Date   

Re: Tektronix 465M - only scale illumination

paronaram
 

I have no luck with this ready to recycle scopes...
This one is with no exception just is bad :)
Almost every single screw stripped, this scope was probably is Tektronix repair school or something...
Missing :
- one IC U2900 from (Horizontal assy)
- some screws

Voltages:
Right by the PT:
o o o o
+32V +5V -5V GND -- ALL Good

on the other side of this same board:
o o o o
-5V GND +5V +32V -- ALL Good

+95V - is missing all I have is +10V and I checked C582 (2.2mF 200V) is leaking it's visible without removing this cap

Is it possible by restoring this +95V will bring both Channels up?

Also what type of IC I'm missing? Can I find replacement

Any help will be greatly appreciated.

Thank you.


Re: 7000 series plug-in ribbon cable extender

John Griessen
 

On 4/28/20 11:03 PM, Brian Symons wrote:
It only takes a few extra holes to allow someone to add the stress relief if they wish but you will have to allow space for the holes & screws / P-Clips between traces & it would be nice to add a marking on the PCB at nice easy distances for measuring.
You don't need to drill the holes
They can be ordinary plated through holes as in my extender kits:

https://www.ecosensory.com/tek/TEK_7K_FLEX_assy-013-sm.jpg


Re: 7000 series plug-in ribbon cable extender

John Griessen
 

On 4/28/20 5:37 PM, Ke-Fong Lin wrote:
The EDAC connector is quite expensive (around $16).
This can be down to about $3.5 each from Taiwan.

The company I bought from is called Chartron Inc. They have a modified drawing for making the 38x2 pin from a more generic 40x2 pin connector.


Re: A question about a failing 7B53A plugin

 

Raymond and Dan, I appreciate your recommendations.

I would never file a cam (god forbid)l I was speculating of slightly filing the legs of the cam assembly where the screws attach it to the board, therefore reducing the clearances of all the contacts.

If a deformed contact would remain in the closed position this would be easy to see after disassembly, and could be bent away from the board as close as possible to its attaching point until it has the same separation as the others.
The big problem I see is that while assembled one cannot see the clearance of the open contacts. To reduce the guess work one could make a note of which positions of the time/div setting are failing, and then from the schematics find the corresponding contact or contacts to check continuity before disassembling and verify the incorrect open circuits, and then pay them attention when disassembled.

This really sounds like precision work, like microsurgery. Fortunately I am quite nearsighted, and I can see well small details without glasses...

Ernesto


Re: 7000 series plug-in ribbon cable extender

Brian Symons
 

Hi,

You can add some degree of stress relief at low cost by adding a few holes.

Use plastic P-Clips for the coax & just use a hole each side of the ribbon cable to allow a short piece of aluminium with a covering or two of heatshrink for cushioning to be screwed down to it.

It only takes a few extra holes to allow someone to add the stress relief if they wish but you will have to allow space for the holes & screws / P-Clips between traces & it would be nice to add a marking on the PCB at nice easy distances for measuring.
You don't need to drill the holes if you don't want to - someone wishing to add the stress relief can do that themselves.

Even a two pairs of holes (a hole each side of the coax) to allow for two cable ties per coax may work - I wouldn't trust just one - & you need to use wide enough cable ties & preferably some sort of cushioning so the cable ties do not distort the coax & cause reflections.
If you do that then please add a note - "Do not distort cables when applying cable ties" - because a maker may not think of that problem.

A silly TV antenna installer here used tight cable ties on our TV coax & the UHF really did not like the distorted coax.
There was no water ingress & the connectors were well done but the distorted coax really affected our signal.
Removing just the flattened section immediately fixed our signal so that was definitely the problem.

You can also get ribbon cable IDC connectors that instead of the wire coming immediately out at right angles, have the wire doing a loop back at the top under a plastic bar to provide some stress relief but I prefer a bar to take the stress off the connector.
I've worked on boards dome like that & they lasted well in applications that had the cable regularly being flexed.
Using Box Headers with the plastic surround around the pins also gives some stress relief & you can use that along with the stress relief type IDC to give some strss relief without using a clamping bar.

I'm sure others here could make some informed comment.


Regards,
Brian.

On 29/04/2020 8:37 am, Ke-Fong Lin wrote:
Hi everyone,

I've already done a TM500/TM5000 extender project that kinda work (with a bodge wire).
I've fixed the routing and hopefully v1.0 will be fine right of the box.

With experience gained, I proceed to make an extender for 7000 series scope plug-ins.

https://drive.google.com/file/d/1xezAJTyPtNLSpBsh3BvpF2dpQ337-xTt/view
https://drive.google.com/file/d/1TJj_3dmeTkXguoz1MMEFxS_noeKs2LK5/view

Intended use is calibration and repair.

It is fully "passive", with optional molex connectors for power supply (don't populate if not needed).
These allow connection to a bench supply and debug short circuits problems (say tantalums).
And (abnormal) power consumption can be a good indicator.

The extender is designed for use with ribbon cable (24AWG for everything non-signal) and coxial (RG174) cables.
Here are few motivations for that:

1) For high frequency (which is the whole point of 7000 series after all), coaxial cables are needed anyway.
2) Flexible cable allows, well, more flexibility, as the plug-in to calibrate/repair can be turned and moved around.
3) It's not that tedious (with ribbon cables) to do the assembly, and it's still cost effective.

Ribbon cable with 2.54mm (.1 inch) is readily available, can be bought in small quantities (1m/3ft) at Farnell for example
26AWG ribbon cable should even be fine for non-power.
With a length of 50cm (around 20 inches), and 10 conductors, that would require 4m of ribbon cable.
Plus 3m (6 * 50cm) of RG174 coax.
Both ribbon cable and RG174 are not that expensive if bought in 30m (100ft) spools (let's do a group buy!).

The tedious part of assembly would be to strip the coaxials and ground the shields. On the PCB, there's one GND pad for each set of 2 signals pads.
The problematic part is on the "female" connector (EDAC 345-076-500-*)
The shields would have to be stripped. Soldered to a wire (shrink tube to protect the soldering), with that wire soldered to the GND pins.
That's a bit of work! I've considered making a small PCB for the connector, but unfortunately, the plug-in have plastic that prevents it.

The PCB extender doesn't have strain relief for wires. For non-industrial use, that shouldn't be necessary, especially if careful!
From the experience with the TM500 extender, that's not an issue.
Also note that front face is right side, this is so the wirings are on the left, and hence, make insertion vertical compartment (on right) easier.

Can you guys have a look give a few comments and suggestions?
I can send you the Kicad and gerber files if interested.

In particular, pins 21A and 21B are both marked as signal here:

http://w140.com/tekwiki/wiki/7000_Series_plug-in_interface

But note 5 of the same document, says they're grounded.
And the "official" Tektronix extender on ebay, don't seem to use coax on it.

When it's done, I'll put everything on github.

One more thing, the total cost would be around 35-40 euros.
The EDAC connector is quite expensive (around $16).
The PCB at JLPCB is quite cheap (5 with "gold fingers" and ENIG for less $20 without shipping).
As for ribbon cable and coaxial, if we do a group buy and take spools. I would estimate at less than 10 euros.

Best regards,


Re: SG503 with flashing display and amplitude problems.

Mlynch001
 

Dan,

Thanks for your reply. I have been looking at this thing earlier today. I did some additional checking LR190 checks as good. I checked all the other inductors and found no opens. One thing that happened is the unit began to function properly for a short time. So something is intermittent in the unit.

Q180 tests good on a component tester.

Measuring the voltages at Q180, I found these voltages:

Emitter -18.290V
Base -18.686V
Collector -22.054V

Measuring Q190 I found these voltages:

Collector 14.7 p-p sine wave with a 3.656 DC offset.
Base. 6.3V p-p sine wave with a -18.43 DC offset.
Emitter 1 and 2 -18.873V

I don’t know if these measurements reveal anything useful.

I appreciate your feedback.

Sincerely,

--
Michael Lynch
Dardanelle, AR


Re: Tektronix 465M - only scale illumination

Mlynch001
 

No problem, good luck. You will find this forum full of people who are very willing to help and know a lot about this equipment. I have fixed several of these scopes, so I watch he 4xx series posts with great interest.
--
Michael Lynch
Dardanelle, AR


Re: Tektronix 465M - only scale illumination

paronaram
 

Thank you, OK now I know, it's a B053102

Thanks again!


Re: Tektronix 465M - only scale illumination

Mlynch001
 

I don’t think that is the unit serial number. The TEKTRONIX serial numbers usually begins with B, followed by a series of numbers. There is a downloadable copy of the 465M manual on the TEK Wiki site.

Here is the link:

http://w140.com/tekwiki/wiki/465M

--
Michael Lynch
Dardanelle, AR


Tektronix 465M - only scale illumination

paronaram
 

Wile I'm trying to figure out what's going on with my 2235 horizontal sweep, I got another scope, maybe this will be easier to fix.
I have Tektronix 465M with only scale illumination and broken CH 2 Volts/Div knob.
I have heard that schematic diagram can vary from s/n to s/n
Can you help find correct schematics?
s/n 6625-01-032-6914

Thank you


Re: A question about a failing 7B53A plugin

Dan G
 

Hi Ernesto,

On Tue, Apr 28, 2020 at 08:12 PM, Ernesto wrote:

You said that adjustment may be necessary? What can dis-adjust the
contacts? I thought that dirt is the only cause of non-continuity.
In most cases, spring contact adjustment is not necessary -- just very
careful cleaning with IPA.

In extreme cases, some contacts may have lost their spring action if they
were stored in the closed position for decades on end. Then you may find that
they won't lift off the PCB even when the cam moves out of the way
(contact always closed). Similarly, they may have settled into the pressed
position to the point where they no longer exert sufficient force on the PCB
(contact has high resistance when closed).

In either case, we are talking about *tiny* adjustment in the range of fractions
of a millimeter. Do not bend anything more than that, and under no circumstance
should filing ever be required!

Be careful not to bend the contacts out of shape, or to scratch either the
PCB gold pad, or the underside of the spring contact. Having a watchmaker's
touch really helps here, as does having watchmaker tools.

In general, I would not recommend adjusting the contacts unless absolutely
necessary, and then only after thorough cleaning with IPA does not resolve
the problem. When cleaning, make sure not to leave tiny fibers on the
contact surfaces -- these are a well known source of switch reliability problems.


dan


Re: A question about a failing 7B53A plugin

 

On Wed, Apr 29, 2020 at 02:12 AM, Ernesto wrote:


You said that adjustment may be necessary?
Two main adjustments, on top of the obvious positioning of the knobs against the scale:
- It takes some adjustment during assembly to have *one* good, positive clicking action when the A- and B-knobs are interlocked. The hex screws must be quite tight. The open side of the C-shaped filler tube must sit between the two screws in the B-timebase (grey) knob.
- Sometimes, one or several contact fingers aren't firmly/positively touching/pushed against the PCB by the cams, caused by minute wear of the cam or slight metal deformation. Filing a cam should not be done and brings the finger *away* from the board. Very slightly bending the notch of the finger "up" where it touches the cam will do the job. If ever necessary, I hold the free end of the contact finger close to the board and very slightly bend the notch up with a small tool (you have to move up about 1 mm to overcome the elasticity of the finger). Check by observing the movement of the finger when lifted by the cam; you should be able to observe a very very slight movement (less than 0.5mm is enough) of the finger across the PCB just when the cam is reaching its top position against the finger's notch. Again, seldomly necessary but avoids repeated assembly/disassembly looking for the problem if it is.

Raymond


Re: A question about a failing 7B53A plugin

 

Thank you Raymond. I will disassemble the 7B53A and clean the contacts with 99% IPA. You said that adjustment may be necessary? What can dis-adjust the contacts? I thought that dirt is the only cause of non-continuity. Is there an adjustment of the space between board and cams? Maybe some cams wear out due to use? Maybe one can file off some of the plastic and get the cams closer to the board? It seems that there is enough clearance that an accidental contact closure is not probable. (the experimental engineer is speaking, ha ha). Will see, will give it a try.

Ernesto


Re: 7854 ROMs: the whole story

Paul
 

Hello Christian -

Regarding edge connectors - why should you never put Sn/Pb surfaces in a gold connector? I have made extender cards for my Type 230 readout unit and the cost difference between Sn/Pb and hard gold was too much for me to spend. I didn't go with ENIG because of the concern of it rubbing off, but will I be doing damage to my connectors with my boards?

regards -
Paul


Re: respectable 7A13 plugin disabled by common failure...

 

On Tue, Apr 28, 2020 at 11:22 PM, Ernesto wrote:


In over a quarter of a century of dealing with Tektronix oscilloscopes, my
most interesting experiences happened and are happening in the last few
months. And they have one thing in common: tantalum capacitors. The
traditional design excellence may have slipped a little in the selection of
component tolerances. I had found a 15V tantalum on a 15V supply, and now it
is a 20V one also with 15V applied. I have now an electrolytic (not tantalum)
100 uF capacitor of the same size of the blown one it will replace, and it is
a 35V, not 20V.
Maybe this is a bit less straightforward than you think, for the following reasons:
- Most tantalums have been inside these 'scopes for well over 30 years. Many still survive, though I won't say replacing all would be a bad idea...
- When these circuits were designed, the characteristics of epoxy-dipped Ta caps were less well known. Manufacturers specified hardly any DC voltage derating generally, as opposed to Al caps. Surge current was to be limited though. The R's or L's you often find in the power lines helped to reduce surge current during power up.
- Hermetic (cylindrical) Ta caps *very* seldomly fail, even when only slightly DC derated. The epoxy-encapsulated caps die (again, after a long life), so humidity may be a factor. Documentation on failure mechanisms is available on the Internet.

Re. the 7A13: It is a very nice vertical plugin. It may be used as a diff amp but it also allows observing very small low frequency AC voltages on a large, finely adjustable DC offset.

Raymond


Re: TEK 5054 CPU Fan

BobH
 

If it is any help, here is the writeup I did on the project including source code for windows or Linux

https://hackaday.io/project/25558-screen-capture-utility-for-tektronix-ths720-scopes

Bob

On 4/28/20 3:02 PM, David Kuhn wrote:
" BMP format for the printer, then told the scope to print
the screen and captured the data from the serial line to a file."

I would love to learn to do that. I have one utility (that I did not
write) that captures the screen, or at least the waveform part, with the
TDS3000 series scope. It's part of an automatic test system. So I am sure
it can be done. I don't know that the programming manual shows how to do
that, but I will need to study it more closely.

One little chuckle I had to have with the TDS5054. I use a custom 100:1
probe for what I do. It does not have a pin to put the scope in to 100:1
mode automatically. With the TDS3054, I just have to tell it to go to
100:1 mode. I forget the exact command but if I recollect correctly, it is
a'most exactly telling "100". Thru the front panel it's also very easy to
tell it you are using a 100:1 probe.

Well, with the TDS5054, you can not even through the user interface, tell
it manually you have a 100:1 probe. No... You have to tell it you
are attenuating the input signal by 40dB. This scope is an engineer's
scope more than a tech's scope. You have to think a little more to use it
compared to the TDS3000 series.

dave

On Tue, Apr 28, 2020 at 2:55 PM BobH <wanderingmetalhead@dakotacom.net>
wrote:

It may not be too difficult to write a utility to do that. I wrote one
for my THS720 a few years ago. Not sure about using the GPIB bus though,
the THS720 only has a serial port. Via the serial port, I configured the
THS720 to use a BMP format for the printer, then told the scope to print
the screen and captured the data from the serial line to a file. Tek
programmers manuals are quite thorough.

BobH

On 4/28/20 9:07 AM, David Kuhn wrote:
" TE together via GPIB with a NI USB-HS feeding to my computer - that
lets
me do anything remotely that I want without the risks inherent in having
it
live on the network."

I used GPIB too, with Agilent Controller and VISA (along with some TEK
visa
sprinkled around here and there). I wish they was a utility to get a
screen shot from the TDS5054. That is why I was looking at network
conductivity. Doing a Scrprt pasting it in to paint and the saving it
to a
thumb drive to transfer the screen shot to a computer that is editing a
documents is a lot of steps and a pain in the ass. With the TDS3054
scopes, I just use openchoice desktop and in a few seconds after
clicking a
button, I have a screen shot and copy it to the clipboard..

Other than some different way of doing the same tasks, adapting it
software
wise to utilities it easy enough. It's the screen shot that has me
stumped.


Re: A question about a failing 7B53A plugin

 

On Tue, Apr 28, 2020 at 11:39 PM, Ernesto wrote:


has any of you had this kind of problem, and were you able to fix it by
disassembling the switch, removing the cylinder with the cams from the board
and spraying the contact fingers with contact cleaner?
Hi Ernesto,
Disassembling and reassembling for cleaning is often necessary with these old units. It isn't as much work nor as difficult as you perhaps think. It's also a very good exercise, since many of the dual timebases in this kind of Tek gear have a similar construction. That applies at least to the 7000-series 'scopes and the 400-series portables. You'll be ready for your first 7B92A afterwards.
The spring-and-cam construction is used in just about all Tek (7000-series) amplifiers and timebases, so it's an exercise for working on those as well. Reassembling is not difficult if you prepare yourself by studying how the unit is put together and why, because it takes some manual adjustment.
Some people seem to clean without disassembly but I always disassemble if I want to be sure that I'm done in one go, avoiding a need to revisit after a short period.

Spraying the contact fingers with contact cleaner is hardly ever necessary and only advisable if you thorougly clean the contacts (rinse with IPA) afterwards. *Don't* leave contact cleaner on the contacts.
I almost always clean the contacts with IPA. They have a tiny wiping action on the PCB when lifted by the cams. In exceptional cases, a tiny bit of adjustment may be advisable, just by bending, using the right tools.
Many people clean the contacts by moving a "wet" piece of *non-abrasive* paper between contact and PCB. Use either IPA or contact cleaner to wet the contacts. Again, rinse if you use contact cleaner.

Raymond


7000 series plug-in ribbon cable extender

Ke-Fong Lin
 

Hi everyone,

I've already done a TM500/TM5000 extender project that kinda work (with a bodge wire).
I've fixed the routing and hopefully v1.0 will be fine right of the box.

With experience gained, I proceed to make an extender for 7000 series scope plug-ins.

https://drive.google.com/file/d/1xezAJTyPtNLSpBsh3BvpF2dpQ337-xTt/view
https://drive.google.com/file/d/1TJj_3dmeTkXguoz1MMEFxS_noeKs2LK5/view

Intended use is calibration and repair.

It is fully "passive", with optional molex connectors for power supply (don't populate if not needed).
These allow connection to a bench supply and debug short circuits problems (say tantalums).
And (abnormal) power consumption can be a good indicator.

The extender is designed for use with ribbon cable (24AWG for everything non-signal) and coxial (RG174) cables.
Here are few motivations for that:

1) For high frequency (which is the whole point of 7000 series after all), coaxial cables are needed anyway.
2) Flexible cable allows, well, more flexibility, as the plug-in to calibrate/repair can be turned and moved around.
3) It's not that tedious (with ribbon cables) to do the assembly, and it's still cost effective.

Ribbon cable with 2.54mm (.1 inch) is readily available, can be bought in small quantities (1m/3ft) at Farnell for example
26AWG ribbon cable should even be fine for non-power.
With a length of 50cm (around 20 inches), and 10 conductors, that would require 4m of ribbon cable.
Plus 3m (6 * 50cm) of RG174 coax.
Both ribbon cable and RG174 are not that expensive if bought in 30m (100ft) spools (let's do a group buy!).

The tedious part of assembly would be to strip the coaxials and ground the shields. On the PCB, there's one GND pad for each set of 2 signals pads.
The problematic part is on the "female" connector (EDAC 345-076-500-*)
The shields would have to be stripped. Soldered to a wire (shrink tube to protect the soldering), with that wire soldered to the GND pins.
That's a bit of work! I've considered making a small PCB for the connector, but unfortunately, the plug-in have plastic that prevents it.

The PCB extender doesn't have strain relief for wires. For non-industrial use, that shouldn't be necessary, especially if careful!
From the experience with the TM500 extender, that's not an issue.
Also note that front face is right side, this is so the wirings are on the left, and hence, make insertion vertical compartment (on right) easier.

Can you guys have a look give a few comments and suggestions?
I can send you the Kicad and gerber files if interested.

In particular, pins 21A and 21B are both marked as signal here:

http://w140.com/tekwiki/wiki/7000_Series_plug-in_interface

But note 5 of the same document, says they're grounded.
And the "official" Tektronix extender on ebay, don't seem to use coax on it.

When it's done, I'll put everything on github.

One more thing, the total cost would be around 35-40 euros.
The EDAC connector is quite expensive (around $16).
The PCB at JLPCB is quite cheap (5 with "gold fingers" and ENIG for less $20 without shipping).
As for ribbon cable and coaxial, if we do a group buy and take spools. I would estimate at less than 10 euros.

Best regards,


Re: TEK 5054 CPU Fan

David Kuhn
 

" BMP format for the printer, then told the scope to print
the screen and captured the data from the serial line to a file."

I would love to learn to do that. I have one utility (that I did not
write) that captures the screen, or at least the waveform part, with the
TDS3000 series scope. It's part of an automatic test system. So I am sure
it can be done. I don't know that the programming manual shows how to do
that, but I will need to study it more closely.

One little chuckle I had to have with the TDS5054. I use a custom 100:1
probe for what I do. It does not have a pin to put the scope in to 100:1
mode automatically. With the TDS3054, I just have to tell it to go to
100:1 mode. I forget the exact command but if I recollect correctly, it is
a'most exactly telling "100". Thru the front panel it's also very easy to
tell it you are using a 100:1 probe.

Well, with the TDS5054, you can not even through the user interface, tell
it manually you have a 100:1 probe. No... You have to tell it you
are attenuating the input signal by 40dB. This scope is an engineer's
scope more than a tech's scope. You have to think a little more to use it
compared to the TDS3000 series.

dave

On Tue, Apr 28, 2020 at 2:55 PM BobH <wanderingmetalhead@dakotacom.net>
wrote:

It may not be too difficult to write a utility to do that. I wrote one
for my THS720 a few years ago. Not sure about using the GPIB bus though,
the THS720 only has a serial port. Via the serial port, I configured the
THS720 to use a BMP format for the printer, then told the scope to print
the screen and captured the data from the serial line to a file. Tek
programmers manuals are quite thorough.

BobH

On 4/28/20 9:07 AM, David Kuhn wrote:
" TE together via GPIB with a NI USB-HS feeding to my computer - that
lets
me do anything remotely that I want without the risks inherent in having
it
live on the network."

I used GPIB too, with Agilent Controller and VISA (along with some TEK
visa
sprinkled around here and there). I wish they was a utility to get a
screen shot from the TDS5054. That is why I was looking at network
conductivity. Doing a Scrprt pasting it in to paint and the saving it
to a
thumb drive to transfer the screen shot to a computer that is editing a
documents is a lot of steps and a pain in the ass. With the TDS3054
scopes, I just use openchoice desktop and in a few seconds after
clicking a
button, I have a screen shot and copy it to the clipboard..

Other than some different way of doing the same tasks, adapting it
software
wise to utilities it easy enough. It's the screen shot that has me
stumped.




A question about a failing 7B53A plugin

 

A third 7B53A plugin I have has shown many problems. They are: the inability to transition correctly to a delayed sweep, some funny readouts, and the failure in many of the time/div switch positions.

First I localized that a slide switch on the main board that is activated by pulling in and out the knob that sets the delay sweep had been YANKED lose (the ends of the metal cover that hold the insulating piece with the terminals bent out of shape), another case of maybe a dyslectic bodybuilder pulling from the knob beyond its limit. I fixed this, and then I removed the long narrow readout board with the finger contacts activated by the cylinder of cams, and sprayed the contacts with contact cleaner. This improved somewhat the performance of the unit. But what remains is the sweep stopping on many of the time/div settings, and this must be a switch bad contact too, where the R in the RC time constant for the sweep is open.

Here is my question: has any of you had this kind of problem, and were you able to fix it by disassembling the switch, removing the cylinder with the cams from the board and spraying the contact fingers with contact cleaner?
I can do this, but it is a lengthy procedure, and if the chances of success are not very good I might just give up on this dual time base, since I have plenty of these.

Thanks for your answer,
Ernesto

20081 - 20100 of 186480