Topics

7904 excessive shadow/flare

unclebanjoman
 

Hi,
I've just repaired a 7904. Usual problem, SMPS not working.
Then I've checked all voltages, HV voltages, etc.. all it's fine, very good.
Just some adjustments to Z board, focus and so on.
The scope is in very good conditions, the trace is well defined, bright and sharp.

But there is an annoying problem that I don't think is normal.

It is the usual shadow / halo / flare effect but while in my two 7854s and 465s that already own this effect only manifests itself with somewhat high brightness, in this particular 7904 with low/normal brightness the halo/flare effect it is already visible and definitely annoying.
To give an idea I took a picture that you can see at

https://groups.io/g/TekScopes/album?id=90746

I reiterate that the photo was taken with the intensity controls were set for a normal brightness even if in the photo it seems that the brightness is too high.

I've compared side-by-side my 7904 with my 7854 (they should use the same tube model) :while the 7854 the flare/halo effect is unnoticeable, the 7904 is definitely visible already with moderate brightness.

I'm puzzled.
Could it be the crt damaged by some mechanical shock?
What do you think about it?

Thanks,
Max

radioconnection@...
 

Looks like an out of phase ghost image? Signal reflection due to SWR on the probe cable?

Try another signal source and different cable.

Pete

unclebanjoman
 

No, the artifact is visible with any type of signal.
It also appears using the signal standardizer. I used the latter today to perform some calibrations on my 7904 and I was very irritated by this halo / flare effect.
With sweep rates of 0.1-0.5 sec and no signal applied is clearly visible the usual "globus" leading/lagging the spot.
Already with moderate brightness the artifact is visible.

Max

Chuck Harris
 

The 7904, because of its very high anode voltage, and its
mesh lens is one of the worst scopes in existence for reflections
and flare inside of the CRT. But I like it anyway.

-Chuck Harris

unclebanjoman wrote:

No, the artifact is visible with any type of signal.
It also appears using the signal standardizer. I used the latter today to perform some calibrations on my 7904 and I was very irritated by this halo / flare effect.
With sweep rates of 0.1-0.5 sec and no signal applied is clearly visible the usual "globus" leading/lagging the spot.
Already with moderate brightness the artifact is visible.

Max



Roger Evans
 

To my eye each of the shadow peaks and troughs of the sine wave lies on a straight line joining the real peak/trough to the CRT centre, but farther out. This should be very obvious at say 100msec/div. I would guess the probable cause is the CRT cathode voltage dropping out of regulation due to a HV capacitor failure. If you don't have a HV probe you should be able to see the ripple on the feedback to the regulator.

Regards,

Roger

unclebanjoman
 

@Chuck:
I agree that. But 7854 and 7904 share the same CRT model (154-0644-05).
Why in my 7904 the exhibits more, more halo/flare effects than my 7854???
There's really a big difference between the two.
It's really the first time I've ever been disturbed by such an effect. In all my other oscilloscopes (465, 475, 7633, 7854) this effect is almost imperceptible.


@Roger:
As I stated previously, all voltages (high and low) are perfetly O.K. (I recapped ALL the SMPS capacitors with new Nichicon PWM grade ones).
Checking the -2960V catode supply (HV test point) with my Fluke H.V. probe I measured -2992V. It seems somewhat reasonable value to me, since Tek specs stated that this voltage should be -2960 V +/- 1%.
I'm unable to check H.V ripple because I don't know where to probe. The manual doesn't mention that check.

Max

unclebanjoman
 

Just to show what happens, I made this short video. Intensity controls are set to a very moderate brightness but the effects is noticeable:

https://www.dropbox.com/s/tfy91d8tq1b6ql9/7904_shadow_flare%20003.AVI?dl=0

Max

Chuck Harris
 

I know that, but my 7904's beat my 7854 in the flare department
by easily double. They also have much brighter maximum intensity
than does the 7854.

I haven't explored the differences, but the 7904's all have a worse
time than *any* other scope I have ever seen.

As to what it looks like, if you go to a slow sweep, something on
the order of 20ms/div, and allow a single bright spot to draw a
center-line trace, you will see several other spots moving both
faster, more diffuse, and towards the outer reaches of the CRT screen.

It is clear that they all are "tethered" at the center of the
screen.

It is the inevitable cost of having such a high anode voltage with
a mesh lens. The mesh lens is an imperfect lens. It produces a
repeating series of ghosts images that are at intervals from the
desired image. The intervals are related to the beam size, and the
mesh's pitch. I think they are like the wavelets you see when a
collimated light source is aimed through a narrow slit... which is
exactly how a mesh (window screen) lens works.

-Chuck Harris

unclebanjoman wrote:

@Chuck:
I agree that. But 7854 and 7904 share the same CRT model (154-0644-05).
Why in my 7904 the exhibits more, more halo/flare effects than my 7854???
There's really a big difference between the two.
It's really the first time I've ever been disturbed by such an effect. In all my other oscilloscopes (465, 475, 7633, 7854) this effect is almost imperceptible.


@Roger:
As I stated previously, all voltages (high and low) are perfetly O.K. (I recapped ALL the SMPS capacitors with new Nichicon PWM grade ones).
Checking the -2960V catode supply (HV test point) with my Fluke H.V. probe I measured -2992V. It seems somewhat reasonable value to me, since Tek specs stated that this voltage should be -2960 V +/- 1%.
I'm unable to check H.V ripple because I don't know where to probe. The manual doesn't mention that check.

Max



Chuck Harris
 

That is a 100% normal image for a 7904 scope.

-Chuck Harris

unclebanjoman wrote:

Just to show what happens, I made this short video. Intensity controls are set to a very moderate brightness but the effects is noticeable:

https://www.dropbox.com/s/tfy91d8tq1b6ql9/7904_shadow_flare%20003.AVI?dl=0

Max



Chuck Harris
 

That's wonderful, but the flare exists when there is only the single
spot on the screen. Why in your theory, does the center spot not
change?

This is a completely normal effect caused by the mesh lens, and the
high energy electron beam.

The high energy beam allows this diffraction to be high enough energy
to light up the phosphor. The slow speed allows the phosphor even
more time to bank up its electrons and make light.

This is normal for the 7904 CRT. It happens on my 7904's, and my
7854 to exactly the same degree.

If you don't like it, turn the intensity down to a more reasonable
value.

-Chuck Harris

Roger Evans via Groups.Io wrote:

To my eye each of the shadow peaks and troughs of the sine wave lies on a straight line joining the real peak/trough to the CRT centre, but farther out. This should be very obvious at say 100msec/div. I would guess the probable cause is the CRT cathode voltage dropping out of regulation due to a HV capacitor failure. If you don't have a HV probe you should be able to see the ripple on the feedback to the regulator.

Regards,

Roger



Roger Evans
 

My comments about cathode voltage regulation were based on seeing a similar problem (and fixing it) in a much lower bandwidth scope without PDA.

I hardly ever use my 7904 at low sweep speeds, but lo and behold I get a similar displaced, diffuse arc (almost a complete circle) at 200msec/div. Poor regulation of the cathode voltage gives a diagonal line (or just a bright spot and a weaker diffuse spot) pointing to the CRT centre. When using the 7904 normally. I am used to a little 'flare' if the brightness is too high but it is never a problem and never a clearly defined second trace.

For what it is worth the HV regulator test point is TP1635, connected to pin 3 of the op-amp U1635 (schematc <11> CRT Circuit). It should be very close to 0V DC, it is difficult to say what ripple is excessive unless you can find a reference in the manual. The AC divider ratio at TP 1635 is very different to the DC divider ratio because of R1642/C1642. A DC measurement of the cathode voltage may not show a problem, especially if you have a digital meter, it mostly samples the correct, plateau, voltage and occasionally catches the dropoff before the capacitors charge again. An analogue meter would show the average voltage to be slightly low.

Roger

unclebanjoman
 

On Thu, May 23, 2019 at 02:42 PM, Chuck Harris wrote:


I know that, but my 7904's beat my 7854 in the flare department
by easily double. They also have much brighter maximum intensity
than does the 7854.

I haven't explored the differences, but the 7904's all have a worse
time than *any* other scope I have ever seen.
O.K. indeed the artifact is two times more pronounced with my 7904 with respect to my 7854 (grossly speaking).
The question that arise is: why?
Both the two models share the same tube type.
The post acceleration voltage is the same (21 kV).
Catode voltage is the same.(I presume).

If this is due to the dome mesh, where is it located on the diagram?
is there a regulation that affects it?
On the 7904 scheme I also see an adjustment for the preset shield volt (I adjusted as per manual): what's the use?

Max

Chuck Harris
 

The dome mesh is an anode, though it is called a "Slot Lens", on
the schematic. It is at pin 4 on the CRT.

I don't know if "Slot Lens" is the more formal name, or if perhaps
it isn't a dome?

The traditional dome mesh lens, as I have seen them is a super fine
stainless steel window screen with a pronounced bow in the screen
direction. Maybe the 7904 is using a special lens that only magnifies
the Y direction, and not the X direction, using slots instead of holes,
like the mesh lens?

I haven't broken a 7904 CRT as yet, so I haven't seen the lens.

They use two quad pole lenses to focus the beam before and after
the vertical plates.

I looked at my 7904 and 7854 more closely, and the 7904 is adjusted
to have a much hotter beam, just about always, than the 7854. The
way I can tell, is the beam spot, although P31 phosphor, is always
a lot bluer than the 7854, which is simply a boring green spot on a
green trace.

The 7854 is also 100MHz lower bandwidth than the screaming 500MHz
7904. It is different, even though it uses the same CRT.

And, (there is always an and with me...) the 7854 is 10 years newer
than the 7904. Surely there were some CRT improvements in that time?

-Chuck Harris

unclebanjoman wrote:

On Thu, May 23, 2019 at 02:42 PM, Chuck Harris wrote:


I know that, but my 7904's beat my 7854 in the flare department
by easily double. They also have much brighter maximum intensity
than does the 7854.

I haven't explored the differences, but the 7904's all have a worse
time than *any* other scope I have ever seen.
O.K. indeed the artifact is two times more pronounced with my 7904 with respect to my 7854 (grossly speaking).
The question that arise is: why?
Both the two models share the same tube type.
The post acceleration voltage is the same (21 kV).
Catode voltage is the same.(I presume).

If this is due to the dome mesh, where is it located on the diagram?
is there a regulation that affects it?
On the 7904 scheme I also see an adjustment for the preset shield volt (I adjusted as per manual): what's the use?

Max



unclebanjoman
 

O.K.! Thanks to everyone for the comprehensive answers, very instructive to me!

Undoubtedly the driving circuits of the 7904 and 7804 CRTs are completely different and this could partly explain the different behavior regarding the halo / flare effect.
Over time I will get used to the remarkable brilliance of my new 7904 (and his associated halo /flares) which, moreover, is really in excellent condition and after careful calibration it matches perfectly the manual's specifications.

My 7904 has serial number B269718; reading the manufacture date of a pair of TI SNxxxx ICs, it should be from 1979.

If another 7904 to be repaired happens to me, I will be curious to observe the halo / flare effect, whether it is identical or not. I will let you know if the case.

Thanks again,
Max

 

Hi Max,
The expansion mesh was an engineering compromise which had significant drawbacks but all the alternatives had worse drawbacks.
Tek scopes were always known for their razor sharp trace. When the 7000 series was being designed it became apparent that the high frequencies the scope had to attain were going to require a new breakthrough if the length of the CRTs were going to be reasonable. The result was the expansion mesh which, unfortunately traded the super sharp trace for higher bandwidth.

An alternative is available however. The 7104 does not use the expansion mesh. It uses a box lens so the trace is finer. It also has the Micro Channel Plate which improved the brightness of the trace so much it was now possible to see a single 300pSec pulse in normal room light at 200pSec/Div sweep speed.

You need to get yourself a 7104. There is so much more it can do than a 7904.

Dennis Tillman W7PF

-----Original Message-----
From: unclebanjoman, Sent: Friday, May 24, 2019 6:21 AM

O.K.! Thanks to everyone for the comprehensive answers, very instructive to me!

Undoubtedly the driving circuits of the 7904 and 7804 CRTs are completely different and this could partly explain the different behavior regarding the halo / flare effect.
Over time I will get used to the remarkable brilliance of my new 7904 (and his associated halo /flares) which, moreover, is really in excellent condition and after careful calibration it matches perfectly the manual's specifications.

My 7904 has serial number B269718; reading the manufacture date of a pair of TI SNxxxx ICs, it should be from 1979.

If another 7904 to be repaired happens to me, I will be curious to observe the halo / flare effect, whether it is identical or not. I will let you know if the case.

Thanks again,
Max
--
Dennis Tillman W7PF
TekScopes Moderator

unclebanjoman
 

Thank you for your explanation Dennis. Very instructive!
I take this opportunity to inform you that a few weeks ago I sent you an e-mail regarding the repair of my 7854 smps (finished with success), but I have not received a reply.
Maybe it was lost due to an antispam or blacklisted?

Thanks again,
Max

 

Hi Chuck and Max,
I have a more complete answer about the cause of the halo / flare after speaking to my friend. He designed the vertical deflection plates of the 7104.

A very important consideration with CRTs 30 or more years ago was their photographic writing rate since photographic film was the only way to capture fast transient events for further study. Photographic writing rate could be increased by raising the accelerating voltage of the CRT. Faster scopes, such as the 7904, required higher HV to transfer enough energy to each electron in the beam since the beam was sweeping across the CRT at faster sweep settings in the 7904.

Inside the CRT the electrons ae accelerated by the HV field. The first thing they strike is a very thin conductive aluminum layer. The aluminum is so thin most of the electrons pass through it and strike the phosphor where they excite the electrons in the phosphor atoms causing them to emit visible photons. The purpose of the aluminum layer is to draw off the charge which would otherwise accumulate on the phosphor layer and repel the beam of electrons.

Unfortunately, as the electrons in the beam are passing through the layer of aluminum there is a small chance some of them will encounter a nucleus. When that happens the electron is repelled (scattered at a random angle) with some force back out of the aluminum layer and into the vacuum of the CRT. It quickly begins to feel the pull of the HV field again and it starts accelerating towards the aluminum again, but it is no longer part of the original focused beam of electrons. It can be scattered fairly far from the original beam by the nucleus it hit. These back scattered electrons are now accelerated toward the aluminum some distance away from the main beam and in a completely unfocused form. This is what is causing the halo/flare you are seeing.

The halo / flare is the same shape as the main focused beam but it is fuzzier and removed from the main trace on the CRT.

Dennis Tillman W7PF

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Chuck Harris
Sent: Wednesday, May 22, 2019 9:11 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 7904 excessive shadow/flare

The 7904, because of its very high anode voltage, and its mesh lens is one of the worst scopes in existence for reflections and flare inside of the CRT. But I like it anyway.

-Chuck Harris

unclebanjoman wrote:
No, the artifact is visible with any type of signal.
It also appears using the signal standardizer. I used the latter today to perform some calibrations on my 7904 and I was very irritated by this halo / flare effect.
With sweep rates of 0.1-0.5 sec and no signal applied is clearly visible the usual "globus" leading/lagging the spot.
Already with moderate brightness the artifact is visible.

Max
--
Dennis Tillman W7PF
TekScopes Moderator

Craig Sawyers
 

I have a more complete answer about the cause of the halo / flare after speaking to my friend. He
designed the vertical deflection plates of the 7104.
Dennis Tillman W7PF
I thought the secondary emission halo was a pretty well-known and understood mechanism. But maybe that
knowledge dropped off the perch with the passage of decades ;-)

Craig

unclebanjoman
 

Thanks again Dennis,

It is always worthwhile to review these things.

So we can say that the halo/flare effects is to both the dome/expansion mesh and the secondary emission, right?

Max

Chuck Harris
 

Hi Dennis,

The mind pictures your description makes are compelling:

I can see the angle the beam impinges on the screen causing the
electrons to reflect at the equal but opposite angle, and the
"wind" of the PDA catching them and dragging them back down to
again splatter on the screen: further to the left when the beam
is left of dead center, dead center when the beam is centered,
and further to the right when the beam is right of dead center...
giving the appearance of a fuzzy magnified version of the original
sweep.

One should be able to calculate the distance the electrons bounced
from the screen by measuring both the distance from screen dead center
of the original spot, and the distance from screen dead center of the
flare.

Pretty wild, Mr Tillman!

-Chuck Harris

Dennis Tillman W7PF wrote:

Hi Chuck and Max,
I have a more complete answer about the cause of the halo / flare after speaking to my friend. He designed the vertical deflection plates of the 7104.

A very important consideration with CRTs 30 or more years ago was their photographic writing rate since photographic film was the only way to capture fast transient events for further study. Photographic writing rate could be increased by raising the accelerating voltage of the CRT. Faster scopes, such as the 7904, required higher HV to transfer enough energy to each electron in the beam since the beam was sweeping across the CRT at faster sweep settings in the 7904.

Inside the CRT the electrons ae accelerated by the HV field. The first thing they strike is a very thin conductive aluminum layer. The aluminum is so thin most of the electrons pass through it and strike the phosphor where they excite the electrons in the phosphor atoms causing them to emit visible photons. The purpose of the aluminum layer is to draw off the charge which would otherwise accumulate on the phosphor layer and repel the beam of electrons.

Unfortunately, as the electrons in the beam are passing through the layer of aluminum there is a small chance some of them will encounter a nucleus. When that happens the electron is repelled (scattered at a random angle) with some force back out of the aluminum layer and into the vacuum of the CRT. It quickly begins to feel the pull of the HV field again and it starts accelerating towards the aluminum again, but it is no longer part of the original focused beam of electrons. It can be scattered fairly far from the original beam by the nucleus it hit. These back scattered electrons are now accelerated toward the aluminum some distance away from the main beam and in a completely unfocused form. This is what is causing the halo/flare you are seeing.

The halo / flare is the same shape as the main focused beam but it is fuzzier and removed from the main trace on the CRT.

Dennis Tillman W7PF