Reply
 
Thread Tools Display Modes
  #1   Report Post  
John L Stewart John L Stewart is offline
Senior Member
 
Location: Toronto
Posts: 301
Smile What is it?

What is it?
Attached Images
 
  #2   Report Post  
Posted to rec.audio.tubes
mick mick is offline
external usenet poster
 
Posts: 10
Default What is it?

On Wed, 16 Jan 2013 17:25:36 +0000, John L Stewart wrote:

What is it?


+-------------------------------------------------------------------+
|Filename: Cascode Amp B 6 4.jpg |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=323|
+-------------------------------------------------------------------+


Two 6N7 twin triodes in push-pull, with both halves paralleled? About 10W
output.

Just a guess...

Nice handles.


  #3   Report Post  
Posted to rec.audio.tubes
Alex Pogossov Alex Pogossov is offline
external usenet poster
 
Posts: 95
Default What is it?


"mick" wrote in message
eb.com...
On Wed, 16 Jan 2013 17:25:36 +0000, John L Stewart wrote:

What is it?


+-------------------------------------------------------------------+
|Filename: Cascode Amp B 6 4.jpg |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=323|
+-------------------------------------------------------------------+


Two 6N7 twin triodes in push-pull, with both halves paralleled? About 10W
output.

Just a guess...

Nice handles.


These metal tubes can also be early 6V6 or quite common video pentodes 6AG7.
The later were not common in audio though. Why would this device have a gas
discharge voltage regulator?


  #4   Report Post  
Posted to rec.audio.tubes
John Byrns John Byrns is offline
external usenet poster
 
Posts: 1,441
Default What is it?

In article ,
John L Stewart wrote:

What is it?


+-------------------------------------------------------------------+
|Filename: Cascode Amp B 6 4.jpg |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=323|
+-------------------------------------------------------------------+


Considering the source we can be fairly certain that it is an audio amplifier.
The name of the photo strongly suggests that the circuit topology includes a
cascode stage, probably a push pull driver given the tube lineup.

Hopefully we will soon see a schematic along with an explanation of the circuits
merits.

--
Regards,

John Byrns

Surf my web pages at, http://fmamradios.com/
  #5   Report Post  
John L Stewart John L Stewart is offline
Senior Member
 
Location: Toronto
Posts: 301
Smile

Quote:
Originally Posted by Alex Pogossov View Post
"mick" wrote in message
eb.com...
On Wed, 16 Jan 2013 17:25:36 +0000, John L Stewart wrote:

What is it?


+-------------------------------------------------------------------+
|Filename: Cascode Amp B 6 4.jpg |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=323|
+-------------------------------------------------------------------+


Two 6N7 twin triodes in push-pull, with both halves paralleled? About 10W
output.

Just a guess...

Nice handles.


These metal tubes can also be early 6V6 or quite common video pentodes 6AG7.
The later were not common in audio though. Why would this device have a gas
discharge voltage regulator?
Thanx for your replies. I did this one about eight years ago. Than I got another contract so the amp never got properly documented.

Some folks will find this configuration a bit odd. It is intended to be. Long ago I decided that enough had already been done with the popular designs of the day, Williamson, Mullard, Dyna & so on. I’ve pretty much always wanted to try something different. Anyone who has read my many articles in Glass Audio & AudioXpress will be aware of that.

The input section is a differential stage, all connected as a PP cascode. Not a mu-follower as some might see it in error. That way the front end has pentode gain but a triode noise figure. And it can be driven from a differential source if desired. That way the effect of ground loops can be minimized.

The second stage is PP cathode followers driving straight into the control grids of PP 6V6’s. And they do look a lot like 6N7s!

My Electronic Workbench software includes no pentodes. The fix is simple. I’ve used the Voltage Controlled Current Source model paralleled with a resistor to simulate the small signal model of the 6V6. That would be a current source of 4.1 ma/volt. On the schema that shows up as 4.1 mmho. That in parallel with a 50K resistor looks like the 6V6 we find in the RCA tube manuals.

So all the way thru, differential, 6V6s running Class AB2. The use of 4BQ7 & 6BQ7 in audio may surprise some. In earlier work I found them to be damn near as good as 6SL7 & 6SN7 at a ¼ of the cost. Something to think about.

The NFB is over one gain stage only to the upper cascode grids. One RC coupling, so LF stability should be unconditional.

The switch labeled ‘(Space)’ lets the laptop Space Bar switch the load in or out.

I’ve included an OP section biasing arrangement that tracks the supply voltage. In a fixed bias cct the bias will increase at the same rate as the B+. As the supply increases the bias will increase in such a way that the OP tubes will be driven toward cutoff. If the OP bias is taken from a regulated source, then the OP stage current will increase. Something in between might offer an advantage. All determined by the triode mu of the OP tubes.

The PS is on a separate schema, I will post it later.

On a ~350 Volt regulated PS this thing can make 26 Watts. Measured by a MetraHit 29S Precision DMM & Wattmeter. At low level it is running Class A. That makes for a lot of headroom.

Keep in mind this is meant from the start to be an experimental amp, something to try few ideas. Better or not than something else, who knows!

I will post more performance measurements later.

And If I can find it I will post a regulated PS cct using 6N7s I built more than 50 years ago.

Cheers to all, John
Attached Images
 


  #6   Report Post  
jazbo8 jazbo8 is offline
Banned
 
Posts: 4
Default

Very interesting cct, is the load 8k?!

Jaz
  #7   Report Post  
Posted to rec.audio.tubes
Alex Pogossov Alex Pogossov is offline
external usenet poster
 
Posts: 95
Default What is it?


"John L Stewart" wrote in message
...

Alex Pogossov;965477 Wrote:
"mick" wrote in message
eb.com...-
On Wed, 16 Jan 2013 17:25:36 +0000, John L Stewart wrote:
-
What is it?



+-------------------------------------------------------------------+
|Filename: Cascode Amp B 6 4.jpg

|
|Download:

http://www.audiobanter.com/attachment.php?attachmentid=323|

+-------------------------------------------------------------------+-

Two 6N7 twin triodes in push-pull, with both halves paralleled? About

10W
output.

Just a guess...

Nice handles. -


These metal tubes can also be early 6V6 or quite common video pentodes
6AG7.
The later were not common in audio though. Why would this device have a
gas
discharge voltage regulator?


Thanx for your replies. I did this one about eight years ago. Than I got
another contract so the amp never got properly documented.

Some folks will find this configuration a bit odd. It is intended to be.
Long ago I decided that enough had already been done with the popular
designs of the day, Williamson, Mullard, Dyna & so on. I've pretty much
always wanted to try something different. Anyone who has read my many
articles in Glass Audio & AudioXpress will be aware of that.

The input section is a differential stage, all connected as a PP
cascode. Not a mu-follower as some might see it in error. That way the
front end has pentode gain but a triode noise figure. And it can be
driven from a differential source if desired. That way the effect of
ground loops can be minimized.

The second stage is PP cathode followers driving straight into the
control grids of PP 6V6's. And they do look a lot like 6N7s!

My Electronic Workbench software includes no pentodes. The fix is
simple. I've used the Voltage Controlled Current Source model paralleled
with a resistor to simulate the small signal model of the 6V6. That
would be a current source of 4.1 ma/volt. On the schema that shows up as
4.1 mmho. That in parallel with a 50K resistor looks like the 6V6 we
find in the RCA tube manuals.

So all the way thru, differential, 6V6s running Class AB2. The use of
4BQ7 & 6BQ7 in audio may surprise some. In earlier work I found them to
be damn near as good as 6SL7 & 6SN7 at a ¼ of the cost. Something to
think about.


Definitely 6BQ7 is a better cathode follower as it has a higher gm. But
something looks missing in the schematic -- pull-down resistors from the
cathodes of 6BQ7 to a negative rail. perhaps those 120K resistors should go
to -300V rail? Also, 330Vdc on the 6BQ7 plates seems too high.



The NFB is over one gain stage only to the upper cascode grids. One RC
coupling, so LF stability should be unconditional.

The switch labeled '(Space)' lets the laptop Space Bar switch the load
in or out.

I've included an OP section biasing arrangement that tracks the supply
voltage. In a fixed bias cct the bias will increase at the same rate as
the B+. As the supply increases the bias will increase in such a way
that the OP tubes will be driven toward cutoff. If the OP bias is taken
from a regulated source, then the OP stage current will increase.
Something in between might offer an advantage. All determined by the
triode mu of the OP tubes.

The PS is on a separate schema, I will post it later.

On a ~350 Volt regulated PS this thing can make 26 Watts. Measured by a
MetraHit 29S Precision DMM & Wattmeter. At low level it is running Class
A. That makes for a lot of headroom.

Keep in mind this is meant from the start to be an experimental amp,
something to try few ideas. Better or not than something else, who
knows!

I will post more performance measurements later.

And If I can find it I will post a regulated PS cct using 6N7s I built
more than 50 years ago.

Cheers to all, John


+-------------------------------------------------------------------+
|Filename: Experimental Amp 10W 7.jpg |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=324|
+-------------------------------------------------------------------+



--
John L Stewart



  #8   Report Post  
John L Stewart John L Stewart is offline
Senior Member
 
Location: Toronto
Posts: 301
Smile

Quote:
Originally Posted by jazbo8 View Post
Very interesting cct, is the load 8k?!

Jaz
In order to take advantage of Class AB2 the loadline needs to be steeper than Class AB1. Otherwise the load simply bumps up against the diode line on the plate characteristics & no more power is available. I found by experiment this particular hookup worked best with a 6.8 K loadline.

Refer to the 6V6GTA composite plate family. These are for Eg2 of 250 volts, but demonstrate well the problem.

I usually set the loadline steeper than optimum. That is because in a real application the load is not a resistor but rather a loudspeaker. That would put an elliptical loadline on the amplifier, often several times the speaker nominal R.

It has been shewan by some that the steeper loadline also reduces 3H.

Cheers, John
Attached Images
 
  #9   Report Post  
John L Stewart John L Stewart is offline
Senior Member
 
Location: Toronto
Posts: 301
Default

[quote=Alex Pogossov;965492]"John L Stewart" wrote in message
...[color=blue][i]

Alex Pogossov;965477 Wrote:
"mick" wrote in message
eb.com...-
On Wed, 16 Jan 2013 17:25:36 +0000, John L Stewart wrote:
-
What is it?



+-------------------------------------------------------------------+
|Filename: Cascode Amp B 6 4.jpg

|
|Download:

http://www.audiobanter.com/attachment.php?attachmentid=323|

+-------------------------------------------------------------------+-

Two 6N7 twin triodes in push-pull, with both halves paralleled? About

10W
output.

Just a guess...

Nice handles. -


These metal tubes can also be early 6V6 or quite common video pentodes
6AG7.
The later were not common in audio though. Why would this device have a
gas
discharge voltage regulator?



Definitely 6BQ7 is a better cathode follower as it has a higher gm. But
something looks missing in the schematic -- pull-down resistors from the
cathodes of 6BQ7 to a negative rail. perhaps those 120K resistors should go
to -300V rail? Also, 330Vdc on the 6BQ7 plates seems too high.

Alex, you are correct, the 6BQ7 CFs need a pull down. I just looked at the chassis & the pull downs are to -150 volts. That supplied by the gas regulator. I need to correct the schema before it is finished. 330 volts on the BQ7s is pushing it. I simply wanted to see what this thing could do. There are other warts I could & might fix as well!

Those metal 6V6s were in a cache I got from a local broadcaster when they went SS. That size metal configuration also showed up in the 25L6. IBM used a lot of them in mark sense readers in the 50's. The PS was a FWB Selenium. Not many failures though. The B+ rail was 200 volts.

Cheers, John
  #10   Report Post  
jazbo8 jazbo8 is offline
Banned
 
Posts: 4
Default

Quote:
Originally Posted by John L Stewart View Post
In order to take advantage of Class AB2 the loadline needs to be steeper than Class AB1. Otherwise the load simply bumps up against the diode line on the plate characteristics & no more power is available. I found by experiment this particular hookup worked best with a 6.8 K loadline.

Refer to the 6V6GTA composite plate family. These are for Eg2 of 250 volts, but demonstrate well the problem.

I usually set the loadline steeper than optimum. That is because in a real application the load is not a resistor but rather a loudspeaker. That would put an elliptical loadline on the amplifier, often several times the speaker nominal R.

It has been shewan by some that the steeper loadline also reduces 3H.

Cheers, John
That made good sense - so set it "lower" and let the ellipse take it up, thanks!

Jaz


  #11   Report Post  
Posted to rec.audio.tubes
John Byrns John Byrns is offline
external usenet poster
 
Posts: 1,441
Default What is it?

In article ,
John L Stewart wrote:

What is it?


+-------------------------------------------------------------------+
|Filename: Cascode Amp B 6 4.jpg |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=323|
+-------------------------------------------------------------------+


Awesome, 26 Watts from a pair of 6V6s is pretty impressive! Did you measure 26
Watts on the bench, or was it determined from the simulation?

How come you used a 100k resistor and -152 volt power supply in the tail circuit
of the differential cascode voltage amplifier instead of a closer approximation
of a real current source?

Patrick T. would have used a diode current source, and I would have given a
6/4BQ7 a try with both sections paralleled and a large un bypassed cathode
resistor, maybe about 33k, taken to the -152 volt supply.

Speaking of 4BQ7s, what is the point of using a 4BQ7 in the upper part of the
cascode instead of another 6BQ7?

Shouldn't there be resistors from the cathodes of the two 6BQ7 followers to the
-152 volt supply to provide bias to the 6V6s?

Does your simulation model for the 6V6 include grid current?

What is the plate-plate load on the PP 6V6 output stage?

What is the quiescent current of the 6V6 output stage?

While there may only be one RC coupling inside the negative feedback loop, there
are two low frequency poles inside the feedback loop which may prevent low
frequency oscillation, but won't eliminate the possibility of low frequency
peaking without a careful choice of time constants.

--
Regards,

John Byrns

Surf my web pages at, http://fmamradios.com/
  #12   Report Post  
John L Stewart John L Stewart is offline
Senior Member
 
Location: Toronto
Posts: 301
Smile

Quote:
Originally Posted by John Byrns View Post
In article ,
John L Stewart wrote:

What is it?


+-------------------------------------------------------------------+
|Filename: Cascode Amp B 6 4.jpg |
|Download: http://www.audiobanter.com/attachment.php?attachmentid=323|
+-------------------------------------------------------------------+


Awesome, 26 Watts from a pair of 6V6s is pretty impressive! Did you measure 26
Watts on the bench, or was it determined from the simulation?

How come you used a 100k resistor and -152 volt power supply in the tail circuit
of the differential cascode voltage amplifier instead of a closer approximation
of a real current source?

Patrick T. would have used a diode current source, and I would have given a
6/4BQ7 a try with both sections paralleled and a large un bypassed cathode
resistor, maybe about 33k, taken to the -152 volt supply.

Speaking of 4BQ7s, what is the point of using a 4BQ7 in the upper part of the
cascode instead of another 6BQ7?

Shouldn't there be resistors from the cathodes of the two 6BQ7 followers to the
-152 volt supply to provide bias to the 6V6s?

Does your simulation model for the 6V6 include grid current?

What is the plate-plate load on the PP 6V6 output stage?

What is the quiescent current of the 6V6 output stage?

While there may only be one RC coupling inside the negative feedback loop, there
are two low frequency poles inside the feedback loop which may prevent low
frequency oscillation, but won't eliminate the possibility of low frequency
peaking without a careful choice of time constants.

--
Regards,

John Byrns

Surf my web pages at, http://fmamradios.com/
That is real power measured by a precision DMM/Wattmeter. See it at this link- http://www.gossenmetrawatt.com/engli...etrahit29s.htm
I have two of them.

The 100K tail seems to do just fine. I measured the CMR today. At One KHz it is 72 db. The CM Tolerance is more than 17 volts, the limit of my HP200CD Audio SG.

4BQ7 is elevated, so I used the 5V wdg on the PT to drive its heater thru a One R resistor.

The CFs are returned to the -152 volts. There was an error on the schema.

The simulation does not measure grid current. Gotta do that on the model.

The P-P load is 6.8K, all covered in an earlier post.

The 6V6s are running about 35 ma each. The cct could benifit with some DC balancing.

I had a quick look at the LF end. I think the HF end is going to be more of a problem.

So we will see, Regards, John
Attached Images
 
  #13   Report Post  
Posted to rec.audio.tubes
Phil Allison[_3_] Phil Allison[_3_] is offline
external usenet poster
 
Posts: 500
Default What is it?


"John Byrns"

Awesome, 26 Watts from a pair of 6V6s is pretty impressive!


** Fraid it is quite commonplace.

Fender amps like this example from 1963 did the same.

http://www.webphix.com/schematic%20h...a763_schem.pdf

Note the use of -35V grid bias.


.... Phil


  #14   Report Post  
Posted to rec.audio.tubes
John Byrns John Byrns is offline
external usenet poster
 
Posts: 1,441
Default What is it?

In article ,
"Phil Allison" wrote:

"John Byrns"

Awesome, 26 Watts from a pair of 6V6s is pretty impressive!


** Fraid it is quite commonplace.

Fender amps like this example from 1963 did the same.

http://www.webphix.com/schematic%20h...m/fenderamps/d
eluxe_reverb_aa763_schem.pdf

Note the use of -35V grid bias.


I note also the use of 415 volts on the screens!

In any case this amp is a wimp compared with John's amp as the Fender only does
22 Watts.

And John's amp uses real 6V6s while your Fender uses 6V6GTs.

Using 6V6s in this fashion may be "commonplace" in guitar amps however I doubt
that it is all that common in straight audio amps as used in peoples parlors.

Fender pushes the poor little 6V6GTs within an inch of their life running class
AB1, while John uses a more sophisticated class AB2 circuit which is easier on
the poor little tubes.

Not being a Guitar guy, having built only one guitar amp in my entire life, I
find Fenders model numbering/naming scheme somewhat confusing as the model name
and number parts apparently must be considered together to arrive at the correct
schematic. Looking just at the number part, aa763, I find Fender amps using PP
6V6GTs, PP 6L6GCs, and PPP 6L6GCs ranging in power from 22 Watts to 85 Watts.

--
Regards,

John Byrns

Surf my web pages at, http://fmamradios.com/
  #15   Report Post  
John L Stewart John L Stewart is offline
Senior Member
 
Location: Toronto
Posts: 301
Smile

Quote:
Originally Posted by Phil Allison[_3_] View Post
"John Byrns"

Awesome, 26 Watts from a pair of 6V6s is pretty impressive!


** Fraid it is quite commonplace.

Fender amps like this example from 1963 did the same.

http://www.webphix.com/schematic%20h...a763_schem.pdf

Note the use of -35V grid bias.


.... Phil
Thanx for that link Phil. I'm always amazed at how these guitar pickers keep their stuff running with the OP section at these gross overvoltages. That particular amp & many others are in 'The Tube Amp Book' by Aspen Pitman along with 463 pages of other amps. Every page different. My copy is the 4th Edition, got it about 10 years ago. It is a great resource for anyone working on tube or hybrid amps.

JJ claims to have a better 6V6GT version, see it here-

http://frank.pocnet.net/sheets69.html

Has anyone tried JJ?

About 20 years ago I bought some Russian 6L6GCs for use in a PP UL amp running 450 volts B+. They did not last long. The construction was the old kind using the glass press for the leads, not the button base used in the 1950s version. I don't think the vacuum was too good either.

My thoughts anyway, Cheers, John


  #16   Report Post  
Posted to rec.audio.tubes
Phil Allison[_3_] Phil Allison[_3_] is offline
external usenet poster
 
Posts: 500
Default What is it?


"John Byrns"
"Phil Allison"
"John Byrns"

Awesome, 26 Watts from a pair of 6V6s is pretty impressive!


** Fraid it is quite commonplace.

Fender amps like this example from 1963 did the same.

http://www.webphix.com/schematic%20h...m/fenderamps/d
eluxe_reverb_aa763_schem.pdf

Note the use of -35V grid bias.


I note also the use of 415 volts on the screens!

In any case this amp is a wimp compared with John's amp as the Fender only
does
22 Watts.



** They regularly do 24 to 26 watts at the clipping point.

I have a recent example here now( Deluxe Reverb 65) that outputs 25 watts
from a pair of Electro Harmonix 6V6s.


And John's amp uses real 6V6s while your Fender uses 6V6GTs.


** Purest nonsense.


Using 6V6s in this fashion may be "commonplace" in guitar amps however I
doubt
that it is all that common in straight audio amps as used in peoples
parlors.


** Irrelevant to what I posted about.

Not being a Guitar guy, having built only one guitar amp in my entire
life,


** You have not lived.


.... Phil


  #17   Report Post  
Posted to rec.audio.tubes
Phil Allison[_3_] Phil Allison[_3_] is offline
external usenet poster
 
Posts: 500
Default What is it?


"John L Stewart"
'Phil Allison
John Byrns"
-
Awesome, 26 Watts from a pair of 6V6s is pretty impressive!-


** Fraid it is quite commonplace.

Fender amps like this example from 1963 did the same.

http://tinyurl.com/b8hbs62

Note the use of -35V grid bias.



Thanx for that link Phil. I'm always amazed at how these guitar pickers
keep their stuff running with the OP section at these gross
overvoltages.


** The printed numbers are at idle conditions - at full output the HT
falls by about 20%.


JJ claims to have a better 6V6GT version, see it here-

http://frank.pocnet.net/sheets69.html

Has anyone tried JJ?


** Yep.

JJ have cheated - the plates of their " 6V6S " are identical to Sovtek
5881s so the available dissipation is way more. The heater / cathode
structure is modified to give the right numbers for a 6V6.

While this makes the tube ideal for use in amps like Fender Delux Revebs
that really " push the envelope" it sets up nasty a future problem if the
tube ever goes out of production.



..... Phil



Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT +1. The time now is 06:51 PM.

Powered by: vBulletin
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 AudioBanter.com.
The comments are property of their posters.
 

About Us

"It's about Audio and hi-fi"