Topics

Back from vacation

kb6nax
 

I bowed out of posting to TekScopes a few years ago because of the dearth of vacuum tube 'scope postings. More recently I've become a fan of Tek 455 'scopes. For 465M lovers the too models have identical power supplies. 455's frequently appear on the 'bay and I've bought a few "for parts/repair" units and now have eight restored. Two to go with bad HV supplies that are now guinea pigs for building from scratch replacement HV multiplier modules. Both the scopes appear to have good transformers and will produce +95 volts when run with a module that has the 3x multiplier disconnected. I've been on and off that project for about a year now but when time is available I continue to solve some of the riddles of how to make a good module. I'm about to construct a breadboard to tackle the mysteries of the bias clamp circuit. The 3x multiplier will be outboard and far enough to avoid getting zapped by 10KV, If anyone is interested I'll post my progress/results.

 

Hi Allen,

If you are serious about troubleshooting and repairing scopes you should get yourself a Tek curve tracer. Based on your post you would probably have the most fun with the vacuum tube based 575 transistor curve tracer which you can find in Craig's List for ~$50. But it doesn't really matter which one you get. If you already have a curve tracer I know you are serious about what you are doing.

The good news is for about $100 more you can add the capability of testing vacuum tubes to any Tek curve tracer. It is simple. You may have missed it but about 9 months ago I stumbled on a clever, inexpensive, way to test almost every vacuum tube (and definitely the ones in Tek vintage scopes) by displaying a full set of characteristic curves on any Tek curve tracer. The curve tracer will still test semiconductors because no modifications are made to the curve tracer.

I wrote a paper describing how to do it and I made some PC boards with my adapter design. The very latest and most complete version of the paper I wrote on my Vacuum Tube Curve Tracer Adapter is always located at

http://www.ke5fx.com/A_VTCT_Adapter_for_All_Tektronix_SCTs_W7PF.pdf


Dennis Tillman W7PF

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of kb6nax
Sent: Tuesday, December 19, 2017 2:10 PM
To: TekScopes@groups.io
Subject: [TekScopes] Back from vacation

I bowed out of posting to TekScopes a few years ago because of the dearth of vacuum tube 'scope postings. More recently I've become a fan of Tek 455 'scopes. For 465M lovers the too models have identical power supplies. 455's frequently appear on the 'bay and I've bought a few "for parts/repair" units and now have eight restored. Two to go with bad HV supplies that are now guinea pigs for building from scratch replacement HV multiplier modules. Both the scopes appear to have good transformers and will produce +95 volts when run with a module that has the 3x multiplier disconnected. I've been on and off that project for about a year now but when time is available I continue to solve some of the riddles of how to make a good module. I'm about to construct a breadboard to tackle the mysteries of the bias clamp circuit. The 3x multiplier will be outboard and far enough to avoid getting zapped by 10KV, If anyone is interested I'll post my progress/results.
--
Dennis Tillman W7PF
TekScopes Moderator

Tam Hanna
 

Dennis,
that paper is insane! Wow.

Given that I just resumed - albeit at a very slow pace, it's Christmas and the terrible stress some people make about it is extremely bad for my health - work on my digital upgrade for the Danaher 57x series; please allow me to ask if there is any software feature I could implement to simplify tube testing!

Tam
---
With best regards
Tam HANNA (emailing on a BlackBerry PRIV)

Enjoy electronics? Join 6500 other followers by visiting the Crazy Electronics Lab at https://www.instagram.com/tam.hanna/

 

Hi Tam.
How do you convert an analog curve tracer to digital???
Wouldn't it be easier to start from scratch and build a digital curve tracer rather than try to modify an existing analog curve tracer.
They do exist and they have lots of capabilities you can't get in a Tek analog curve tracer. BUT...
As far as I know none of them has the capability to test devices at the same voltage, current, and overall power levels of the Tek curve tracers. At the opposite end of the spectrum I believe the Tek curve tracers are far more sensitive at micro- and nano-amp currents. Versatility would be another consideration where Tek would probably excel over a digital curve tracer.

A few advantages to a digital curve tracer would be network interconnectivity, GPIB programmability, hard copy output, the ability to save and recall results to/from a file, automatic calculation of all parameters, scale factor readout,

I think the big problem you face is how much this is going to cost to build. It is comparatively simple to capture a trace on a scope because there is a one to one mapping of vertical amplitude values to horizontal time values. In the case of XY displays it is essentially the same - a one to one mapping of vertical amplitude values to horizontal amplitude values. But a curve tracer displays parametric values. For every horizontal amplitude value there can be anywhere from one to a hundred vertical amplitude values on a Tek curve tracer depending on how many vertical base steps you use So you have a one to many mapping. That gets very complicated to try and capture which you would have to do if you were modifying an existing Tek curve tracer. But if you design from scratch it gets easier to capture any number of vertical values since you control each base step and horizontal voltage increment.

Good luck with your project.

Dennis

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Tam Hanna
Sent: Tuesday, December 19, 2017 8:30 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] Back from vacation

Dennis,
that paper is insane! Wow.

Given that I just resumed - albeit at a very slow pace, it's Christmas and the terrible stress some people make about it is extremely bad for my health - work on my digital upgrade for the Danaher 57x series; please allow me to ask if there is any software feature I could implement to simplify tube testing!

Tam
---
With best regards
Tam HANNA (emailing on a BlackBerry PRIV)

Enjoy electronics? Join 6500 other followers by visiting the Crazy Electronics Lab at https://www.instagram.com/tam.hanna/
--
Dennis Tillman W7PF
TekScopes Moderator

Chuck Harris
 

A digital curve tracer is simply a digitally controlled
couple of power supplies, and a digitally read voltmeter,
and a program to do the reading and controlling.

Apply current to the base, current to the collector, ground
the emitter, and measure the voltages everywhere... wash rinse
repeat.

I made my first curve tracer using two adjustable bench power
supplies, two resistors, one for the collector load, one for
the base, and an X-Y chart recorder.

I set the base supply's voltage, dropped the pen, adjusted
the collector supply through its range, and let the recorder
draw whatever I was measuring... Vce, Ic, whatever. Lifted
the pen, reset the collector supply to zero, set a new base
supply voltage...

It is adding the human usable displays that added most of the
complexity to the 57X curve tracers.

-Chuck Harris

Dennis Tillman W7PF wrote:

Hi Tam.
How do you convert an analog curve tracer to digital???
Wouldn't it be easier to start from scratch and build a digital curve tracer rather than try to modify an existing analog curve tracer.
They do exist and they have lots of capabilities you can't get in a Tek analog curve tracer. BUT...
As far as I know none of them has the capability to test devices at the same voltage, current, and overall power levels of the Tek curve tracers. At the opposite end of the spectrum I believe the Tek curve tracers are far more sensitive at micro- and nano-amp currents. Versatility would be another consideration where Tek would probably excel over a digital curve tracer.

A few advantages to a digital curve tracer would be network interconnectivity, GPIB programmability, hard copy output, the ability to save and recall results to/from a file, automatic calculation of all parameters, scale factor readout,

I think the big problem you face is how much this is going to cost to build. It is comparatively simple to capture a trace on a scope because there is a one to one mapping of vertical amplitude values to horizontal time values. In the case of XY displays it is essentially the same - a one to one mapping of vertical amplitude values to horizontal amplitude values. But a curve tracer displays parametric values. For every horizontal amplitude value there can be anywhere from one to a hundred vertical amplitude values on a Tek curve tracer depending on how many vertical base steps you use So you have a one to many mapping. That gets very complicated to try and capture which you would have to do if you were modifying an existing Tek curve tracer. But if you design from scratch it gets easier to capture any number of vertical values since you control each base step and horizontal voltage increment.

Good luck with your project.

Dennis

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Tam Hanna
Sent: Tuesday, December 19, 2017 8:30 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] Back from vacation

Dennis,
that paper is insane! Wow.

Given that I just resumed - albeit at a very slow pace, it's Christmas and the terrible stress some people make about it is extremely bad for my health - work on my digital upgrade for the Danaher 57x series; please allow me to ask if there is any software feature I could implement to simplify tube testing!

Tam
---
With best regards
Tam HANNA (emailing on a BlackBerry PRIV)

Enjoy electronics? Join 6500 other followers by visiting the Crazy Electronics Lab at https://www.instagram.com/tam.hanna/




Tam Hanna
 

Hello folks,

no, no. The idea is to replace the CRT and leave the rest of the unit as it is!


Tam

--
With best regards
Tam Hanna
---

NEW: Enjoy electronics? Like seeing oscilloscopes get repaired? Please subscribe to my new YouTube channel -> http://www.youtube.com/user/MrTamhan

 

Hi Tam,
My Vacuum Tube Curve Tracer Adapter uses all of the Tek Curve tracers just as they are. No modifications are necessary. So as long as your CRT replacement is faithful to the original curve tracer all I would add is storage capability like the 577D1 has, and the ability to printout what is on the display (which would be a new and useful feature).

Dennis

-----Original Message-----
Sent: Wednesday, December 20, 2017 2:51 PM
Subject: Re: [TekScopes] Back from vacation

Hello folks,

no, no. The idea is to replace the CRT and leave the rest of the unit as it is!

Tam

--
With best regards
Tam Hanna
---

NEW: Enjoy electronics? Like seeing oscilloscopes get repaired? Please subscribe to my new YouTube channel -> http://www.youtube.com/user/MrTamhan
--
Dennis Tillman W7PF
TekScopes Moderator

 

On Tue, 19 Dec 2017 23:44:16 -0800, you wrote:

How do you convert an analog curve tracer to digital???
Pick the horizontal and vertical signals off at the inputs to the CRT
amplifier. Ideally this is done with a pair of instrumentation
amplifiers to grab the low noise differential signals and reject the
common mode signals but that might not be necessary.

Wouldn't it be easier to start from scratch and build a digital curve tracer rather than try to modify an existing analog curve tracer.
Think of it instead as using the power supplies and range circuits of
an existing curve tracer.

They do exist and they have lots of capabilities you can't get in a Tek analog curve tracer. BUT...
As far as I know none of them has the capability to test devices at the same voltage, current, and overall power levels of the Tek curve tracers. At the opposite end of the spectrum I believe the Tek curve tracers are far more sensitive at micro- and nano-amp currents. Versatility would be another consideration where Tek would probably excel over a digital curve tracer.
Designing a source/measurement unit, and it will take two of them,
with high voltage, high current, and high dynamic range is a challenge
by itself so in that respect, it makes sense to add digitizing
capability to an existing analog curve tracer.

 

Hi David,
How will you handle the parametric data (multiple vertical traces for each
single horizontal point) a curve tracer generates? Unlike a conventional
digital scope with one vertical data point for every horizontal data point
(a "ONE TO ONE" function) a curve tracer's parametric display is "MANY TO
ONE" and not a function. Tek curve tracers can generate as many as 100
vertical points for each horizontal data point on the screen During some
measurement setups you might actually end up with a unique complete function
for each horizontal data point. That's a lot of complex data. How are you
going to capture it? That step will be more difficult than storing it.

Dennis Tillman W7PF

-----Original Message-----
Sent: Thursday, December 21, 2017 12:04 PM
Subject: Re: [TekScopes] Back from vacation

On Tue, 19 Dec 2017 23:44:16 -0800, you wrote:

How do you convert an analog curve tracer to digital???
Pick the horizontal and vertical signals off at the inputs to the CRT
amplifier. Ideally this is done with a pair of instrumentation amplifiers
to grab the low noise differential signals and reject the common mode
signals but that might not be necessary.

Wouldn't it be easier to start from scratch and build a digital curve
tracer rather than try to modify an existing analog curve tracer.

Think of it instead as using the power supplies and range circuits of an
existing curve tracer.

They do exist and they have lots of capabilities you can't get in a Tek
analog curve tracer. BUT...
As far as I know none of them has the capability to test devices at the
same voltage, current, and overall power levels of the Tek curve tracers. At
the opposite end of the spectrum I believe the Tek curve tracers are far
more sensitive at micro- and nano-amp currents. Versatility would be another
consideration where Tek would probably excel over a digital curve tracer.

Designing a source/measurement unit, and it will take two of them, with high
voltage, high current, and high dynamic range is a challenge by itself so in
that respect, it makes sense to add digitizing capability to an existing
analog curve tracer.
--
Dennis Tillman W7PF
TekScopes Moderator

 

Capturing multiple sweeps is not much of a problem but distinguishing
them may be difficult where the traces crowd together usually near the
origin. If this is a problem, then one solution is to also capture
the digital pulse from the step generator which may be used to
indicate the start of the next trace.

On Thu, 21 Dec 2017 16:25:14 -0800, you wrote:

Hi David,
How will you handle the parametric data (multiple vertical traces for each
single horizontal point) a curve tracer generates? Unlike a conventional
digital scope with one vertical data point for every horizontal data point
(a "ONE TO ONE" function) a curve tracer's parametric display is "MANY TO
ONE" and not a function. Tek curve tracers can generate as many as 100
vertical points for each horizontal data point on the screen During some
measurement setups you might actually end up with a unique complete function
for each horizontal data point. That's a lot of complex data. How are you
going to capture it? That step will be more difficult than storing it.

Dennis Tillman W7PF

Jason A.
 

Hi Dennis,

I have been combing through the posts here and didn't see where I would be able to get one of your boards. I have a 7CT1N and plan to make the power supplies and tube plugin boards myself (I have a Jackson 658A which is allegedly a dynamic mutual conductance tester, or Dynamic Output as they phrase it, but due to its layout it would be very difficult to modify per your article).

Please hold me to task if you see anything here that is a red-flag to you... :-)

I have a plethora of transformers, tube sockets, project boxes and sundry parts to make it all happen, so that part doesn't daunt me (unless I missed something along the way). I plan to use one of these with a variable resistor for the screen supply: http://www.antiquewireless.org/uploads/1/6/1/2/16129770/37-a_solid-state_filter_choke_or_field_coil_replacement.pdf - that should do the trick of getting anywhere up to the 300+V screen values you mentioned in your article. In fairness, I only plan on testing small signal tubes in either Octal, Noval, or B7G, so that limits the number of sockets I would need to wire up. I have a 0-30V, 0-3A supply that could work well for filaments. I don't see why a couple of wall warts couldn't be used for the -50 and +6V supplies - I have a 48v 500mA as well as several 6V ones in my parts bin and I have a feeling that would be close enough for my purposes.

If you don't see anything wrong with my plans, having your board would certainly make things easier from that standpoint, but I could always proto-board something together otherwise.

Thanks for showing a good way to do this! Thankfully I have a 7834, so I can even use the screen memory to compare sections.

Thanks again and best regards,

Jason