DIY TD Pulser


Sergey Kubushyn
 

If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1О©╫308О©╫ / О©╫О©╫308О©╫ in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


 

It is interesting how the 2465B or 2467B does not even show a hint of
the overshoot shown by the 7104/7A29. I think the preshoot shown
would drive me crazy but my 2440 produces the same type of results
with very fast edges.

Even if you 7T11/7S11/S4 works, in my experience it takes some
practice to get reliable results. I often install mine in the center
slots and a conventional amplifier and timebase in the outer slots to
use as a sanity check.

On Tue, 14 May 2013 22:38:12 -0700 (PDT), Sergey Kubushyn
<ksi@koi8.net> wrote:

If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1?308? / ??308? in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


Sergey Kubushyn
 

On Wed, 15 May 2013, David wrote:

Yep, those are very interesting pictures :)

As for my 7104/7A29 -- they are still waiting to be properly refurbished and
calibrated. I think I'd be able to get better results when I have them
rebuilt. This particular one is just an old instrument, no service has been
done on it since it was new. But it definitely WAY faster than 2467B :) That
pulse edge looks like 30 ps or so on its screen and who knows if it is
really that slow :) or it is just best that 7104 can do and it is really
faster :)

And 7S11/7T11/S4 are quite new for me, I have never used those so they might
even be working but not all that obvious to use. They require manual reading
and they definitely deserve their own dedicated scope. I have several 7633s
for this purpose but not much spare time...

It is interesting how the 2465B or 2467B does not even show a hint of
the overshoot shown by the 7104/7A29. I think the preshoot shown
would drive me crazy but my 2440 produces the same type of results
with very fast edges.

Even if you 7T11/7S11/S4 works, in my experience it takes some
practice to get reliable results. I often install mine in the center
slots and a conventional amplifier and timebase in the outer slots to
use as a sanity check.

On Tue, 14 May 2013 22:38:12 -0700 (PDT), Sergey Kubushyn
<ksi@koi8.net> wrote:

If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1?308? / ??308? in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************

------------------------------------

Yahoo! Groups Links


---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


 

Any of the 76xx mainframes which include variable persistence are
great for the 7S11/7T11 and usually better than even the 7854 although
the later is useful when you want to minimize noise.

The slope switch on my 7T11 is flakey and I have not completed a full
calibration on it yet. One thing to beware of is that external
triggering is DC coupled and has a limited range so it is often
necessary to use a DC block. The internal triggering is AC coupled.

On Tue, 14 May 2013 23:48:08 -0700 (PDT), Sergey Kubushyn
<ksi@koi8.net> wrote:

On Wed, 15 May 2013, David wrote:

. . .

And 7S11/7T11/S4 are quite new for me, I have never used those so they might
even be working but not all that obvious to use. They require manual reading
and they definitely deserve their own dedicated scope. I have several 7633s
for this purpose but not much spare time...

. . .

Even if you 7T11/7S11/S4 works, in my experience it takes some
practice to get reliable results. I often install mine in the center
slots and a conventional amplifier and timebase in the outer slots to
use as a sanity check.

On Tue, 14 May 2013 22:38:12 -0700 (PDT), Sergey Kubushyn
<ksi@koi8.net> wrote:

. . .

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

. . .


Ladislao Bastetti <lbastetti001@...>
 

Hi Sergey,

can you please elaborate the reason why there must be three resistance instead of just one ? It's still not so clear

Laszlo


2013/5/15 Sergey Kubushyn <ksi@...>

 

If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1И308Д / ГИ308Д in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************



 

Three resistors in series minimize the parasitic capacitance and
spiral cut leaded resistors act as high frequency inductors.

On Wed, 15 May 2013 09:26:24 +0200, Ladislao Bastetti
<lbastetti001@gmail.com> wrote:

Hi Sergey,

can you please elaborate the reason why there must be three resistance
instead of just one ? It's still not so clear

Laszlo


victor.silva
 

Sergey,

Great work! I think it's the first time I've seen someone make a duplicate of Tek's 067-0681-01 TD Pulser using a Russian diode.

I do have one comment. In the 24xx response the waveform should be on the 0% dotted line and the first cursor should be at the 10%/Trace crossing.

--Victor


vdonisa
 

Great work Sergey, thanks for sharing!

--- In TekScopes@yahoogroups.com, Sergey Kubushyn <ksi@...> wrote:

If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1é308ä / çé308ä in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


g_kupka <ghm1@...>
 

Where can I get the 067-0681-01 manual? I have only the TU-5 Pulser.

Gerhard

--- In TekScopes@yahoogroups.com, "vdonisa" <vdonisa@...> wrote:

Great work Sergey, thanks for sharing!

--- In TekScopes@yahoogroups.com, Sergey Kubushyn <ksi@> wrote:

If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1é308ä / çé308ä in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


Sergey Kubushyn
 

On Wed, 15 May 2013, David wrote:

Yep, exactly.

Three resistors in series minimize the parasitic capacitance and
spiral cut leaded resistors act as high frequency inductors.

On Wed, 15 May 2013 09:26:24 +0200, Ladislao Bastetti
<lbastetti001@gmail.com> wrote:

Hi Sergey,

can you please elaborate the reason why there must be three resistance
instead of just one ? It's still not so clear

Laszlo

------------------------------------

Yahoo! Groups Links


---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


Sergey Kubushyn
 

On Wed, 15 May 2013, victor_j_silva wrote:

Yeah, I know but I did not try to make exact measurements, just to
illustrate how fast that pulse is...

Sergey,

Great work! I think it's the first time I've seen someone make a
duplicate of Tek's 067-0681-01 TD Pulser using a Russian diode.

I do have one comment. In the 24xx response the waveform should be on the
0% dotted line and the first cursor should be at the 10%/Trace crossing.

--Victor





------------------------------------

Yahoo! Groups Links


---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


Sergey Kubushyn
 

On Wed, 15 May 2013, g_kupka wrote:

http://w140.com/tekwiki/wiki/067-0681-01

Where can I get the 067-0681-01 manual? I have only the TU-5 Pulser.

Gerhard


--- In TekScopes@yahoogroups.com, "vdonisa" <vdonisa@...> wrote:

Great work Sergey, thanks for sharing!

--- In TekScopes@yahoogroups.com, Sergey Kubushyn <ksi@> wrote:

If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1�308� / ��308� in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


------------------------------------

Yahoo! Groups Links



---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


 

Hello Sergey,
 
That is a real nice piece of work. Probably the fastest pulser put her to date.
 
Comment - you might get a little better waveform out if you shorten the center pin and the ground lug as best as possible. It has a relatively high inductance before launching into the BNC connector.
 
Best regards,
Tom
 
 
 
 
 

----- Original Message -----
Sent: Wednesday, May 15, 2013 1:38 AM
Subject: [TekScopes] DIY TD Pulser

 

If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1И308Д / ГИ308Д in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


Egge Siert
 

--- In TekScopes@yahoogroups.com, "g_kupka" <ghm1@...> wrote:



Where can I get the 067-0681-01 manual? I have only the TU-5 Pulser.

Gerhard
Hi Gerhard,

As far as I know Rens Tel provided you the schematic on 30 July 2012. Your request for it and the reply of Rens Tel is still into the archives. Intriging.

Greetings,

Egge Siert


Sergey Kubushyn
 

On Wed, 15 May 2013, Tom Miller wrote:

Yep, I know. If someone wanted to squeeze the last picosecond out of this
that TD->R7 structure should've been made coaxial with the BNC connector
with R7 butt-soldered to a short central pin stub and TD Anode butt-soldered
to R7 other terminal. TD Cathode should be soldered to something like a
spider with multiple connections to BNC ground evenly spread around the
entire circumference to reduce parasitic inductance to the ground. But this
is not one-evening project any more and huge overkill for the intended
purpose :) It is more than adequate for the task in its present form and
very easy and quick to build.

BTW, I can see that all those TDs on Ebay are already sold out. For those
who want to build that pulser -- don't you worry guys, that guy probably
have more of those and will relist that item as he did before. He did sold
10 of those some time ago and relisted another 10 that are sold out by now.

You can send him a message so he would relist them sooner than later. I'm
not 100% sure he has more of those but he probably has. Those were abundant
in Russia and nobody seems to have any use for them so they should be still
available.

Hello Sergey,

That is a real nice piece of work. Probably the fastest pulser put her to
date.

Comment - you might get a little better waveform out if you shorten the
center pin and the ground lug as best as possible. It has a relatively
high inductance before launching into the BNC connector.

Best regards,
Tom





----- Original Message -----
From: Sergey Kubushyn
To: TekScopes@yahoogroups.com
Sent: Wednesday, May 15, 2013 1:38 AM
Subject: [TekScopes] DIY TD Pulser



If somebody interested...

Here is DIY Tektronix 067-0681-01 pulser I had to make in order to calibrate
my 2467Bs and 2465Bs. Took a whole evening to build :)

Works like a charm, I'm pretty impressed. I used NOS Russian tunnel diode,
1I308D (aka GI308D or 1О©╫308О©╫ / О©╫О©╫308О©╫ in Cyrillic.) 1 or G stands for
Germanium, 1 is military rated device, G is industrial. Those are still
available on Ebay (no association with the Seller but that is the guy I
bought a couple of GI308D TDs from):

http://www.ebay.com/itm/190769968180

If someone wanted to make his own pulser make sure the last letter is _D_
(it MUST be 1I308_D_, devices with other letters either won't work or will
have poor performance.)

I uploaded several photos in "TD Pulser" album, everybody is welcome to take
a look. There are enough pictures there to make this nice little gem. There
are also 3 photos of the generated pulses, one on 2467B limited by its
bandwidth and 2 on way better 7104 with 7A29/7B10/7B15. I somehow failed to
get my 7S11/7T11 with S4 head to work (they were purchased as "For Parts
Only" so might be not working and require some job to revive them) so I can
not tell the exact rise time but judging from 7104 display it should be way
faster than 100 ps. I expected something like 30 ps and it looks like it is
somewhere in that range.

The full schematic (that is rather simple) is shown in 067-0681-01 manual.
The most tricky and tiny part is TD diode. It is soldered to the output BNC
ground lug with it's cathode, then that lug is bent in place so 0603 43 Ohm
resistor (R7) just fits between TD anode and BNC central pin, R7 soldered
in, then one lead of R6 soldered on top of TD Anode / R7 joint. There are
several photos showing that part so there should be no difficulties in
understanding how it is done.

Do NOT replace R4/R5/R6 chain with a single 3kOhm resistor! It MUST be a
string of 3 1/4W resistors -- it doubles as a filter isolating TD from the
rest of schematic, not just current limiter. Do not substitute anything else
with SMD parts either. While everything else would probably work R4/R5/R6
string must be regular thruhole parts. The only SMD part is R7 that is
chosen to be tiny so parasitic inductance is low.

Diodes used are MUR160 for CR1/CR2, 1N4735 as CR3. Transistor is BC560C from
Fairchild. Capacitor is Rubycon LLE, 10uF 160V, DigiKey p/n 1189-1079-ND.
Please note the Fairchild BC560C pinout that is different from other ones (I
might be wrong by my dementia tells me others have different pinout.)

The thing is powered from PG506 as it is suggested in original datasheet,
starts firing somewhere at R1 midpoint.

I will answer all questions if there are any...

Enjoy :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


Dave Daniel
 

On 5/15/2013 10:34 AM, Egge Siert wrote:
 



--- In TekScopes@..., "g_kupka" wrote:
>
>
>
> Where can I get the 067-0681-01 manual? I have only the TU-5 Pulser.
>
> Gerhard

Hi Gerhard,

As far as I know Rens Tel provided you the schematic on 30 July 2012. Your request for it and the reply of Rens Tel is still into the archives. Intriging.

Greetings,

Egge Siert



Mike
 

Great project, thanks!

Oleg ('alexer1' from Kiev) is a good source and a very nice guy, but at one point he sent me about 50% duds on this paricular part. Even worse yield on the 20 mA and 50 mA TD's. He has no control over it; low-C/I parts are intrinsically fragile, these are old and of "dubious provenance", and he's not set up to test.

So be patient and consider ordering double what you plan to use.

I had a little better yield from Evgeniya ('russhama' from Moscow, also a pleasant and honest dealer) but I think she sold out of the low-C high-current parts a while ago.

BTW I got the most out of these button packages by sandwiching them in a PCB microstrip with an end-launch connector. As Tom points out, the inductance of a single turn the size of your lug will kill your risetime. Also I don't recommend directly soldering them (see "fragile" above). Use spring clips, conductive epoxy, silver paint, etc.

There might be something wrong with your 7104 or vertical plugin, that overshoot doesn't look right to me. Welcome to the arms race (only way out is faster and faster...).

Happy pulsing,

mike

BTW, I can see that all those TDs on Ebay are already sold out. For those
who want to build that pulser -- don't you worry guys, that guy probably
have more of those and will relist that item as he did before. He did sold
10 of those some time ago and relisted another 10 that are sold out by now.


Mike
 

David-

Good advice all round. Even so, I found it's very hard to measure one of these (067-0681-01) pulsers with a 7T11/7S11 rig at all, because you have to rely on the "random" sampling function, which is sketchy at best. The firing jitter with respect to the (high-amplitude) drive waveform is way too random for pretriggering. A 7M11-style series delay works, but slows down the edge.

OTOH, with about the same parts count/form factor, one could instead make something more like an HP213B TD pulser. This can be triggered deterministically by any low-amplitude, moderate risetime source (e.g., TTL). Thus you can pretrigger the sampler and measure your edges without excess jitter or delay line distortion.

As a bonus, with this design you just increase the bias a bit and it'll free-run as a relaxation oscillator (feel free to mail me your PG506 in gratitude ;^)

One take on this is shown at

http://groups.yahoo.com/group/TekScopes/photos/album/983015428/pic/425637240/view

I carry one of these in my pocket to swap meets.

mike

--- In TekScopes@yahoogroups.com, David <davidwhess@...> wrote:

Even if you 7T11/7S11/S4 works, in my experience it takes some
practice to get reliable results. I often install mine in the center
slots and a conventional amplifier and timebase in the outer slots to
use as a sanity check.


 

Under ideal conditions, random sampling on my 7T11 works just as well
as sequential sampling. The only problem it has is that using the
internal trigger allows the sampling strobe kickback which gets into
the trigger pickoff to cause distortion right at the trigger point.
The sampler and timebase "see" their own strobe signal when it is in
proximity to the trigger point.

Tektronix mentions this issue in the 7T11 service manual. It is
actually kind of useful in some cases because if you display a long
pulse train, the distortion caused by the kickback shows where the
trigger point occurred.

The Tektronix 284 pulse generator shows how it is possible to make a
low jitter tunnel diode pulser that also includes a pretrigger signal.

On Thu, 16 May 2013 02:19:11 -0000, "Mike" <zuckerme@gmail.com> wrote:

David-

Good advice all round. Even so, I found it's very hard to measure one of these (067-0681-01) pulsers with a 7T11/7S11 rig at all, because you have to rely on the "random" sampling function, which is sketchy at best. The firing jitter with respect to the (high-amplitude) drive waveform is way too random for pretriggering. A 7M11-style series delay works, but slows down the edge.

OTOH, with about the same parts count/form factor, one could instead make something more like an HP213B TD pulser. This can be triggered deterministically by any low-amplitude, moderate risetime source (e.g., TTL). Thus you can pretrigger the sampler and measure your edges without excess jitter or delay line distortion.

As a bonus, with this design you just increase the bias a bit and it'll free-run as a relaxation oscillator (feel free to mail me your PG506 in gratitude ;^)

One take on this is shown at

http://groups.yahoo.com/group/TekScopes/photos/album/983015428/pic/425637240/view

I carry one of these in my pocket to swap meets.

mike

--- In TekScopes@yahoogroups.com, David <davidwhess@...> wrote:

Even if you 7T11/7S11/S4 works, in my experience it takes some
practice to get reliable results. I often install mine in the center
slots and a conventional amplifier and timebase in the outer slots to
use as a sanity check.


g_kupka <ghm1@...>
 

Hi Egge Siert,

I got from Rens the TD Pulser - 067-0681-00 Original Datasheet and not the 067-0681-01 Datasheet

Gerhard

--- In TekScopes@yahoogroups.com, "Egge Siert" <eggeja2@...> wrote:



--- In TekScopes@yahoogroups.com, "g_kupka" <ghm1@> wrote:



Where can I get the 067-0681-01 manual? I have only the TU-5 Pulser.

Gerhard
Hi Gerhard,

As far as I know Rens Tel provided you the schematic on 30 July 2012. Your request for it and the reply of Rens Tel is still into the archives. Intriging.

Greetings,

Egge Siert