OK, I've never seen/heard a decent explanation of what goes on with a valve
amp with 4 and 8 ohm taps when it is hooked up to speakers of 6 ohm nominal
impedance.

Instinctively I use the 8 ohm tap but, quite recently, one of the yoofs in
HFW wrote 'simply use which one sounds the loudest' or words to that effect.
I know it doesn't appear to make a lot of odds, but summat must happen
differently in the amp on each of these tappings. I've seen references to
the valves 'working harder' one way and the trannies 'not liking it' another
and so it goes on.

Anybody really know? - I would love to hear the correct explanation....

I won't go into technical details here - somebody else can - but I think the
helpful concept to have is that a transformer can transform down "less" or
"more" according to the output taps (if there are any). Put it on its 15 ohm
tap and it transforms down "less". On the 4 ohm tap it transforms down "more".
So with the same speaker, the "less" the transformer transforms down the louder
it will be, the "more" the transformer transforms down the higher the impedence
the valves will see, so the happier they are in terms of less distortion, but
the quieter they are in terms of output. There is an optimum point where it's
both quite loud and quite undistorted. If it goes louder than that it sounds
woolly, if you go the other way it tightens up the sound, which is initially
satisfying but steadily becomes unnecessary in terms of diminishing returns. If
you had 8 ohm speakers, putting the amp on the 4 ohm tap would be cleaner and
tighter, using the 15 ohm tap would be louder and a bit more ragged. Neither
will damage anything, and frankly you can set it at whatever sounds preferable
to your ears.

=== Andy Evans ===
Visit our Website:- http://www.artsandmedia.com
Audio, music and health pages and interesting links.

Andy Evans wrote:
If
you had 8 ohm speakers, putting the amp on the 4 ohm tap would be cleaner and
tighter, using the 15 ohm tap would be louder and a bit more ragged. Neither
will damage anything, and frankly you can set it at whatever sounds preferable
to your ears.

Not so. I clearly remember seeing an old valve amp go up in smoke when
the wrong tap was used.

Well yeah and I wish my valve amp, and others, had multifilar
secondaries, so you could choose, for example, whether to
connect 2 4-ohm secondaries in series for 8 ohms, or in
parallel for 2 ohms with less voltage but a lower o/p impedance.


Martin
--
M.A.Poyser Tel.: 07967 110890
Manchester, U.K. http://www.fleetie.demon.co.uk

In article ,
Fleetie wrote:
Well yeah and I wish my valve amp, and others, had multifilar
secondaries, so you could choose, for example, whether to
connect 2 4-ohm secondaries in series for 8 ohms, or in
parallel for 2 ohms with less voltage but a lower o/p impedance.

As a rule of thumb, the greater the turns ratio the more inefficient a
transformer becomes.

--
*Corduroy pillows are making headlines.

Dave Plowman London SW
To e-mail, change noise into sound.

On Thu, 4 Nov 2004 19:22:55 -0000, "Keith G"
wrote:

OK, I've never seen/heard a decent explanation of what goes on with a valve
amp with 4 and 8 ohm taps when it is hooked up to speakers of 6 ohm nominal
impedance.

Instinctively I use the 8 ohm tap but, quite recently, one of the yoofs in
HFW wrote 'simply use which one sounds the loudest' or words to that effect.
I know it doesn't appear to make a lot of odds, but summat must happen
differently in the amp on each of these tappings. I've seen references to
the valves 'working harder' one way and the trannies 'not liking it' another
and so it goes on.

Anybody really know? - I would love to hear the correct explanation....

Pretty obvious, really. On the 4 ohm tap, it will have .7 of the
output voltage, 1.4 times the available current, and half the output
impedance. Generally, this should give better results.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

In article , Keith G
wrote:
OK, I've never seen/heard a decent explanation of what goes on with a
valve amp with 4 and 8 ohm taps when it is hooked up to speakers of 6
ohm nominal impedance.

Instinctively I use the 8 ohm tap but, quite recently, one of the yoofs
in HFW wrote 'simply use which one sounds the loudest' or words to that
effect. I know it doesn't appear to make a lot of odds, but summat must
happen differently in the amp on each of these tappings. I've seen
references to the valves 'working harder' one way and the trannies 'not
liking it' another and so it goes on.

Anybody really know? - I would love to hear the correct explanation....

Two basic effects from one cause.

The cause is that the choice of transformer tap/ratio changes the load
'seen' by the plates of the o/p valves. This means that the amount of
current the valves are asked to produce for a given swing in plate voltage
will be changed.

The effects a

1) Using a tap that isn't optimum for your speaker may mean that either the
amp voltage limits, or current limits, at a lower power level at the
speaker. Thus the result is either quieter or more distorted.

2) Change in the level of distortion.

A third effect is possible, This is that an unwise tapping choice may lead
to instability or valve damage.

Can't be specific without knowing a fair bit about both the amp and the
speaker.

The problem for valve amps of 'traditional' designs is that speaker
impedance is very variable. So in general no tap is 'perfect', and you have
to choose one that is the 'best available'.

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

"Andy Evans" wrote in message
...
I won't go into technical details here - somebody else can - but I think
the
helpful concept to have is that a transformer can transform down "less" or
"more" according to the output taps (if there are any). Put it on its 15
ohm
tap and it transforms down "less". On the 4 ohm tap it transforms down
"more".
So with the same speaker, the "less" the transformer transforms down the
louder
it will be,

OK


the "more" the transformer transforms down the higher the impedence
the valves will see, so the happier they are in terms of less distortion,
but
the quieter they are in terms of output.


OK


There is an optimum point where it's
both quite loud and quite undistorted. If it goes louder than that it
sounds
woolly, if you go the other way it tightens up the sound, which is
initially
satisfying but steadily becomes unnecessary in terms of diminishing
returns. If
you had 8 ohm speakers, putting the amp on the 4 ohm tap would be cleaner
and
tighter, using the 15 ohm tap would be louder and a bit more ragged.
Neither
will damage anything, and frankly you can set it at whatever sounds
preferable
to your ears.


OK, so you're really saying, all in all, the 4 ohm tap would be best then?
I'll give it a try and see if I can hear an improvement.

Thanks for that - nice to get information from someone who *knows*.... ;-)

Keith G wrote:




OK, so you're really saying, all in all, the 4 ohm tap would be best then?
I'll give it a try and see if I can hear an improvement.

Thanks for that - nice to get information from someone who *knows*.... ;-)




Only other thing to consider (don't disagree with what Stew has posted)
but using the 4R tap with 8R speakers, will mean the output valves see a
reflected load thats 4 times bigger. Thi sis good in that it will tend
to reduce the distortion, but it will also severly limit to output
power, maybe more than the advantages of using the 4R tap give.

Thats why I originally mentioned reflected loads and load lines.

--
Nick

Keith G wrote:

OK, I've never seen/heard a decent explanation of what goes on with a valve
amp with 4 and 8 ohm taps when it is hooked up to speakers of 6 ohm nominal
impedance.

Instinctively I use the 8 ohm tap but, quite recently, one of the yoofs in
HFW wrote 'simply use which one sounds the loudest' or words to that effect.
I know it doesn't appear to make a lot of odds, but summat must happen
differently in the amp on each of these tappings. I've seen references to
the valves 'working harder' one way and the trannies 'not liking it' another
and so it goes on.

Anybody really know? - I would love to hear the correct explanation....

OK, I will explain it all, but will you ever understand it?

Before using a valve amp, you need to know what the
impedance profile of the speakers.

Speakers have varying impedances, and an 8 ohm speaker
may vary between 4 and 40 ohms, but
have an average of 8 ohms.

The most critical part of the audio band is
between 100 and 1,000 Hz, so if the average in that band is 4 ohms,
then always use the 4 ohm outlet.
A speaker where most of the power between 100 and 1k is 6 ohms
should be connected to the 4 ohm outlet.

( it should be remebered that an OPT is like an electronic gear box,
where high voltage swings with low current swings in the tubes are converted to
low voltage swings and high current swings needed by the speakers, so power is
transfered
from a pair of tubes happy to work with a 5 kOhm load,
but still able to supply energy to a 6 ohm speaker.
If the speaker was connected to the tubes directly, there
soon would be smoke and no music.
It'd be like driving up a real steep mountain in top gear; the engine would
overheat and die,
so revs and torque have to be matched to the load.
Volts and amps also have to be matched betwen devices and load.
And even transistor amps can suffer with incorrect loading, and not
many SS amps like 2 ohms at loud levels for very long.)

It should also be ok to connect any value of speaker impedance between 4 ohms
and above
including ESL speakers to the 4 ohm outlet.
The valves will then see a higher ohmic load value, and they will work
without so much thd and more class A % of the total class AB power is available.

Tube amps should never have loads lower than the output setting, so
4 ohm speakers shouln't be used on the 8 ohm outlet, since
the distortion will be 3 times worse, and even though it seems louder for a
given
volume setting, they may not be any greater power.
When the load connected is less than the outlet rating, the tubes see a lower
than
designed for load, their gain reduces, the applied FB therefore reduces,
the % of class A reduces, hence the tripling of thd at full power,
and moderate increase in thd at low power.

Well designed tube amps won't mind being run flat out with no load at all;
they will have clamping diodes on the OPT to prevent the
primary voltage swing at each anode from swinging more than a peak value
equal to the supply voltage.

Well designed tube amps have been stabilised for unconditional stability,
and despite the increase in open loop gain and increase in effective NFB applied

when no load is connected, the amp should remain quite stable.
Strapping 0.2 uF across the output of poorly designed tube amps
can provoke them to immediately oscillate at some low RF.
But well designe tube amps will take any cap load of any value,
and never oscillate.
Quite a few tube amps are more stable than solid state amps.

Well designed tube amps have RC networks across the 1/2 primaries of the OPT,
and perhaps elsewhere in the circuit to reduce gain and phase shift
where one wants that to occur, at F above 30 kHz.
This way there *is* a load connected to the amp at high frequencies, so gain is
controlled
and the amp does not oscillate at RF.
This is important because many speakers are inductive as F rises, so the speaker

presents no load to the amp at say 100 kHz, which is about where some badly
designed tube amps will decide to oscillate.

Anything less than all this is a **** quality amp.

The un-clamped voltage swing with no load is worst in pentode
or tetrode PP amps, because the tube gain becomes
10 times greater than with the rated load, and it only takes
a few volts of grid swing to cause a couple thousand vrms across the primary
of the OPT.
It happens because as one tube of the pair turns with a +ve going signal and
pulls the 1/2 primary
down to 0V, it then becomes disconnected to the OPT, and the other tube
which has a -ve going grid is still controlling the DC current in that 1/2
primary,
and as the idle DC current is cut off, the voltage in the inductance rises very
high,
maybe 1,500 volts positive above 0V,
and the other side of the transformer goes perhaps a thousand volts negative
below 0V.
Diodes prevent the ends of the OPT going negative, so the swing is prevented,
and the insulation is not stressed so much.
The diodes don't conduct during normal operation.

There has been a lot of discussion on these subjects at rec.audio.tubes.

Patrick Turner.