatsink, and trying to make it get hot is difficult.
Aluminium is better than fans or heartbreak.
It took some taming to stop the cross-conduction at about 30 kHz.
I feel this will be the first and last large bjt amp I will ever bother to
build,
I'd much rather work with tubes or mosfets.
Patrick Turner.
| Quote: |
.............. Phil
|
|
|
| Back to top |
|
 |
Patrick Turner
Guest
|
| Posted:
Thu Dec 30, 2004 8:01 pm Post subject:
Re: More Turneroid Bull |
|
|
Phil Allison wrote:
| Quote: | "Patrick Turner"
Phil Allison wrote:
Nobody has a factual reference in black and white, so what's the use of
a
discussion?
** You did **not** ask !!!!
You were INVENTING numbers willy nilly with such GAY abandon it would
have spoilt your fun to supply facts.
I posted a URL for the ESL 63 today - with results from James Moir and
Assoc.
There is a Louis Challis review in Electronics Today, Nov 1982 that has
similar results.
I don't beg for info from anyone.
** More Turneroid paranoia posturing as an excuse.
You didn't volunteer any info.
** The Turneroid was never a party to the conversation.
I haver the '63 info, but its the '57 we also need.
** You posted an impedance table here yourself.
WTF else did you want ????
I have a hand drawn impedance curve for an ESL57 ( my own) that covers
DC
to 50 kHz - never seen the like published anywhere.
I had a graph of the impedance also of the '57 and it gave the following
figures,
which I posted last week, which nobody objected to.
They could be BS.
20Hz, 20ohms,
** Should be 6 ohms
The impedance doubles with each octave from 3 Hz up to 80 Hz - as expected
for a 55mH, 0.3 R inductor.
80Hz, 33,
200Hz, 16,
500Hz, 11,
2kHz, 10,
5 kHz, 9,
16kHz, 2,
30 kHz, 4.
** The rest are near enough.
|
OK, so your'e saying that the Z curve tallies about with what your view of '57
are.
You should know; you owned '57 for a long time did you not?
There was some long discussions of your experiences involving '57 and a certain
ancient sony CD player at aus.hi-fi as you know.
| Quote: |
Saturation is at 10 volts rms at 20 Hz.
Go weave your asinine, Quad hating, autistic bricklayer's swill around
that.
|
I don't hate Quads.
I have a "reserved opinion".
I neither confirm or deny that I love them.
I have never owned a pair, so WTF would I know?
10vrms at 20Hz would mean 20v would be Fsat at 40 Hz for the transformer.
Its about the same frequencies for saturation in Quad II amps, and most other
old tube amps of that time.
To better these figures so that 20v at 20 Hz is allowable,
there has to be twice the turns, or twice the Afe, and so the tranny becomes
much larger and heavier.
In 1957, Quad wasn't into making gear which was perfect; they made gear
which suited 95% of the folks who purchased it, and who thought the other makers
were
wasting time.
Deep bass wasn't a major concern for anyone in 1957.
Most of the population was used to awful systems or radios
which went from 100Hz to 3 kHz if they were lucky.
Deep bass wasn't included in recordings, and was carefully
filtered out because it was a darn nuisance.
ppl playing drums were asked to set up a long way away from the
studio mic.
But the next generation in 1970s had more expectations, and
Bob Marley said let dere be de bass,
and then we all grooved to good bass.
The somebody recorded a song called "And then he kissed me"
and all this junk was crammed onto a recording.....
Nowdays, young folks like sound levels at the gigs they go to and in their cars
which I thought were quite injurous to health.
ESL57 are junk to them.
Loud, in 1957, was anyone shouting louder than my mother
complaining about aspects of my behaviour, maybe 90 dB she could go.
Patrick Turner.
| Quote: |
............. Phil |
|
|
| Back to top |
|
|
Patrick Turner
Guest
|
| Posted:
Thu Dec 30, 2004 8:16 pm Post subject:
Re: More Turneroid Bull |
|
|
Phil Allison wrote:
| Quote: | "Patrick Turner"
Phil Allison wrote:
** Full power at 40 Hz is *extremely* rare on a commercial recording -
plus the ESL57 has no output down that low anyhow. 2 ohms at 18 kHz is
of
no consequence whatever.
Pinkie is a damn LIAR !!!!
Full power is rarely seen on any recording at 40 Hz or below.
Certainly not in 1955.
It arcs with inputs above 20 volts rms.
** Then use an amp with no more.
But you said 33 vrms was possible with '57.....
** Where was that ????
The Turneroid moron is a misquoting fucking arsehole !!
|
What I said :-
| Quote: | But then if the Quad ESL57 can only handle 15 watts before arcing,
|
Then you said :-
** More invented Turneroid garbage.
The ESL57 is rated to take +/- 33 volts peak safely.
I make that to be 23.331 vrms, so that makes me wrong.
Big deal.
| Quote: |
There is only one way to do it that works - and it is NOT the one where
the curve is simply extended.
So what does work?
** Go figure it out yourself - arsehole.
|
That's our boy, he never gives away any useful information....
Don't let it worry you, it will be 1,000 years before I get around to a stacked
'57 arrangement.
Why would one just "extend the curve" anyway?
Then the path lengths for the HF at least would be quite different.
But have there not been exotic examples of ESL arranged in columns,
and radiating all F in all directions? Not my choice for my room, that's for
sure.
There are a few websites devoted to '57 set ups fo anyone who wants to explore
the idea.
Patrick Turner.
| Quote: |
............. Phil |
|
|
| Back to top |
|
|
Chris Morriss
Guest
|
| Posted:
Thu Dec 30, 2004 9:11 pm Post subject:
Re: More Turneroid Bull |
|
|
In message <5lp5t098ccbfnoiuri2aaq2ckh27kccch8@4ax.com>, Stewart
Pinkerton <patent3@dircon.co.uk> writes
| Quote: | On Wed, 29 Dec 2004 19:01:29 +1100, "Phil Allison"
philallison@tpg.com.au> wrote:
"Stewart Pinkerton"
"Pat Turner"
..... the speaker that is designed to appear mainly resistive from the
amplifiers point of view
is a better designed speaker,
I'd agree with that, and I suspect that it's one reason why large
planars mostly sound very good.
** Large magnetic planers sound just awful and their resistive nature is
not of any benefit.
So, you're deaf as well as stupid and ignorant?
What a surprise.............................
|
Very few people seem to know how to position large planar speakers in a
room. Although my Maggies are stored under my bed at the moment (until
I can afford to buy a house with a room big enough to use them again), I
can't say they sounded bad in the days when I was using them in a large
room.
The problem with the ribbon tweeter Maggies is their tiny vertical
'sweet-line' caused by the 'comb-filter' effect of a long vertical
line-source ribbon with a length much longer than the wavelength of the
sound being emitted. The horizontal radiation pattern is of course
superb.
The Quad 63s and 988/989 don't suffer the same problem with their more
sophisticated driver zoning.
--
Chris Morriss |
|
| Back to top |
|
|
Stewart Pinkerton
Guest
|
| Posted:
Thu Dec 30, 2004 11:18 pm Post subject:
Re: More Turneroid Bull |
|
|
On Thu, 30 Dec 2004 18:37:58 +1100, "Phil Allison"
<philallison@tpg.com.au> wrote:
| Quote: |
"Stewart Pinkerton"
"Phil Allison"
- it's around 110-130 Hz
(varies quite a bit due to varying skin tension with humidity and
temperature), and it isn't above 8 ohms for more than an octave or so
either side, which is about as sharp as it gets for decent speakers.
** The 8 ohm points are at 50 Hz and 1 kHz - makes it over 4 octaves
wide.
It is certainly not "massive" or "sharp" .
Utter crap ...
** Fuck you - shithead.
|
Argued with your usual wit and brilliance..............
| Quote: | - the 8 ohm points I have from *measurements* made by Hi-Fi
Choice are at 95 and 500Hz, which is pretty sharp.
** 3.5 octaves wide is not a " massive sharp peak" - you stinking
LIAR.
|
For a speaker, it's sharp, and when did quoting facts become a lie?
Too many tinnies yet again, you colonial clown? :-)
BTW, if you actually knew anything about anything, the Q of a
resonance is determined by the -3dB points, not the reduction to a
nominal 8 ohms. For the '57, that gives a Q of about 10, which is
pretty damn sharp by any standard.
--
Stewart Pinkerton | Music is Art - Audio is Engineering |
|
| Back to top |
|
|
Phil Allison
Guest
|
| Posted:
Fri Dec 31, 2004 4:24 am Post subject:
Re: More Turneroid Bull |
|
|
"Patrick Turner"
| Quote: | Phil Allison wrote:
** Think of 2N3055s driven by BD139/140s with say 33 volts and 9 amps
for
about 100 mS, repeated a few times rapidly.
Sayonara baby.
Perhaps so.
|
** The SOA curves for the 2N3055 indicate that 30 volts at 9 amps is OK for
1mS bursts.
The BD drivers would be operating at 30 volts and about 1 amp.
A kamikaze pilot has more chance of surviving his final dive.
| Quote: | BTW
Nothing stops a simple RC combination ( say 2.2 ohms // 470 uF) being
installed in series.
Now you will have all the audiophiles screaming that raising the amp Ro
to 2 ohms at about 50 Hz will ruin the sound.....
|
** Doesn't ruin the Quad ESL 63.
http://www.quadesl.com/schematics/quad63_schematic.gif
See R1, R2 and R15 ???
| Quote: | 2.2 ohms and 470 uF have a reactive impedance of 1.54 ohms at 154 Hz.
|
** The ESL 57 has an impedance 28 ohms at 154 Hz.
| Quote: |
Help heaps with any DC offsets and nasty sub sonics.
Perhaps.
|
** A nearly 10 times reduction in DC offset current is not *perhaps*.
An impedance minimum of 3 ohms in the sub sonic region is not
*perhaps*.
Even unprotected 3055s would then survive.
.................. Phil |
|
| Back to top |
|
|
Phil Allison
Guest
|
| Posted:
Fri Dec 31, 2004 6:16 am Post subject:
Re: More Turneroid Bull |
|
|
"Stewart Pinkerton" ...
"Phil Allison"
| Quote: |
** The 8 ohm points are at 50 Hz and 1 kHz - makes it over 4
octaves
wide.
It is certainly not "massive" or "sharp" .
Utter crap ...
** Fuck you - shithead.
Argued with your usual wit and brilliance..............
|
** Wit is wasted on a witless, abusive, lying moron like Pinkie.
| Quote: | - the 8 ohm points I have from *measurements* made by Hi-Fi
Choice are at 95 and 500Hz, which is pretty sharp.
** 3.5 octaves wide is not a " massive sharp peak" - you stinking
LIAR.
For a speaker, it's sharp,
|
** Bullshit.
| Quote: | and when did quoting facts become a lie?
|
** The " massive sharp peak" bit is a lie - you pommy arsehole.
| Quote: | BTW, if you actually knew anything about anything, the Q of a
resonance is determined by the -3dB points, not the reduction to a
nominal 8 ohms. For the '57, that gives a Q of about 10, which is
pretty damn sharp by any standard.
|
** ROTFL - the topic is the ESL 63 not the ESL 57.
The 3 dB bandwidth of the impedance peak is about 70 Hz with a centre
frequency of 100 Hz .
So the Q is no more than 1.4 - you complete and utter moron.
The only fucking engineering Pinkie ever did must have been with toy trains
on the floor.
.............. Phil |
|
| Back to top |
|
|
Patrick Turner
Guest
|
| Posted:
Fri Dec 31, 2004 8:25 am Post subject:
Re: More Turneroid Bull |
|
|
Phil Allison wrote:
| Quote: | "Patrick Turner"
Phil Allison wrote:
** Think of 2N3055s driven by BD139/140s with say 33 volts and 9 amps
for
about 100 mS, repeated a few times rapidly.
Sayonara baby.
Perhaps so.
** The SOA curves for the 2N3055 indicate that 30 volts at 9 amps is OK for
1mS bursts.
The BD drivers would be operating at 30 volts and about 1 amp.
A kamikaze pilot has more chance of surviving his final dive.
BTW
Nothing stops a simple RC combination ( say 2.2 ohms // 470 uF) being
installed in series.
Now you will have all the audiophiles screaming that raising the amp Ro
to 2 ohms at about 50 Hz will ruin the sound.....
** Doesn't ruin the Quad ESL 63.
http://www.quadesl.com/schematics/quad63_schematic.gif
See R1, R2 and R15 ???
|
Yes, there is about 2.3 ohms in series with the primary,
and then there is the DCR of the primary.
| Quote: |
2.2 ohms and 470 uF have a reactive impedance of 1.54 ohms at 154 Hz.
** The ESL 57 has an impedance 28 ohms at 154 Hz.
|
So there would be no great effect of having the 2r2 and 470 uF in series
with the 28 ohms.
| Quote: |
Help heaps with any DC offsets and nasty sub sonics.
Perhaps.
** A nearly 10 times reduction in DC offset current is not *perhaps*.
An impedance minimum of 3 ohms in the sub sonic region is not
*perhaps*.
Even unprotected 3055s would then survive.
|
Well yes.
I wasn't aware that Walker provided such load tailoring to
cope with transformer saturation, it gives me ideas about doing the equivalant
in a tube amp which was to be used at LF.
In a typical anode to anode load of 5k, and
where the class B load in one tube was 1.25k, then 1k plus 3 uF
would have a pole at 50 Hz, but at 10 Hz, the tube sees
just under a k, mainly resistive.
Ppl don't bother to take such measures, and a saturation transient is fast,
so the cap would charge up since its in a circuit with current flow more in one
direction than the other.
Tubes laugh at the occasional saturation transient, although sometimes when the
tube saturates at
LF signals there is a burst of HF oscillations during the part
of the cycle where saturation occurs.
With a well critically damped and stabilised amp this does not occur.
In my 50 watt mosfet class A amp I have an OPT that is the same size as a 50
watt
OPT for a tube amp, and I was worried what the saturation might do.
Loading is 35 ohms drain to drain to a 5 ohm secondary.
There is about 2 ohms of DCR in the primary, so 1 ohms per two
mosfets per side of the PP circuit, so theroetically if the inductance collapses
the load is 1 ohm, and a peak current of 34 amps can flow since the supply is
+34v to the CT,
so and there is possibly 17 amps through each mosfet.
However, mosfets are transconductance devices, and I have 0.22 ohms in the
source
circuit of each mosfet so with a 1 ohm load the gain plummets to a low figure,
and the gate-source voltage is limited to 7.5v.
Its impossible to generate such large currents in the mosfets.
They don't try to cross conduct as all are N types, set up like one would in a
tube amp.
http://www.turneraudio.com.au/htmlwebpgs02/schema5050mosfet.htm
No matter what I did with the input signal, the Hiatchi mosfets showed no sign
of perishing due to saturation effects.
but I still didn't like the sudden saturation effects which I did observe below
10 Hz,
at full power..
I found that gapping the OPT C-cores with one layer of thin plastic
( from a shopping bag ) stopped the tranny from saturation and made the
tranny behave as a more benign inductance as the full power signal level
was applied below 20 Hz.
If one has 35 ohms as a load, then only 280 mH of primary inductance
for the OPT is required.
This gives a load reduction to 25 ohms where ZL = RL at 20 Hz.
This is quite tolerable for most amplifiers, and this ratio of L : RL
can be a guide.
At 5 Hz, the load would be nearly fully inductive, and about 6.5 ohms.
Gapping the core is a necessity in SE amps, where a large DC flow requires a
large gap
in the core, reducing the effective GOSS u from say 17,000 in an interleaved E&I
core in a PP amp
to maybe only 500 in the same core material used in an SE amp.
Below 20Hz, the inductance could rise depending on the material since the u
may have a peak value higher at 10 Hz than 20 Hz.
So exactly what imdedance one gets as F is reduced needs measuring to find out.
u is also highest when the voltage is highish.
Several variables are at work.
In my case the Lp of the 35 ohm winding was a couple of H, and when gapped this
fell
about 2.5 times, and could have been lower, but gapping definately
made the amp less prone to viciously sudden saturation, yet still
allowed for enough primary inductance to do its job, and superlative bass
performance.
When I build my next PP tube amp I will try gapping the cores.
The Australian made GOSS E&I material I get has a max u = 17,000.
I could afford to gap the cores so the u was 1,000.
Cores with u = 3,000, or Lycore 150, ie, medium grade power tranny iron
are quite OK in my designs of OPT,
and typical maximum inductance for a 5 k load with u = 3,000 gives
around 200H, falling to say 20H at some low voltage.
the Z 20H = 5k at 40 Hz, and where Lp = 200H,
the Z 200H = 5k at 4 Hz.
Using u = 17,000 at full power gives 1,100 H and a LF pole that
is hard to measure were it not for the cut off caused by saturation at say 15
Hz.
But at 5v a-a the material still has a large u, so lots of Lp,
and the pole between load and Lp is very low.
So above 20 Hz, the load is mainly resistive.
I don't normally gap PP cores with such high u material because
so far under normal musical programme conditions where the average signal level
using busy bass rich rock and roll is 10dB below sine wave clipping,
there is no sign of saturation distress.
My friend in Sydney who is entranced by ESLs has been playing around trying to
wind
better transformers for the ESL57 panels he has also altered to
apparently get far better sensitivity.
Getting a flat response, or one that is acceptable to the ears is taking some
time.....
Perhaps its pointless to try to extend the bass response of the base panels.
A sub would be a better solution.
Patrick Turner.
| Quote: |
................. Phil |
|
|
| Back to top |
|
|
|
|
|
|
Profile
Log in to check your private messages
Log in