[Norman Simmington]

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

Please let me know if further information is required.

This is not an audiophile type application, but a driver for a lab
frequency generator. Only the most basic type of design is needed. No
gain control, filtering or OT required. It will be wired directly to
the 350 ohm load, via a series resistance if necessary.

The designs I have seen in books are music-oriented and seem overly
complex for this. Finding the correctly specified tube is another
issue. I know I could use transistors, but tubes are part of the
experiment..

Hopefully, someone here can guide me in the right direction.

Thank you,

Norman Simmington

[Chris Hornbeck]

On Sun, 20 May 2007 09:16:49 +1000, Norman Simmington
wrote:

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

This is not an audiophile type application, but a driver for a lab
frequency generator. Only the most basic type of design is needed. No
gain control, filtering or OT required. It will be wired directly to
the 350 ohm load, via a series resistance if necessary.

You've got your work cut out for ya. The required 237
volts peak-to-peak from a 300 volt supply implies some
pretty high perveance.

You might also want to think about an output coupling
capacitor and DC drain.

Multiple voltage-regulator triodes would be a conventional
approach. Maybe a search on "OTL amplifiers" or somesuch
might provide inspiration.

All good fortune,

Chris Hornbeck

[Ernst]

On May 19, 6:16 pm, Norman Simmington
wrote:
Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

That will be quite the trick to get that kind of bandwidth and power
output considering the low load impedance. Typically, regulator tubes
(i.e. 6080, 6082, 6AS7) have been used in an OTL topology to directly
drive low impedance loads; however, this requires several tubes in
parallel and requires a bipolar power supply. Here is an example:
http://www.triodeel.com/rcaotl.gif.

Probably more complex than what you are seeking.

Best of luck,

Ernst

[Phil Allison]

"Norman Simmington"

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

Please let me know if further information is required.


** OK.

1. What exactly is this mysterious 350 ohm load ?

2. Is the 350 ohm resistive or not?

3. How is the sine wave being modulated?



This is not an audiophile type application, but a driver for a lab
frequency generator. Only the most basic type of design is needed. No
gain control, filtering or OT required.


** Having no OT creates a big problem - it's not a simplification.


It will be wired directly to
the 350 ohm load, via a series resistance if necessary.


** So the 350 ohm load IS reactive ??


The designs I have seen in books are music-oriented and seem overly
complex for this.


** Only seems that way to a tube novice.


Finding the correctly specified tube is another issue.


** So is MUST be one tube only?

Can't possibly have two tubes in parallel ?

So far, what you have posted look like yet another bonkers, dangerous
crackpot scheme.

Go ahead - prove me wrong.

Explain what this is all about.




......... Phil

[Chris Hornbeck]

On 19 May 2007 16:58:45 -0700, Ernst wrote:

Output impedance 350 ohms

That will be quite the trick to get that kind of bandwidth and power
output considering the low load impedance.

And we're both considering the "output impedance" spec
to be in error, because it's (literally) impossible to
generate that power with that B+ with that source
impedance.

A typo or a troll; who knows? Or maybe the class project
requires a DC-DC convertor...

All good fortune,

Chris Hornbeck

[Eeyore]

Norman Simmington wrote:

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

The 10Hz - 100kHz bandwidth is going to be a big problem ! Transformers aren't
often that good although you might get it with a toroid but that would have to
be gapped for SET use.

Why do you want a tube design ? It's almost trivially simple to do it with solid
state and you won't have the high standing dissipation of Class-A unless you
want to go down that route for some reason.

Graham

[Patrick Turner]

Norman Simmington wrote:

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

Please let me know if further information is required.

This is not an audiophile type application, but a driver for a lab
frequency generator. Only the most basic type of design is needed. No
gain control, filtering or OT required. It will be wired directly to
the 350 ohm load, via a series resistance if necessary.

The designs I have seen in books are music-oriented and seem overly
complex for this. Finding the correctly specified tube is another
issue. I know I could use transistors, but tubes are part of the
experiment..

Hopefully, someone here can guide me in the right direction.

Thank you,

From what you have said comments following the list above and in the
list above, you
have confused the load and output impedance.

But let me assume the load resistance is 350 ohms, and we don't know
what the output impedance of the
amp needs to be.

We understand that the signal will be an amplitude or frequency
modulated sine wave.

What distortion, THD and IMD, phase distortions, amplitude levels are
acceptable?

What is the wacky woowoo lab experiment you are doing that needs a tube
amp?

A figment of your imagination at work?

You can tell us. Your secret will reamin unsafe with us.


But about 15 paralleled 6BQ5/EL84 in triode and driven by
a 6SN7 with CF buffer will give you what you need with a well designed
OPT
with P:S interleaved, and 1:1 ratio for 350:350 ohms ratio.
Output impedance will be 133 ohms.

Patrick Turner.

Norman Simmington

[Phil Allison]

"Chris PITA Idiot Hornbeck"

And we're both considering the "output impedance" spec
to be in error, because it's (literally) impossible to
generate that power with that B+ with that source
impedance.


** Nonsense.

A choke in the plate cct with load connected across it doubles the available
swing.

The OP's use of the phrase "output impedance" in his wish list is
ambiguous - he likely meant load impedance.


A typo or a troll; who knows?


** A grandiose loony is more like it.



........ Phil

[Patrick Turner]

Chris Hornbeck wrote:

On Sun, 20 May 2007 09:16:49 +1000, Norman Simmington
wrote:

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

This is not an audiophile type application, but a driver for a lab
frequency generator. Only the most basic type of design is needed. No
gain control, filtering or OT required. It will be wired directly to
the 350 ohm load, via a series resistance if necessary.

You've got your work cut out for ya. The required 237
volts peak-to-peak from a 300 volt supply implies some
pretty high perveance.

You might also want to think about an output coupling
capacitor and DC drain.

Multiple voltage-regulator triodes would be a conventional
approach. Maybe a search on "OTL amplifiers" or somesuch
might provide inspiration.

All good fortune,

Chris Hornbeck

Let us say that the specification is vague and unscientific for this
amp.
He says 15 to 20 watts. Well just how many?

But let's assume 17.5 watts, which is 78.26 Vrms into 350 ohms.

This would appear to be easily doable with 15 x 6BQ5 in triode,
and if an OPT isn't needed as I suggested, a dc supply through an anode
choke plus capacitor
coupling anode to the 350 ohm load resistance would work OK.

about 7 x EL34 or 5 x KT88 would also work OK.

Patrick Turner.

[Chris Hornbeck]

On Sun, 20 May 2007 12:05:56 +1000, "Phil Allison"
wrote:

A choke in the plate cct with load connected across it doubles the available
swing.

The OP's use of the phrase "output impedance" in his wish list is
ambiguous - he likely meant load impedance.

The OP could have meant any durned thing in the whole
wide world. As usual for this kind of posts, the OP is
gone to the winds.

If he (and, let's face it; it's *always* a he) could
properly define the problem, he wouldn't have posted it.

Your recommendation of a plate choke likely bends the
(completely unspoken and mysterious) rules of the
game much less than mine of a DC-DC convertor.

Here's another: a high voltage supply derived from
the (unspecified) filament supply.

Much thanks, as always,

Chris Hornbeck

[Patrick Turner]

Phil Allison wrote:

"Norman Simmington"

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

Please let me know if further information is required.


** OK.

1. What exactly is this mysterious 350 ohm load ?

2. Is the 350 ohm resistive or not?

3. How is the sine wave being modulated?

This is not an audiophile type application, but a driver for a lab
frequency generator. Only the most basic type of design is needed. No
gain control, filtering or OT required.

** Having no OT creates a big problem - it's not a simplification.

It will be wired directly to
the 350 ohm load, via a series resistance if necessary.

** So the 350 ohm load IS reactive ??

The designs I have seen in books are music-oriented and seem overly
complex for this.

** Only seems that way to a tube novice.

Finding the correctly specified tube is another issue.

** So is MUST be one tube only?

Can't possibly have two tubes in parallel ?

So far, what you have posted look like yet another bonkers, dangerous
crackpot scheme.

Go ahead - prove me wrong.

Explain what this is all about.

........ Phil

The OP should fully disclose what he wants to actually do with the
amplifier
he says he needs.

Is it floating with its own 300V battery supply?

With what other gear is it connected to or beside or involved with.
Would there be problems associated with such issues?

Finding ONE tube that puts 78Vrms into 350 ohms and has input voltage =
0.53Vrms
means the gain is 147.

There are no single tubes which would achieve this.

Hence I suggested 15 x EL84 in triode, gain = 15, easily
giving the required power but needing a 6SN7 driver though.
No NFB is required, and the Rout = 133 ohms approx or less,
because if each EL84 has Ra = 2,000 ohms, 15 in parallel give 133 ohms.
The OP makes no mention of the dampoing factor needed,
and with 133 ohms, DF = a lousy 2.6; maybe its OK for his lunatic lab
experiments.

A parafeed arrangement would allow the anode to be capacitor coupled
to the load and if the choke is large and well designed with low shunt C
the BW of 10Hz to over 100kHz is easily found.

If the load is reactive, there cannot be 17.5 watts into the load,
because reactive
loads do not create power. A load such as a capacitor
with reactance = 350 ohms at 1kHz would be 0.454uF.
The tubes could not produce 78Vrms into 0.454uF at say 50kHz,
but could produce smoke.

At 10Hz, current into 0.454uF is miniscule.

Patrick Turner.

[Phil Allison]

"Chris PITA ****wit Hornbeck"



A choke in the plate cct with load connected across it doubles the
available
swing.

The OP's use of the phrase "output impedance" in his wish list is
ambiguous - he likely meant load impedance.

The OP could have meant any durned thing in the whole
wide world. As usual for this kind of posts, the OP is
gone to the winds.

If he (and, let's face it; it's *always* a he) could
properly define the problem, he wouldn't have posted it.

Your recommendation of a plate choke likely bends the
(completely unspoken and mysterious) rules of the
game much less than mine of a DC-DC convertor.


** What asinine ********.

Suitable chokes are cheap, readily available parts.

Ground the B+ & make the drive amp safe to use.


Here's another: a high voltage supply derived from
the (unspecified) filament supply.


** ???????????????

Take a dose of rat bait - ****head.




........ Phil

[Chris Hornbeck]

On Sun, 20 May 2007 02:08:40 GMT, Patrick Turner
wrote:

Let us say that the specification is vague and unscientific for this
amp.
He says 15 to 20 watts. Well just how many?

But let's assume 17.5 watts, which is 78.26 Vrms into 350 ohms.

This is the majorly flakey part of the OP. He specifies
the 350 Ohms as a source impedance, and we all *assume*
something else. And the OP is, as usual, missing from
the scene. Gee; that's never happened before...

and if an OPT isn't needed as I suggested, a dc supply through an anode
choke plus capacitor
coupling anode to the 350 ohm load resistance would work OK.

It's a stupid project, so knowing the rules as set forth by
the originating body are essential. If a plate choke is
allowable, is a plate autotransformer? Etc. Etc.

Still strikes me more like a troll, but, hey, it's a quiet
Saturday night around here.

Much thanks, as always,

Chris Hornbeck

[Phil Allison]

"Chris Idiot Hornbeck"

And the OP is, as usual, missing from
the scene. Gee; that's never happened before...



** It's Sunday morning here in NSW Australia - where the OP lives.

So he could be at church praying for guidance from above .....




......... Phil

[Chris Hornbeck]

On Sun, 20 May 2007 12:36:12 +1000, "Phil Allison"
wrote:

Here's another: a high voltage supply derived from
the (unspecified) filament supply.


** ???????????????

This is a (guessing) game. Unspecified factors are
open-ended fair game. Please don't tell me that you
couldn't pick up on this hint; I'd be terribly
disappointed.

Take a dose of rat bait - ****head.

Maybe later, but thanks for the generous offer,

Chris Hornbeck

[Patrick Turner]

Eeyore wrote:

Norman Simmington wrote:

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

The 10Hz - 100kHz bandwidth is going to be a big problem ! Transformers aren't
often that good although you might get it with a toroid but that would have to
be gapped for SET use.

Graham, getting 10hz to 100kHz with an OPT is child's play and the 65
steps involved are at my website.

That people say all this is difficult is because accountants dominate
design
procedures with words of "NO!".

Toroids would be fine if cap coupled to a paraffed lot of tubes.

But the OP hasn't psecificied all the requirments and circumstances of
what,
where or how the amp will be used.

Why do you want a tube design ? It's almost trivially simple to do it with solid
state and you won't have the high standing dissipation of Class-A unless you
want to go down that route for some reason.

He has stated he does want to go down the path of tube and class A SE.

So he won't be needing your expertise with solid state.


Patrick Turner.
Graham

[Chris Hornbeck]

On Sun, 20 May 2007 12:44:40 +1000, "Phil Allison"
wrote:

** It's Sunday morning here in NSW Australia - where the OP lives.

So he could be at church praying for guidance from above .....

All of which explains a lot. Thanks, as always, for the
lively banter. Been missing ya' around here of late,

Chris Hornbeck

[Phil Allison]

"Chris PITA Autistic ****ing ASS Hornbeck"


Here's another: a high voltage supply derived from
the (unspecified) filament supply.


** ???????????????

This is a (guessing) game.



** No it is NOT - you ****ING ASS !!


Take a GIANT dose of rat bait &

- drop dead ****head.




........ Phil

[Eeyore]

Patrick Turner wrote:

Eeyore wrote:
Norman Simmington wrote:

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

The 10Hz - 100kHz bandwidth is going to be a big problem ! Transformers aren't
often that good although you might get it with a toroid but that would have to
be gapped for SET use.

Graham, getting 10hz to 100kHz with an OPT is child's play

100kHz (and 10Hz) @ full power ?

Graham

[Chris Hornbeck]

On Sun, 20 May 2007 03:53:02 GMT, Eeyore
wrote:

Graham, getting 10hz to 100kHz with an OPT is child's play

100kHz (and 10Hz) @ full power ?

At these (assumed, 'cause the gig is just a wing and
a prayer assumption this go-'round) impedances and
turns ratios, it's quite do-able.

DC in the primary winding exaggerates a lot of parasitic
reactance issues, but they're still just issues.

But!

The mysterious OP *seems* to have excluded output
transformers; to assume otherwise might prematurely
ruin the sport. Such as it is...

Much thanks, as always,

Chris Hornbeck

[Patrick Turner]

Eeyore wrote:

Patrick Turner wrote:

Eeyore wrote:
Norman Simmington wrote:

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

The 10Hz - 100kHz bandwidth is going to be a big problem ! Transformers aren't
often that good although you might get it with a toroid but that would have to
be gapped for SET use.

Graham, getting 10hz to 100kHz with an OPT is child's play

100kHz (and 10Hz) @ full power ?

It can be done, but you have to use sufficient turns per volt,
interleaving, core size, and insulation thickness.

The 300W PP trannies I wound have 20Hz to 270kHz at 300W.
At 1/2 output voltage Fs is 10Hz, at 75 watts, and there is seldom
huge amounts od energy in music at 10Hz, so the saturation at full power
at 20Hz is fine for my amps.
Interleaving is 5P x 6S, PtoS insulation = 0.8mm.

The HF response does not change for PP or SE.

If the same cores were gapped and used for SE
you'd get about 75watts for the same BW, because the ac Bmax
would be half that of the PP tranny at 300W,
and the dc magnetization would be the same number of Tesla.
To get 10Hz, you'd have to halve the applied voltage again so
you'd get 18 watts.

There is scope for further optimisation of the design if done
specifically for SE use,
and P turns could be increased 1.41 times to allow 1.41 times the
applied voltage without sat
thus doubling the rated LF power. This would increase the LL
by twice which would halve the BW from 270kHz to 135kHz, but it would
all still be fine.

The trannies for my 300Watters are 25Kg.
Recent trannies I wound for 50W SET for 845 are about 18Kg.

In the case now being considered, because the Load is only 350 ohms,
Cshunt is a minor concern,
but for low LL the same interleaving is required.

I use ordinary E&I lams.
Toroids can be used, but then there is the gap problem,
so C-cores are better.

Just why anyone would bother to use tubes in a lab
apart from just illustrating cathode emissions and current across
vacuums remains a mystery.

Patrick Turner.



Graham

[Patrick Turner]

Chris Hornbeck wrote:

On Sun, 20 May 2007 03:53:02 GMT, Eeyore
wrote:

Graham, getting 10hz to 100kHz with an OPT is child's play

100kHz (and 10Hz) @ full power ?

At these (assumed, 'cause the gig is just a wing and
a prayer assumption this go-'round) impedances and
turns ratios, it's quite do-able.

DC in the primary winding exaggerates a lot of parasitic
reactance issues, but they're still just issues.

But!

The mysterious OP *seems* to have excluded output
transformers; to assume otherwise might prematurely
ruin the sport. Such as it is...

Much thanks, as always,

Chris Hornbeck

Our careless sloppy original poster
says 10-100kHz. maybe he does mean 10kHz to 100kHz.

Who knows what he means; maybe he wants to do some supersonic heating or
excitation of some other gear or chemical process.

But the tranny could be air cored for this range of F,
or ferrite cored, but if OPTs are forbidden then
gapped ferrite core with cap coupling to the load seems the easiest.

Patrick Turner.

[Eeyore]

Patrick Turner wrote:

Eeyore wrote:
Patrick Turner wrote:

Graham, getting 10hz to 100kHz with an OPT is child's play

100kHz (and 10Hz) @ full power ?

It can be done, but you have to use sufficient turns per volt,
interleaving, core size, and insulation thickness.

The 300W PP trannies I wound have 20Hz to 270kHz at 300W.
At 1/2 output voltage Fs is 10Hz, at 75 watts, and there is seldom
huge amounts od energy in music at 10Hz, so the saturation at full power
at 20Hz is fine for my amps.

Indeed for audio but he wanted some 'lab' amplifier which is why I mention it.

Graham

[Eeyore]

Patrick Turner wrote:

if OPTs are forbidden then gapped ferrite core with cap coupling to the load
seems the easiest.

I have no idea how anyone got the idea that OPTs are forbidden, although the
required voltage swing of about 230V meams that it would likely be a 1:1.

Graham

[Phil Allison]

"Eeysore ****wit has wanked itself totally Blind"


Patrick Turner wrote:

if OPTs are forbidden then gapped ferrite core with cap coupling to the
load
seems the easiest.

I have no idea how anyone got the idea that OPTs are forbidden,


** Quote from the initial post:

" Only the most basic type of design is needed. No
gain control, filtering or OT required. "


I seriously doubt the OP is practising Scientologist alluding to " Operating
Thetans " ~!~!!


LOL ............




........ Phil

[Eeyore]

Phil Allison wrote:

"Eeysore ****wit has wanked itself totally Blind"

Patrick Turner wrote:

if OPTs are forbidden then gapped ferrite core with cap coupling to the
load seems the easiest.

I have no idea how anyone got the idea that OPTs are forbidden,

** Quote from the initial post:

" Only the most basic type of design is needed. No
gain control, filtering or OT required. "

That'll be fun driving 350 ohms direct from the plate.!

Graham

[Eeyore]

Eeyore wrote:

Phil Allison wrote:
"Eeysore ****wit has wanked itself totally Blind"
Patrick Turner wrote:

if OPTs are forbidden then gapped ferrite core with cap coupling to the
load seems the easiest.

I have no idea how anyone got the idea that OPTs are forbidden,

** Quote from the initial post:

" Only the most basic type of design is needed. No
gain control, filtering or OT required. "

That'll be fun driving 350 ohms direct from the plate.!

Actually, how would a cathode follower fare here with a decent size tube ? The
voltage swing is rather large but maybe 2 of them operating in anti-phase ?

Graham

[Patrick Turner]

Eeyore wrote:

Patrick Turner wrote:

if OPTs are forbidden then gapped ferrite core with cap coupling to the load
seems the easiest.

I have no idea how anyone got the idea that OPTs are forbidden, although the
required voltage swing of about 230V meams that it would likely be a 1:1.

230V? if pk to pk, maybe, that's 81 Vrms.

The load is 350 ohms.

With 15 x EL84 in triode, each triode sees 15 x 350 = 5.25kohms if
parafed by choke
and the anodes are cap coupled to 350 ohms.
Each triode biased minimally could make about 100Vrms into 5.25k = 1.9
watts
with Pda = 9.5 watts each, so 15 would make 28 watts easily,
but we don't need to have Pda as high as this.


For 18 watts into 350 ohms, we need approx 80Vrms, as you said, which is
an Ia change of 16mArms,
or +/- 22mA peak, so bias I would be about 30mA for low THD.
The 15 tubes need Ia total of 450mA, which means wire size has to be
large,
and choke core large, but then because it isn't an OPT which has 1/2 the
window occupied by sec windings,
it's easy to make a choke for this app, and reactance of the choke needs
to be 350 ohms at 5Hz,
so L = 11H, and this is quite easy to do. Two 6H chokes in series would
even be better.

My rough calcs give me a choke with 5,000 turns of 0.5mm dia wire around
a
50mm stack of 50mm tongue GOSS material.
The gap needs critical adjustment, but can be calculated more precisely
by reference to the simple steps at my website.
The choke heat dissipation could be up to around 10 watts,
so wire resistance needs to be below 100 ohms....

More than enough L I would guess, and perhaps the stack could be
reduced, hence also dcr reduced for less
Pd....

A toroidal wide band ac un-gapped OPT can be cap coupled off the anodes
if voltage reduction is needed but
otherwise cap coupling is is fine by 100uF from anode to the load of 350
ohms.

The use of the 15 x EL84 seems excessive, but it makes a very linear
triode
of gm 0.15A/V, Ra = 133 max ohms, µ = 20.
Total Pda = 180 watts easily,
and the heater power is 12A x 6.3V, or 4A x 19V.
Alternatively, a pair of trioded 13E1 could be used, but the gain is
much less.

I would add that such an amp would be a very nice music amplifier.





Graham

[Patrick Turner]

Eeyore wrote:

Eeyore wrote:

Phil Allison wrote:
"Eeysore ****wit has wanked itself totally Blind"
Patrick Turner wrote:

if OPTs are forbidden then gapped ferrite core with cap coupling to the
load seems the easiest.

I have no idea how anyone got the idea that OPTs are forbidden,

** Quote from the initial post:

" Only the most basic type of design is needed. No
gain control, filtering or OT required. "

That'll be fun driving 350 ohms direct from the plate.!

Actually, how would a cathode follower fare here with a decent size tube ? The
voltage swing is rather large but maybe 2 of them operating in anti-phase ?

Graham

The alternative arrangement for cathode follower would have all
the previously mentioned parafeed arrangement between cathode and 0V,
and the choke dcr
would perhaps give the wanted bias voltage for the grid.
Since gain is about 15, some 86Vrms drive to grid is needed to make
80Vrms at the cathode,
so one needs a pair of 6SN7, one with a choke in its anode to circuit
get a large easy low thd swing.
Rout then becomes 1/gm = 1/0.15 = 6.7 ohms for the 15 x EL84 I
previously suggested,
but the load for the BW still needs to be 350 ohms.
THe CF gives a good DF of 53, not bad for an SET without global loop
NFB.

BW at 15 watts would also be below 10Hz to maybe 1MHz.

Using the EL84 in pentode with choke feed to the screens which are
bypassed to the cathode
would be more efficient, and 8 tubes count could be enough.

The excess performance isn't needed.

Patrick Turner.

[robert casey]

Norman Simmington wrote:
Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

Please let me know if further information is required.

This is not an audiophile type application, but a driver for a lab
frequency generator. Only the most basic type of design is needed. No
gain control, filtering or OT required. It will be wired directly to
the 350 ohm load, via a series resistance if necessary.

Is it possible to have the load operating at B+ and plate voltage? Or
is one end tied to ground? Safety issues if the load is at B+ voltage?

If you can safely have the load at B+ voltage, you could connect the
output plate(s) of the output tube(s) directly to the other end of the
load. But this assumes that the load won't mind a DC component creating
extra heat in it.

Or a cathode follower circuit designed to have an impedance of 350 ohms
connected to the load, the other end of the load grounded. Again,
there's a DC component warming the load. But safer.

Or the load is only to see the signal without any DC, then a coupling
cap would be needed. One big enough to avoid significant phase errors
on the signal (you'll have to decide how much phase error you can have).
The cap will block the DC, so a tube with a 350 ohm load from its plate
to B+, and a cap coupling the plate to the 350 ohm load may do what you
need. (Oh, the plate load resistor would need to be a big larger than
350 ohms, as the tube's "plate resistance" would look to be a parallel
resistor. Plate resistance values are up in the tens of thousands of
ohms, so it's not a biggie).

You'll need several output tubes in parallel, because the single ended
operation would have current peaks around 2/3 amp, and most power tubes
can do around 100ma at most.

[Arny Krueger]

"Eeyore" wrote in
message
Phil Allison wrote:

"Eeysore ****wit has wanked itself totally Blind"

Patrick Turner wrote:

if OPTs are forbidden then gapped ferrite core with
cap coupling to the load seems the easiest.

I have no idea how anyone got the idea that OPTs are
forbidden,

** Quote from the initial post:

" Only the most basic type of design is needed. No
gain control, filtering or OT required. "

That'll be fun driving 350 ohms direct from the plate.!

Just run about 90 tubes in parallel to get down to 4 ohms.

[Patrick Turner]

Arny Krueger wrote:

"Eeyore" wrote in
message
Phil Allison wrote:

"Eeysore ****wit has wanked itself totally Blind"

Patrick Turner wrote:

if OPTs are forbidden then gapped ferrite core with
cap coupling to the load seems the easiest.

I have no idea how anyone got the idea that OPTs are
forbidden,

** Quote from the initial post:

" Only the most basic type of design is needed. No
gain control, filtering or OT required. "

That'll be fun driving 350 ohms direct from the plate.!

Just run about 90 tubes in parallel to get down to 4 ohms.

You could call your new proposed model the Kroogamplifier MkI.

There's something eclectically esoteric for you eh what.

But it'd give 225 watts class A triode, or 450 watts class A pentoad.

But either way heater power = 450 watts.

( and no cheatin with PP to get 1,000 watts...)

Just great for winter music.

Since triode THD reduces towards zero from a maximum 5% at 225W,
at a civilised 0.22 watt level the THD would be quite low.

Patrick Turner.

[Iain Churches]

"Eeyore" wrote in message
...


Norman Simmington wrote:

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

The 10Hz - 100kHz bandwidth is going to be a big problem ! Transformers
aren't
often that good although you might get it with a toroid but that would
have to
be gapped for SET use.



Errm. I have an original Williamson with bandwidth in excess of 200 kHz.
It was built nearly sixty years ago.

Maybe it's time for you to try some of the better winders, Graham?

Regards
Iain

[Patrick Turner]

Iain Churches wrote:

"Eeyore" wrote in message
...


Norman Simmington wrote:

Can anyone assist me with a circuit diagram for a single ended output
stage with the following characteristics? This is for a college
physics project. I am fully experienced in solid state, but not tubes.
Any expert advice would be sincerely appreciated.

Bandwidth 10-100KHz
Available B+ 300V
Signal input 1.5Vpp
Continuously modulated sinewave
Single ended
Output impedance 350 ohms
Output power 15-20W

The 10Hz - 100kHz bandwidth is going to be a big problem ! Transformers
aren't
often that good although you might get it with a toroid but that would
have to
be gapped for SET use.


Errm. I have an original Williamson with bandwidth in excess of 200 kHz.
It was built nearly sixty years ago.

Maybe it's time for you to try some of the better winders, Graham?

Regards
Iain

The original OPT for Williamson without NFB would battle to make
200kHz loaded with R rated for it and with flat signal at the anodes.

The old data for W amps does show they did make 100kHz, but was
expecting a bit
much, and with 0.22uF, they oscillated badly.

Later versions had better critical damping.

But the best BW was only possible with OPT made by Savage, Partridge
etc, where they actually
used the wisdom by Williamson and not just pay lip service while dumbing
down the ideas like Leak did.

And W amps were PP, not SE, so getting a good bottom end was easier...

Patrick Turner.

[Eeyore]

Iain Churches wrote:

"Eeyore" wrote in message

The 10Hz - 100kHz bandwidth is going to be a big problem ! Transformers
aren't often that good although you might get it with a toroid but that
would
have to be gapped for SET use.


Errm. I have an original Williamson with bandwidth in excess of 200 kHz.
It was built nearly sixty years ago.

Full power bandwidth ?

Graham