Topics

475 With a bowed display. . . .Ideas?


Michael W. Lynch
 

I recently found a nice 475. Upon receiving the unit, I fired it up, the scope presented a strange garbled trace. Checking the various LV supplies, I found the +15V and the +15V Unregulated to be very low at about 12.2 and 16.5 respectively. Lifted the +15V jumper and there was no change, indicating that the problem was in the power supply, not the scope circuitry. Otherwise the scope looks almost new inside, no signs of tampering or repairs. All other supply voltages are spot on.

Problem was C1442 was bad. High resistance and almost no capacitance, loading the +15V supply. Replaced that cap with a new modern cap and both voltages are now spot on. Ripple checked on all supplies and we are within specs.

Cleaned and exercised all the push button switches.

Trace is bright and fairly clear for a mesh CRT.

Horz and Vert Position controls work as expected.

Focus works well

Trace Rotation R1386 controls rotation either side of horizontal equally.

Set the scope in X-Y and defocus the dot, the resulting spot is nice and round. Astig control R1397 produces the expected change in the spot. The Spot will go "oblong" in both X and Y directions in response to turning R1397.

Horizontal timing is close to right as the timing marker setting generally agrees with the Time/Div switch setting, some adjustment is needed, but the horizontal is only off about 5-7% or so.

Problem is when I put my marker signal from my marker generator into the unit and set Volts/Div knob for 6-8 division vertical markers then I center the middle marker to align with the center vertical graticule line, as you move away from the center, the markers become more and more "leaned in" in at the top. Left and right of center, the tops of the markers lean progressively more and more toward the center as you move farther from the centerline. The bottom trace line is barely, but ever so slightly bowed. The vertical markers are straight but just leaning in toward the centerline. The bottom trace bowing is not nearly as profound as the "leaning" of the markers. Trace can be adjusted to about 1/2 trace width of bow, checking at top middle and bottom of the display.

Y Align R1385 and Geometry R1390 will not bring the vertical markers into proper vertical alignment even though they do have some effect on the trace by making it less severe. Even though they never "Fix" the problem.

I have fixed several of these scopes and never encountered this issue before, where it could not be adjusted out of the instrument.

Ideas or suggestions?

Thanks in Advance.
--
Michael Lynch
Dardanelle, AR


Stephen
 

On Fri, Aug 7, 2020 at 09:38 AM, Michael W. Lynch wrote

Nice find Michael. Congrats.

Problem is when I put my marker signal from my marker generator into the unit
and set Volts/Div knob for 6-8 division vertical markers then I center the
middle marker to align with the center vertical graticule line, as you move
away from the center, the markers become more and more "leaned in" in at the
top. Left and right of center, the tops of the markers lean progressively
more and more toward the center as you move farther from the centerline. The
bottom trace line is barely, but ever so slightly bowed. The vertical markers
are straight but just leaning in toward the centerline. The bottom trace
bowing is not nearly as profound as the "leaning" of the markers. Trace can
be adjusted to about 1/2 trace width of bow, checking at top middle and bottom
of the display.
I’m not sure, but that makes me think « geometry »? Maybe??


 

On Sat, Aug 8, 2020 at 01:26 AM, Stephen wrote:


I’m not sure, but that makes me think « geometry »? Maybe??
Possibly not:
In the paragraph immediately following your longish quote the OP writes:

"Y Align R1385 and Geometry R1390 will not bring the vertical markers
into proper vertical alignment even though they do have some effect on
the trace by making it less severe. Even though they never "Fix" the problem."

Raymond


Chuck Harris
 

Typically, when a proper application of the Geometry,
Focus, Astigmatism, Trace Rotation, and other affiliated
adjustments results in a screen that just doesn't look
as pretty as we would like, it means that there has been
an outside physical influence on the CRT (READ: OOPS! BANG!
$%^! & #!!*^! Why Me???)

The Mumetal shield can get bent, or magnetized. One of
the steering magnets (held with Scotch tape) can fall off
of the CRT neck, causing the beam to partially strike one
many internal shield electrodes that looks like a hole in
a plate of metal.

Or, it can be that we are much more nit-picky than tektronix
was originally.

If you carefully read the specifications in the manual, you
will often find that we have been spoiled rotten by all of
the CRTs we get that are much better than the specifications
say they could be...

-Chuck Harris

Raymond Domp Frank wrote:

On Sat, Aug 8, 2020 at 01:26 AM, Stephen wrote:


I’m not sure, but that makes me think « geometry »? Maybe??
Possibly not:
In the paragraph immediately following your longish quote the OP writes:

"Y Align R1385 and Geometry R1390 will not bring the vertical markers
into proper vertical alignment even though they do have some effect on
the trace by making it less severe. Even though they never "Fix" the problem."

Raymond




Stephen
 

On Fri, Aug 7, 2020 at 11:25 PM, Raymond Domp Frank wrote:

Possibly not:
In the paragraph immediately following your longish quote the OP writes:

"Y Align R1385 and Geometry R1390 will not bring the vertical markers
into proper vertical alignment even though they do have some effect on
the trace by making it less severe. Even though they never "Fix" the
problem."

Raymond
Ok.


Michael W. Lynch
 

On Sat, Aug 8, 2020 at 08:22 AM, Chuck Harris wrote:


Typically, when a proper application of the Geometry,
Focus, Astigmatism, Trace Rotation, and other affiliated
adjustments results in a screen that just doesn't look
as pretty as we would like, it means that there has been
an outside physical influence on the CRT (READ: OOPS! BANG!
$%^! & #!!*^! Why Me???)
No signs of major trauma, but this is still a possibility. What confuses me in this regard is the fact the the unfocused spot on the CRT is beautiful and round, one of the best I have ever seen.

The Mumetal shield can get bent, or magnetized.
Not bent, but magnetized? VERY POSSIBLE. I am curious how such magnetization can occur as this issue has come up in other threads as well?

One of the steering magnets (held with Scotch tape) can fall off
of the CRT neck, causing the beam to partially strike one
many internal shield electrodes that looks like a hole in
a plate of metal.
I can look at this when I pull the CRT. Would this not distort the unfocused spot (mentioned above) somehow?

Or, it can be that we are much more nit-picky than tektronix
was originally.
I'm just not that picky. I do not think I am asking too much for the markers to be vertical and consistent across the entire viewing area (or at least within one minor division). If it was just one minor division of deviation, I could live with this, but the outermost vertical markers are about 6 minor divisions "leaned in" at the top of the marker. It is almost like looking through a "fisheye" lens.

If you carefully read the specifications in the manual, you
will often find that we have been spoiled rotten by all of
the CRTs we get that are much better than the specifications
say they could be...
I totally agree. The CRT is one of those items that is a source of constant amazement for me. How they work and withstand decades of use and many times abuse. They are a marvel.

Bottom Line, sounds like I need to pull the CRT and inspect it for obvious damage. I can try another "known good" CRT, as I have one or two that produce very good and well aligned traces in other machines.
if a "Good CRT" shows the same symptoms in this chassis, then I need to look elsewhere.

I was thinking that I had missed some other related adjustment? I guess not?

Thanks for the valuable input.
--
Michael Lynch
Dardanelle, AR


 

On Sat, Aug 8, 2020 at 04:40 PM, Michael W. Lynch wrote:


If it was just one minor division of deviation, I could live with this, but
the outermost vertical markers are about 6 minor divisions "leaned in" at the
top of the marker. It is almost like looking through a "fisheye" lens.

I'd think that's way out of range for the electrical Geometry adjustments that you mentioned earlier, so probably something is wrong of the sort that Chuck describes. It may be worthwhile to quickly check if the Geometry adjustment pot's range reaches the intended 0 to +110V. It's GND-related, so not dangerous to measure, unlike grid- and focus- voltage levels.

Re. the "fisheye" appearance: Is the distortion mostly (radially) symmetrical? In that case, I wouldn't think it could be just a moved magnet.

Trying another CRT seems to make sense. Not too much hassle with a 465/475. Just the usual care with the lingering anode voltage.


BTW, re. the first post, Michael wrote:

"Problem was C1442 was bad. High resistance and almost no capacitance, loading the +15V supply."

The high resistance and low capacitance resulted in your meter showing that the average or RMS voltage was too low so good regulation wasn't possible.
C1442 wasn't loading the +15V supply ("high resistance") but loading the 15V supply wasn't possible, since the regulation circuit received practically unbuffered rectified AC. It's probably what you meant.

Raymond


Chuck Harris
 

Hi Michael,

The spot we see is a picture of the active cathode
area, as is collimated and distorted by the grids,
anodes and lenses in the CRT.

Because of slight manufacturing variations that can
make for beam motions that would cause the beam to
completely miss the concentric holes in the shields
and anodes, there is a need for one or two steering
magnets judiciously placed on the neck of the CRT.

If one is missing, your spot could on a worst case
disappear completely. In a less drastic case, it would
chop off part of the circular spot leaving semi circular.

Tektronix used cellophane tape to hold the magnets (if any)
in place on the neck. One steers the beam horizontally,
and one steers it vertically. I have seen the cellophane
tape with empty pockets where the magnet used to go. Usually
the magnet is stuck on the inside of the mumetal shield.

The various elements after the gun that deflect, magnify,
and distort the spherically projected image into a
flat image for display on the screen all affect the
circularity of the spot by making it look like the
wind smeared its shape... they make its shape look more
comet like in appearance.

To be as distorted as you describe (I wish I read that
earlier), you have to have one of your astigmatism,
geometry, or focus elements that is electrically in the
wrong place in the tube. If it is truly like a fish eye
lens, centered on the center of the screen, you probably
have a bad power supply going to one of the pots that
drives the astigmatism, geometry or focus grids.

This is just a guess, but in any case, it is time for you
to verify that all adjustment pots in that area are properly
fed, good, and are somewhere in the center of their range.

Pay particular attention to any circuit complexity that
exists around these grids. Sometimes there are transistorized
distortion circuits that are supposed to warp things into
flat for the screen.

-Chuck Harris

Michael W. Lynch via groups.io wrote:

On Sat, Aug 8, 2020 at 08:22 AM, Chuck Harris wrote:


Typically, when a proper application of the Geometry,
Focus, Astigmatism, Trace Rotation, and other affiliated
adjustments results in a screen that just doesn't look
as pretty as we would like, it means that there has been
an outside physical influence on the CRT (READ: OOPS! BANG!
$%^! & #!!*^! Why Me???)
No signs of major trauma, but this is still a possibility. What confuses me in this regard is the fact the the unfocused spot on the CRT is beautiful and round, one of the best I have ever seen.

The Mumetal shield can get bent, or magnetized.
Not bent, but magnetized? VERY POSSIBLE. I am curious how such magnetization can occur as this issue has come up in other threads as well?

One of the steering magnets (held with Scotch tape) can fall off
of the CRT neck, causing the beam to partially strike one
many internal shield electrodes that looks like a hole in
a plate of metal.
I can look at this when I pull the CRT. Would this not distort the unfocused spot (mentioned above) somehow?

Or, it can be that we are much more nit-picky than tektronix
was originally.
I'm just not that picky. I do not think I am asking too much for the markers to be vertical and consistent across the entire viewing area (or at least within one minor division). If it was just one minor division of deviation, I could live with this, but the outermost vertical markers are about 6 minor divisions "leaned in" at the top of the marker. It is almost like looking through a "fisheye" lens.

If you carefully read the specifications in the manual, you
will often find that we have been spoiled rotten by all of
the CRTs we get that are much better than the specifications
say they could be...
I totally agree. The CRT is one of those items that is a source of constant amazement for me. How they work and withstand decades of use and many times abuse. They are a marvel.

Bottom Line, sounds like I need to pull the CRT and inspect it for obvious damage. I can try another "known good" CRT, as I have one or two that produce very good and well aligned traces in other machines.
if a "Good CRT" shows the same symptoms in this chassis, then I need to look elsewhere.

I was thinking that I had missed some other related adjustment? I guess not?

Thanks for the valuable input.


Michael W. Lynch
 

On Sat, Aug 8, 2020 at 10:34 AM, Raymond Domp Frank wrote:


I'd think that's way out of range for the electrical Geometry adjustments that
you mentioned earlier, so probably something is wrong of the sort that Chuck
describes. It may be worthwhile to quickly check if the Geometry adjustment
pot's range reaches the intended 0 to +110V. It's GND-related, so not
dangerous to measure, unlike grid- and focus- voltage levels.
I was thinking of checking the Geometry Pot, just to mark it off the list of possibilities. If this pot is somehow defective, this might affect the range of Geometry adjustment.

Re. the "fisheye" appearance: Is the distortion mostly (radially) symmetrical?
In that case, I wouldn't think it could be just a moved magnet.
The marker peaks and lines are straight (more or less) but simply tilted in at the top. The tops of markers to the left of center lean in toward the center and those to the right do the same. The markers are a mirror image of one another, with the centerline being the mirror point. If that makes any sense.

Trying another CRT seems to make sense. Not too much hassle with a 465/475.
Just the usual care with the lingering anode voltage.
Yes! I will discharge before handling or removing, I have shocked myself before. One does not quickly forget a multi kV shock.
.
BTW, re. the first post, Michael wrote:

"Problem was C1442 was bad. High resistance and almost no capacitance, loading
the +15V supply."

The high resistance and low capacitance resulted in your meter showing that
the average or RMS voltage was too low so good regulation wasn't possible.
C1442 wasn't loading the +15V supply ("high resistance") but loading the 15V
supply wasn't possible, since the regulation circuit received practically
unbuffered rectified AC. It's probably what you meant.
Actually, I used the term "loading", as that was a typical cause of low DC voltage which I was familiar with, especially when "Bad" capacitors are concerned. I "assumed" loading, but after reading your comment, I realized that actual cause of the problem was that the regulation circuit simply could not cope with the unbuffered rectified AC. I knew there was a problem, but did not know the proper term to use to describe it.

One important lesson that I learned early on, was that all the power supplies need to be within specs before trying to make anything else work. This is the one thing that seems to be overlooked by novices like myself,

After isolating the +15V supply from the instrument and still having problems, there were a very limited number of parts that could be the cause. So my diagnosis was a process of elimination of the individual components and the bulk capacitors lke C1442 are the most likely and well know failure point.

I did not attempt to test the capacitor "in circuit". As a confirmation of the failure, I tested C1442 after I had removed it. The ESR showed about 1000 ohms and the capacitance at 196pF.

As always, I appreciate the education that I receive from everyone who has replied!

Sincerely,
--
Michael Lynch
Dardanelle, AR


Michael W. Lynch
 

I am going to post some pictures of the phenomenon later today. I am also going to follow up on all the suggestion or ideas presented in the preceding conversations. I have plenty to check off the list for now.

If the Astigmatism, Geometry and Y Axis Align voltages and component values check out, then I will proceed with a CRT substitution. That will definitely help to isolate the issue to either the CRT or the Instrument circuitry.

I wish that I was better at using the proper terms to describe what I am seeing. I feel like I lead people astray when I use the wrong terms or otherwise poor wording to describe the issue. For example, my using the term "loading" when this was not a case of low voltage due to an overload in the circuit.



Thanks again for all the help!

--
Michael Lynch
Dardanelle, AR


Steph L
 

Michael, I'm not all that familiar with using Markers

So Re: "Problem is when I put my marker signal from my marker generator into the unit and set Volts/Div knob for 6-8 division vertical markers then I center the middle marker to align with the center vertical graticule line, as you move away from the center, the markers become more and more "leaned in" in at the top"

I'm assuming the Marker sig gen is fed in via Z axis coming in through P1452 Pin2 (sheet 10). There is a heap of circuitry that follows.
I guess I would be looking (with another cro) for signal folding/distorted on the positive excursions. In particular if signal is too close to +ve rail.

I was wondering what happens when you feed a short duration pulse into Ch1 for display on screen and same height?
Does the pulse behave the same way or is it only when using Markers?

Steph Lancaster
Melbourne (Aust)


Michael W. Lynch
 

I have posted pictures of the CRT in various configurations.

Signal is from a TEK TG501, into 50 Ohms and the signal is being sent to both channels.

Tilt is consistent, regardless of sweep speed or amplitude. Sine, Square and Triangle signals are distorted as well.

Displayed signal is always more narrow at the top and wider at the bottom.

https://groups.io/g/TekScopes/album?id=251792



--
Michael Lynch
Dardanelle, AR


lilacbarn
 

Micheal, you asked for ideas and I am no expert on Tek scopes like Chuck Harris and others,
but is it possible that the gain of the horizontal (x) amp is being affected by the vertical (y) amp signal in some way?
I do not suggest that I know of a way the circuits could interact in this way, but the "tilt" towards the center at the top of the screen suggests this.
Experts, please say why this is not possible.
Geoff.


Steph L
 

Thanks Michael for added detail and photos.

Process elimination; - I'd be inclined to swap Y output leads at point of connection to tube and try make sense of result.

Also, re your screen shot 475-trace-bent-6.jpeg, in terms of geometry, the image appears to be trapezoid shape. That is, the distance from left to right across the top of the pulses is shorter than across the base line, for the same number of pulses. This is a good clue. Looking at the Horizontal (X) sawtooth linearity between blanking, might offer a reason.

Sorry if you have explained this earlier, but when the Vertical Shift is used to lower the trace, does the top (trapazoid) line width "expand" out such that when the tops of the signal pulses are across the bottom of the tube face, do they now match gratical exactly as they do in that photo?? If so, it reinforces point - the tube geometry is somehow trapezoidal instead of rectangular in a 10:8 ratio.

Steph Lancaster
Melbourne (Australia)


Keith
 

Just thinking out loud here - if a scope sat on a bench (or in storage) in one position for a long time near some relatively powerful fixed nearby magnetic field, (like a big permanent magnet loudspeaker assembly, etc.) I suppose it is possible for some nearby magnetic anomaly like that to gradually induce a very specific magnetism somewhere, leading to visible misalignment? Such an unwanted magnetic influence could produce a very specific distortion and it would seem reasonable for such an influence to be most obvious on traces out at the edges of the CRT.

Maybe you could try degaussing the shield around the CRT? You can borrow my degaussing tool if you want. Sometimes it is the simple stuff. PIA to remove the shield and separate from the CRT, but that is probably the safest way to do it. Of course you would first have to remove those steering magnets like someone mentioned, or the degaussing tool would ruin them. Of course this means even more fiddling, but hey...that’s why we do this hobby, right?

Also, maybe the shield has been damaged by some mishandling in the past? Tek says, and I quote

“CAUTION This Mumetal shield has been carefully annealed after fabrication. Any shock may destroy its magnetic shielding characteristics”

So obviously it is possible for them to go bonkers. Just a thought...

Cheers,
CBG


Chuck Harris
 

I don't think so, and here is why:

When the scope is exposed to a strong magnet,
the field lines extend in loops from N to S poles.
When those flux lines confront a magnetizable metal,
they are distorted and concentrated into that metal,
making for truly weird looking flux arrangements.

On screen, such magnetizations of the mumetal
shield, which really is the only magnetizable material
anywhere near the CRT will make beam deflections in
small areas. This typically shows up as a kink
somewhere on the screen, near one edge.

_____________

_____________

___________/\

____________/

A power supply regulation failure could cause a progressive
pinch of the screen width making a "keystone" shape.

-Chuck Harris

Keith wrote:

Just thinking out loud here - if a scope sat on a bench (or in storage) in one position for a long time near some relatively powerful fixed nearby magnetic field, (like a big permanent magnet loudspeaker assembly, etc.) I suppose it is possible for some nearby magnetic anomaly like that to gradually induce a very specific magnetism somewhere, leading to visible misalignment? Such an unwanted magnetic influence could produce a very specific distortion and it would seem reasonable for such an influence to be most obvious on traces out at the edges of the CRT.

Maybe you could try degaussing the shield around the CRT? You can borrow my degaussing tool if you want. Sometimes it is the simple stuff. PIA to remove the shield and separate from the CRT, but that is probably the safest way to do it. Of course you would first have to remove those steering magnets like someone mentioned, or the degaussing tool would ruin them. Of course this means even more fiddling, but hey...that’s why we do this hobby, right?

Also, maybe the shield has been damaged by some mishandling in the past? Tek says, and I quote

“CAUTION This Mumetal shield has been carefully annealed after fabrication. Any shock may destroy its magnetic shielding characteristics”

So obviously it is possible for them to go bonkers. Just a thought...

Cheers,
CBG




Michael W. Lynch
 

All:

I went back and confirmed that the Geometry and Y Align controls were functioning correctly. Both circuits checked out good.

Reconfirmed all power supplies potentials and ripple.

Checked all physical connections to the CRT.

After that, the CRT needed to come out to see if steering magnets were used and had subsequently become dislodged (which was not the case).

Although I was holding out hope that physical damage to the CRT was not the cause, Chuck Harris hit the nail on the head with his first response to my inquiry.

Chuck Harris
Aug 8 #170283
Typically, when a proper application of the Geometry,
Focus, Astigmatism, Trace Rotation, and other affiliated
adjustments results in a screen that just doesn't look
as pretty as we would like, it means that there has been
an outside physical influence on the CRT (READ: OOPS! BANG!
$%^! & #!!*^! Why Me???)
Cause of Problem found, But problem not solved! After removing the CRT, I could detect the very faint "tinkle" of broken glass. Closer examination revealed that one of the the glass rods that supports for the Deflection plates and the geometry shield had been broken. this obviously skews the deflection plates and the geometry electrodes as well.

See the condition of the glass support rod in New photos at: https://groups.io/g/TekScopes/photo/251792/0

My next problem is that I do not have a replacement CRT. I was thinking (there I go again!) that I have one, but what I have available are a 154-0731-00 and a 154-0731-04 not the required 154-0677-10. So, I need to find a good CRT for this scope. This scope is too nice to just part out, but also worthless without a proper CRT.


Michael Lynch
Dardanelle, AR


toby@...
 

On 2020-08-10 12:25 p.m., Michael W. Lynch via groups.io wrote:
...
My next problem is that I do not have a replacement CRT. I was thinking (there I go again!) that I have one, but what I have available are a 154-0731-00 and a 154-0731-04 not the required 154-0677-10. So, I need to find a good CRT for this scope. This scope is too nice to just part out, but also worthless without a proper CRT.

Hi,

Sphere lists one. https://www.sphere.bc.ca/test/tek-crts.html

--Toby




Michael Lynch
Dardanelle, AR



Tom Gardner
 

On 10/08/20 17:25, Michael W. Lynch via groups.io wrote:
Cause of Problem found, But problem not solved! After removing the CRT, I could detect the very faint "tinkle" of broken glass. Closer examination revealed that one of the the glass rods that supports for the Deflection plates and the geometry shield had been broken. this obviously skews the deflection plates and the geometry electrodes as well.
See the condition of the glass support rod in New photos at: https://groups.io/g/TekScopes/photo/251792/0
My first scope repair, a 465, took a similarly long time. The scope was doing a passable imitation of a torch, which I eventually traced to a broken grid weld, see https://entertaininghacks.files.wordpress.com/2016/03/crtgrid.jpg I learned a lot in that process :)

I went and picked up a CRT, and the bloke let me have a decent deal on a 485 that took a long time to startup first thing in the morning. Locating the dubious startup electrolytic fixed the problem, but lead to the problem of buying more duff scopes :(


toby@...
 

On 2020-08-10 12:25 p.m., Michael W. Lynch via groups.io wrote:
All:
...
My next problem is that I do not have a replacement CRT. I was thinking (there I go again!) that I have one, but what I have available are a 154-0731-00 and a 154-0731-04 not the required 154-0677-10. So, I need to find a good CRT for this scope. This scope is too nice to just part out, but also worthless without a proper CRT.
Aaarg, sorry, ignore my message about Sphere. It's sold.

There is one on ebay UK though:
https://www.ebay.ca/itm/154-0677-10-TEKTRONIX-CRT/264704023173

--Toby



Michael Lynch
Dardanelle, AR