Date   

Re: 6DJ8 substitute idea - reverse engineering

 

I just use the tubes New Sensor have been providing for years. The Electroharmonix brand is excellent with very consistent and quiet tubes. They aren't expensive as far as tubes go either.

I will some day sell all my NOS tubes and replace the ones I need with newly manufactured ones. Audiophiles will allow me to put in a larger stock. :)

-Chris


Re: 6DJ8 substitute idea - reverse engineering

Grayson Evans
 

Hello Keith,
Just saw this thread on the 6DJ8.
I have several sitting in my NOS stock box and would be happy to send you a couple if that will help you out.

Grayson, KJ7UM


Re: 6DJ8 substitute idea - reverse engineering

Liam Perkins
 

FWIW: InterFET make a 300V JFET available on Mouser here
https://tinyurl.com/4xwzebu4 at $US15.00 each; but really, this is getting
to be a tailchase. Just buy some current production ECC88, E88CC, 6922,
7308, 6DJ8 from any of two three easily found good vendors and be done with
it.
Liam @ PEARL, Inc.

On Tue, Jun 22, 2021 at 3:11 PM Tom Lee <tomlee@ee.stanford.edu> wrote:

Fetrons were pseudo-cascode pairs (autocorrect insists on converting
“cascode” into “cascade”). As you note, the decline of JFETs makes it
difficult to reproduce a fetron nowadays. Even back in the day,
high-voltage JFETs were very rare beasts. I wouldn’t know where to get one
today, especially if the goal would be to match the high breakdown voltage
rating of a 6DJ8. [In fact, back in the ‘80s, my PhD advisor cascoded a
bipolar transistor with a vacuum tube to solve a tough high-voltage
problem. He could not find a suitable all-semiconductor alternative at the
time.]

I’d stick with vacuum tubes. There is still plenty of stock out there.

—Cheers
Tom

Sent from my iThing, so please forgive typos and brevity.

On Jun 22, 2021, at 2:58 PM, Michael Perkins <terman08@gmail.com> wrote:

Western Electric made solid state replacements for tubes, calling them
fetrons. They were a cascade of two fets, one a high gm fet and one high
voltage fet. It might be harder to do with Jfets in decline, but they
would
last longer than NOS Russian tubes or trying to find working 6DJ8s. Tek
wrote a paper on creating the 6DJ8. If I can find it , I'll send it to
you.
Mike

On Tue, Jun 22, 2021 at 2:32 PM Michael W. Lynch via groups.io
<mlynch003=
yahoo.com@groups.io> wrote:

On Tue, Jun 22, 2021 at 03:04 PM, David Holland wrote:


When the definition of perfection is the shape of a plate, or its
color,
or
the color of the pins, or even the factory of manufacture, it's not
really
all that hard to find perfection in old 6DJ8's.
David,

Exactly my point, more importantly, the tubes in question need to
solidly
meet the basic spec for the TEK gear in which it is to be used. However,
the tube certainly does not need to be the "pinnacle" of physical
characteristics,electrical "perfection" or identical performance. When
more than one tube of a type is used, a "Perfect" match between all
tubes
is also not generally needed, unless the schematic calls for a close
match
of some certain specification. I do not have a dog in the tube "hunt"
as
most of what I work on has those three legged bugs and little black
caterpillars or centipedes. I just know that when I see a $5 USD tube
selling for $100 or more for a single tube, based on some mystical
combination of characteristics, even a solid state guy such as myself
begins searching around for the "snake oil salesman"; especially when it
comes to test equipment. It will be interesting to hear the results of
your purchase and how these tubes compare to the 6DJ6.

--
Michael Lynch
Dardanelle, AR













Re: 6DJ8 substitute idea - reverse engineering

Tom Lee
 

Fetrons were pseudo-cascode pairs (autocorrect insists on converting “cascode” into “cascade”). As you note, the decline of JFETs makes it difficult to reproduce a fetron nowadays. Even back in the day, high-voltage JFETs were very rare beasts. I wouldn’t know where to get one today, especially if the goal would be to match the high breakdown voltage rating of a 6DJ8. [In fact, back in the ‘80s, my PhD advisor cascoded a bipolar transistor with a vacuum tube to solve a tough high-voltage problem. He could not find a suitable all-semiconductor alternative at the time.]

I’d stick with vacuum tubes. There is still plenty of stock out there.

—Cheers
Tom

Sent from my iThing, so please forgive typos and brevity.

On Jun 22, 2021, at 2:58 PM, Michael Perkins <terman08@gmail.com> wrote:

Western Electric made solid state replacements for tubes, calling them
fetrons. They were a cascade of two fets, one a high gm fet and one high
voltage fet. It might be harder to do with Jfets in decline, but they would
last longer than NOS Russian tubes or trying to find working 6DJ8s. Tek
wrote a paper on creating the 6DJ8. If I can find it , I'll send it to you.
Mike

On Tue, Jun 22, 2021 at 2:32 PM Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

On Tue, Jun 22, 2021 at 03:04 PM, David Holland wrote:


When the definition of perfection is the shape of a plate, or its color,
or
the color of the pins, or even the factory of manufacture, it's not
really
all that hard to find perfection in old 6DJ8's.
David,

Exactly my point, more importantly, the tubes in question need to solidly
meet the basic spec for the TEK gear in which it is to be used. However,
the tube certainly does not need to be the "pinnacle" of physical
characteristics,electrical "perfection" or identical performance. When
more than one tube of a type is used, a "Perfect" match between all tubes
is also not generally needed, unless the schematic calls for a close match
of some certain specification. I do not have a dog in the tube "hunt" as
most of what I work on has those three legged bugs and little black
caterpillars or centipedes. I just know that when I see a $5 USD tube
selling for $100 or more for a single tube, based on some mystical
combination of characteristics, even a solid state guy such as myself
begins searching around for the "snake oil salesman"; especially when it
comes to test equipment. It will be interesting to hear the results of
your purchase and how these tubes compare to the 6DJ6.

--
Michael Lynch
Dardanelle, AR









Re: 6DJ8 substitute idea - reverse engineering

Michael Perkins
 

They were actually a cascode, not a cascade.
Mike

On Tue, Jun 22, 2021 at 2:58 PM Michael Perkins via groups.io <terman08=
gmail.com@groups.io> wrote:

Western Electric made solid state replacements for tubes, calling them
fetrons. They were a cascade of two fets, one a high gm fet and one high
voltage fet. It might be harder to do with Jfets in decline, but they would
last longer than NOS Russian tubes or trying to find working 6DJ8s. Tek
wrote a paper on creating the 6DJ8. If I can find it , I'll send it to you.
Mike

On Tue, Jun 22, 2021 at 2:32 PM Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

On Tue, Jun 22, 2021 at 03:04 PM, David Holland wrote:


When the definition of perfection is the shape of a plate, or its
color,
or
the color of the pins, or even the factory of manufacture, it's not
really
all that hard to find perfection in old 6DJ8's.
David,

Exactly my point, more importantly, the tubes in question need to solidly
meet the basic spec for the TEK gear in which it is to be used. However,
the tube certainly does not need to be the "pinnacle" of physical
characteristics,electrical "perfection" or identical performance. When
more than one tube of a type is used, a "Perfect" match between all tubes
is also not generally needed, unless the schematic calls for a close
match
of some certain specification. I do not have a dog in the tube "hunt" as
most of what I work on has those three legged bugs and little black
caterpillars or centipedes. I just know that when I see a $5 USD tube
selling for $100 or more for a single tube, based on some mystical
combination of characteristics, even a solid state guy such as myself
begins searching around for the "snake oil salesman"; especially when it
comes to test equipment. It will be interesting to hear the results of
your purchase and how these tubes compare to the 6DJ6.

--
Michael Lynch
Dardanelle, AR










Re: 6DJ8 substitute idea - reverse engineering

Michael Perkins
 

Western Electric made solid state replacements for tubes, calling them
fetrons. They were a cascade of two fets, one a high gm fet and one high
voltage fet. It might be harder to do with Jfets in decline, but they would
last longer than NOS Russian tubes or trying to find working 6DJ8s. Tek
wrote a paper on creating the 6DJ8. If I can find it , I'll send it to you.
Mike

On Tue, Jun 22, 2021 at 2:32 PM Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

On Tue, Jun 22, 2021 at 03:04 PM, David Holland wrote:


When the definition of perfection is the shape of a plate, or its color,
or
the color of the pins, or even the factory of manufacture, it's not
really
all that hard to find perfection in old 6DJ8's.
David,

Exactly my point, more importantly, the tubes in question need to solidly
meet the basic spec for the TEK gear in which it is to be used. However,
the tube certainly does not need to be the "pinnacle" of physical
characteristics,electrical "perfection" or identical performance. When
more than one tube of a type is used, a "Perfect" match between all tubes
is also not generally needed, unless the schematic calls for a close match
of some certain specification. I do not have a dog in the tube "hunt" as
most of what I work on has those three legged bugs and little black
caterpillars or centipedes. I just know that when I see a $5 USD tube
selling for $100 or more for a single tube, based on some mystical
combination of characteristics, even a solid state guy such as myself
begins searching around for the "snake oil salesman"; especially when it
comes to test equipment. It will be interesting to hear the results of
your purchase and how these tubes compare to the 6DJ6.

--
Michael Lynch
Dardanelle, AR






Photo Notifications #photo-notice

TekScopes@groups.io Notification <noreply@...>
 

Shaun M <shaun_merrigan@...> added the album 6N23P and 6DJ8 Plate Curves: Plate Curves of the 6N23P (weak and strong examples), vintage Tungsram ECC88 (6DJ8) and Shuguang 6N11. These were run on a uTracer 3+ curve tracer using identical test parameters.


The following photos have been uploaded to the 6N23P and 6DJ8 Plate Curves album of the TekScopes@groups.io group.

By: Shaun M <shaun_merrigan@...>


The following photos have been uploaded to the 6N23P and 6DJ8 Plate Curves album of the TekScopes@groups.io group.

By: Shaun M <shaun_merrigan@...>


Re: 6DJ8 substitute idea - reverse engineering

Michael W. Lynch
 

On Tue, Jun 22, 2021 at 03:04 PM, David Holland wrote:


When the definition of perfection is the shape of a plate, or its color, or
the color of the pins, or even the factory of manufacture, it's not really
all that hard to find perfection in old 6DJ8's.
David,

Exactly my point, more importantly, the tubes in question need to solidly meet the basic spec for the TEK gear in which it is to be used. However, the tube certainly does not need to be the "pinnacle" of physical characteristics,electrical "perfection" or identical performance. When more than one tube of a type is used, a "Perfect" match between all tubes is also not generally needed, unless the schematic calls for a close match of some certain specification. I do not have a dog in the tube "hunt" as most of what I work on has those three legged bugs and little black caterpillars or centipedes. I just know that when I see a $5 USD tube selling for $100 or more for a single tube, based on some mystical combination of characteristics, even a solid state guy such as myself begins searching around for the "snake oil salesman"; especially when it comes to test equipment. It will be interesting to hear the results of your purchase and how these tubes compare to the 6DJ6.

--
Michael Lynch
Dardanelle, AR


Re: 6DJ8 substitute idea - reverse engineering

David Holland
 

When the definition of perfection is the shape of a plate, or its color, or
the color of the pins, or even the factory of manufacture, it's not really
all that hard to find perfection in old 6DJ8's. :-/

I did buy a small pile of "tested" 6N23P's based on this conversation, as I
do have use for a few 6DJ8's in my 560 series scope plugins.

In a couple of months when they arrive, I'll try to remember to put them on
my UTracer3, and see how they compare to some actual 6DJ8's. It'll also
let me know if they're just flat dead. ( I guess technically they may have
been tested - and found dead - sold anyways. :-) )

David

On Tue, Jun 22, 2021 at 3:48 PM Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

It needs to be remembered that TEKTRONIX built their equipment to operate
within specifications while using various tubes and transistors which fell
into a relatively broad envelope of performance. It is absolutely not
necessary to use tubes or transistors on the very bleeding edge of their
performance specifications. One needs components that are in spec, but
certainly NOT (in most cases) perfection. This is the insanity of those
who scavenge tubes from old scopes, these tubes are very likely NOT going
to pass ALL of the various and rigorous inspections that "audiophooles"
seem to require. Sorting through 50-70 year old USED tubes and hoping to
find perfection is one definition of insanity. It would be something akin
to sorting through each and every transistor in a circuit, then attempting
to get every one of the same type to be matched in every respect; simply
impossible. TEK knew that these scopes would run for many thousands of
hours and that if any semblance of reasonable accuracy was to be
maintained, there have to be considerable latitude in the performance
requirements of the individual components. Otherwise the instruments would
require calibration before every use as these components will change ever
so slightly during each use.

--
Michael Lynch
Dardanelle, AR






Re: 6DJ8 substitute idea - reverse engineering

Stephen
 

That makes a lot of sense, Michael.


Re: 6DJ8 substitute idea - reverse engineering

Michael W. Lynch
 

It needs to be remembered that TEKTRONIX built their equipment to operate within specifications while using various tubes and transistors which fell into a relatively broad envelope of performance. It is absolutely not necessary to use tubes or transistors on the very bleeding edge of their performance specifications. One needs components that are in spec, but certainly NOT (in most cases) perfection. This is the insanity of those who scavenge tubes from old scopes, these tubes are very likely NOT going to pass ALL of the various and rigorous inspections that "audiophooles" seem to require. Sorting through 50-70 year old USED tubes and hoping to find perfection is one definition of insanity. It would be something akin to sorting through each and every transistor in a circuit, then attempting to get every one of the same type to be matched in every respect; simply impossible. TEK knew that these scopes would run for many thousands of hours and that if any semblance of reasonable accuracy was to be maintained, there have to be considerable latitude in the performance requirements of the individual components. Otherwise the instruments would require calibration before every use as these components will change ever so slightly during each use.

--
Michael Lynch
Dardanelle, AR


Re: 6DJ8 substitute idea - reverse engineering

Stephen
 

I bought a brand new 6DJ8/ECC88 from Conrad Electronics to replace the one in my 067-0502-01. There are modern productions of this tube.


Re: 6DJ8 substitute idea - reverse engineering

Keith
 

Well, as the OP, I see that I've opened up quite a discussion.

So, in summary (so far) it looks like we've established for future readers and scope restorers the following;

1.) 6DJ8 are scarce and expensive, but not quite unobtanium, yet... as of June 2021.
2.) Russian 6N23P tubes have some definite applicability, but may suffer from reliability and lead time issues. Compensations are required, and available - for a price
3.) 6BQ7 is a viable option, but only with caution in every case since Tek may have pushed particular applications beyond 6BQ7 safe operating zones (Thanks Prof. Lee!)
4.) Some guys on this forum have lots of ECC88 just wasting away in THAT box, and doggone it they haven't sold them to me yet. :-)
5.) I reckon I need to see the little woman about an advance on my tube purchase budget, re; item #4

Future readers and scope builders of tomorrow, we salute you!


Re: 7S12 problem

Richard Steedman
 

I've had the same issue with the case of that pot breaking. It was relatively straightforward to glue back together. The pot is now a little noisy but not bad enough to bother me.


Re: 6DJ8 substitute idea - reverse engineering

Liam Perkins
 

I might be in the minority but knowing the guys at EuroTube and having
dealt with them I know their prices are rational and if section-to-section
curve tracer matching is important, as opposed to single-data-point
"matching", they can meet the need and as much to the point; they know the
difference between the two. While noise grading almost certainly isn't
needed for scope work, the fact that they can do it and get it right is
noteworthy.

On Tue, Jun 22, 2021 at 12:06 AM Tom Lee <tomlee@ee.stanford.edu> wrote:

My guess is that you are in the minority in declaring such a move
"neither productive nor necessary." Given the absurd cost of most 6DJ8s,
using 6BQ7s is often the only economically practical route to repair (or
restoring to service a found carcass that has been stripped of its
tubes). Scopes don't care about the things that matter to audiophiles
(e.g., "noise grading"). Spending hundreds of dollars for a set of tubes
to restore a scope is likely prohibitive to most. If a 6BQ7 will work in
the application, I fail to see why one shouldn't use it.

--Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 6/21/2021 11:31, Liam Perkins wrote:
Or, and if you need them, you could buy current production parts from
Eurotubes https://www.eurotubes.com/store/pc/E88CC-6922-or-6DJ8-c52.htm
who
can provide them curve tracer, section-to-section matched and noise
graded.
They have the RoeTest curve tracer and an ex-Tek guy, Matt Kamina working
with them. They know what they're doing. Here are a couple links to the
RoeTest site http://roehrentest.de/EnglishInfo.html
http://roehrentest.de/RoeTest2.html.
I wouldn't even consider downgrading to the 6BQ7, such a move is neither
productive nor necessary.
Liam

On Mon, Jun 21, 2021 at 12:10 PM Jeff Dutky <jeff.dutky@gmail.com>
wrote:

Keith wrote:
With any non-Tek prepped NOS 6BQ7 OR current production 6N23P we could
do the very same thing. Burn in and section balance measurements are
not very
difficult to do. You just have to tell yourself the truth when
measurements show
that you have to to reject six of those seemingly "economical" 6N23P.
Allowing for
waste, these seemingly less expensive tubes might suddenly "cost" you
$5
or
more a pop, plus the time wasted burning in and measuring.
Of course you just turn around and resell the "failed" tubes to the
audiophools on eBay for $10 each as "burned in and tested." It wouldn't
be
a lie, or even an exaggeration.

-- Jeff Dutky













Re: 6DJ8 substitute idea - reverse engineering

Tom Lee
 

My guess is that you are in the minority in declaring such a move "neither productive nor necessary." Given the absurd cost of most 6DJ8s, using 6BQ7s is often the only economically practical route to repair (or restoring to service a found carcass that has been stripped of its tubes). Scopes don't care about the things that matter to audiophiles (e.g., "noise grading"). Spending hundreds of dollars for a set of tubes to restore a scope is likely prohibitive to most. If a 6BQ7 will work in the application, I fail to see why one shouldn't use it.

--Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 6/21/2021 11:31, Liam Perkins wrote:
Or, and if you need them, you could buy current production parts from
Eurotubes https://www.eurotubes.com/store/pc/E88CC-6922-or-6DJ8-c52.htm who
can provide them curve tracer, section-to-section matched and noise graded.
They have the RoeTest curve tracer and an ex-Tek guy, Matt Kamina working
with them. They know what they're doing. Here are a couple links to the
RoeTest site http://roehrentest.de/EnglishInfo.html
http://roehrentest.de/RoeTest2.html.
I wouldn't even consider downgrading to the 6BQ7, such a move is neither
productive nor necessary.
Liam

On Mon, Jun 21, 2021 at 12:10 PM Jeff Dutky <jeff.dutky@gmail.com> wrote:

Keith wrote:
With any non-Tek prepped NOS 6BQ7 OR current production 6N23P we could
do the very same thing. Burn in and section balance measurements are
not very
difficult to do. You just have to tell yourself the truth when
measurements show
that you have to to reject six of those seemingly "economical" 6N23P.
Allowing for
waste, these seemingly less expensive tubes might suddenly "cost" you $5
or
more a pop, plus the time wasted burning in and measuring.
Of course you just turn around and resell the "failed" tubes to the
audiophools on eBay for $10 each as "burned in and tested." It wouldn't be
a lie, or even an exaggeration.

-- Jeff Dutky







Re: 7S12 problem

ycui7
 

Mine DC knob was separated into two of press too hard. You can try push the pot back into one piece from the back side.

Worth a try.

Yang Cui

On Mon, Jun 21, 2021, at 7:09 PM, Chris van Lint wrote:
Hi all,

My 7S12 sudden stopped working in a TDR configuration, using a S6 and
S52 plug-in.  I know that the S52 and the S6 are OK, as they have been
tested on a different 7S12.  The problem appears to be in the DC balance
circuitry.  Turning either or both the coarse and fine balance control
knobs does not result in any vertical movement of the trace. The
horizontal trace is there and can be varied in terms of speed.  I wonder
whether somebody who has experience with this beast can give me some
idea where to start looking, as this is quite a daunting bit of gear.

Chris







Re: Type CA module in a 545

Morris Odell
 

You should check the voltage divider from pin 15 of the plugin connector. That pin needs to sink 150 mA at 75 volts in all plugins. The 75 volts is derived from the +100 in the mainframe dropped through a couple of 12.6 volt tube heaters. In many plugins including the CA it is connected to a combination of tube heaters and divider resistors with taps to provide bias for various parts of the circuit. In the CA there are also interconnection networks with the +100 and +225 pins. One of those voltages from the divider network is +37.5 that supplies the suppressors of the output amplifier tubes. If that voltage is not right it could affect the gain of those stages. There are also +69 and +12.5 volt taps that are used for DC balance adjustment and to set operating conditions for other parts of the circuit.

Let us know what you find.

Cheers,

Morris


Re: 6DJ8 substitute idea - reverse engineering

Shaun M
 

Foks,
I have some experience with the Russian 6N23P tubes from various vendors.

To summarize:
Most vendors I have dealt ship from Ukraine and the shipping time can be quite long (up to 8 weeks) to Canada.
I check all the tubes on a curve tracer (uTracer) and run a set of plate curves to sort them.
The quality of the tubes is variable, but I have had many more fair-weak tubes than very good-excellent ones.
I bought the cheaper lot quantity tubes (say 20 tubes) rather than the ones marketed to the audio crowd (date matched, matched quad, etc., etc.)
I have received refunds for tube lots that were really junk and sold as "tested" or "new".
I have used the 6N23P in vintage gear in place of the 6DJ8; in fact, the initial reason I investigated the 6N23P was the HP 428B which uses two of them.

Shaun Merrigan


7S12 problem

Chris van Lint
 

Hi all,

My 7S12 sudden stopped working in a TDR configuration, using a S6 and S52 plug-in.  I know that the S52 and the S6 are OK, as they have been tested on a different 7S12.  The problem appears to be in the DC balance circuitry.  Turning either or both the coarse and fine balance control knobs does not result in any vertical movement of the trace. The horizontal trace is there and can be varied in terms of speed.  I wonder whether somebody who has experience with this beast can give me some idea where to start looking, as this is quite a daunting bit of gear.

Chris

4341 - 4360 of 188003