TEK465 - 10 MHz ripple

[francozappa]

Good evening everyone. I introduce myself. My name is Franco, I live in Italy and I just signed up for this forum.
I own a used TEK465 oscilloscope for repair.
The problem is the presence of a ripple at around 10 MHz which modulates the beam. The oscillation comes from the low voltage supply voltages. I have tried in various ways to eliminate it, without success.
If anyone has had the same problem or has any idea of the cause, thank you in advance. I have the schematics and the service manual.
Hi everyone.
FZ

[RoGeorge]

I have tried in various ways to eliminate it, without success.

What exactly did you tried?  Did you replaced all the old electrolytic capacitors already?

[Whales]

How big is the 10MHz ripple?  Can you provide a photo + info on what vertical settings you are using?

[pbarton]

Do you have this problem with all the metal shields in place and the scope within its aluminium case?

[francozappa]

Ciao !!
I changed ALL the tantalum capacitors of the scope and electrolytics C1534, C1404 and C1220.

[francozappa]

Ciao !!
TIME/DIV is 5microsec/DIV and VOLT/DIV is 5 mV/DIV.
I attach the photos of the single and double track.
Thanks for the tips.

[francozappa]

The problem is the same both with the oscilloscope in its metal container and outside.
Ciao !!

[BreakingOhmsLaw]

Ciao Franco!
First, get yourself some near-field-probes. These are great for hinting down gremlins like that. You can order them cheap from China.
For now, do this:
If you have access to another scope (you need a scope to fix a scope, yes), connect the ground crocodile clamp to the tip. Coil the lead into two windings, fix it with a rubber band. Now move the coil around the PCBs to pick up the source of the noise.
 The Signal seems to strong to be a harmonic. It also is around 1,2 MHz, not 10MHz. (6 peaks per 5microseconds = 1/(5microseconds/6).
The only thing that generates such high frequencies in an analog scope should be the timebase circuit, so start there.

[francozappa]

Sorry, I was wrong, the time base was set at 0.05 microsec/DIV.
I put photos of the actual situation.
Sorry again.

[francozappa]

The VOLT / DIV knob does not count, the inputs are grounded.

[tggzzz]

If you have access to another scope (you need a scope to fix a scope, yes),

Not always. If you have one working and one faulty channel, sometimes you can use the workign channel to debug the faulty channel.

connect the ground crocodile clamp to the tip. Coil the lead into two windings, fix it with a rubber band. Now move the coil around the PCBs to pick up the source of the noise.

I'll bet that you could use that technique with the scope, and notice when the "noise" increased. Yes, you would have to mentally subtract the existing noise from the noise detected by the probe, but it would give an indication of the area.

The Signal seems to strong to be a harmonic. It also is around 1,2 MHz, not 10MHz. (6 peaks per 5microseconds = 1/(5microseconds/6).
The only thing that generates such high frequencies in an analog scope should be the timebase circuit, so start there.

If poorly setup internally, oscillations can occur. There are many many trimpots and trimcaps in a 465. It may be worth locating the area, then following the manual to twiddle things. I recommend putting a red mark on a trimmer before moving it, so that it can be moved back to the original position.

[pbarton]

Try turning your internet router off. Turn off all your ethernet items.
I hope that you (or your neighbour) does not have one of those room to room 'internet over the mains powerline' devices.

[francozappa]

279/5000
Thanks for the advice.
I don't have internet devices on the electrical network, only wifi.
I add that in the laboratory, attached to the same electrical network, there are two other oscilloscopes in addition to the 465 / Ripple: another identical 465 and a 466 and neither of them presents the problem.
Ciao!

[BreakingOhmsLaw]

Ciao Franco,
that gives a a noise signal of 120MHz.

Please change the timebase and check if the noise is always 120MHz regardless of the setting of the timebase. Post results. We want to know, if changing the timebase affects the noise or the noise is contantly at 120MHz. This will help determine if it is caused by the horizontal deflection circuit or is coming from something else.

[francozappa]

Hello and thanks for the reply.
If I change the TIME/DIV time base the displayed tracks change in the sense that by decreasing the TIME/DIV the undulations are no longer distinguishable. However, I attach the photos of the signal detected with another oscilloscope on the TP+5V, TP+15V and TP-8V. Changing the time base TIME/DIV does not change the disturbance
Foto1: TP+5/1 microSEC/DIV; FOTO2: TP+5/0,05 microSEC/DIV; FOTO3: TP-8/0,05 microSEC/DIV. VOLT/DIV = 10 mV/DIV.
Ciao !!
FZ

[BreakingOhmsLaw]

If I change the TIME/DIV time base the displayed tracks change in the sense that by decreasing the TIME/DIV the undulations are no longer distinguishable.

use the X10 MAG button

[BreakingOhmsLaw]

Wait, if you can detect the noise on the power rails, it's easy to narrow it down. Disconnect the variuous boards from the power supply (start with A7 timing board) and see which one causes the oscillation.

[francozappa]

Wait, if you can detect the noise on the power rails, it's easy to narrow it down. Disconnect the variuous boards from the power supply (start with A7 timing board) and see which one causes the oscillation.

I get it.
So I disconnected the -8V power supply from the A7/TIMING BOARD board (removed the dummy resistor near J2) and the noise disappears. What can we deduce from you?

[BreakingOhmsLaw]

Well, something on that board is injection noise into the power rails. Try again if any button affects the noise.  And i mean ALL, including all the VAR settings on the timebase, magnification buttons. Try everything. Probe around on that board to see which part of the circuit is causing it.  Check around the high speed timing caps C6015 and C6048.

[francozappa]

Well, something on that board is injection noise into the power rails. Try again if any button affects the noise.  And i mean ALL, including all the VAR settings on the timebase, magnification buttons. Try everything. Probe around on that board to see which part of the circuit is causing it.  Check around the high speed timing caps C6015 and C6048.

Thank you so much for the advice.
Today I can do the tests as soon as I can, then I'll let you know.
Ciao !!

[francozappa]

Well, something on that board is injection noise into the power rails. Try again if any button affects the noise.  And i mean ALL, including all the VAR settings on the timebase, magnification buttons. Try everything. Probe around on that board to see which part of the circuit is causing it.  Check around the high speed timing caps C6015 and C6048.

The abbreviations of the capacitors you have indicated to me are not in the TIMING BOARD. I believe we are looking at different versions of the scheme. My scope has S/N 704952 .My manual is "465 Oscilloscope with options SERVICE S/N 250000 & UP". So far it seems to me that it corresponds to the instrument. However, I attach the photos of the TIMING BOARD wiring diagram and the wiring.

[francozappa]

Trials done tonight: the noise changes only by pressing MIX+AUTO.
With MIX+AUTO pressed two things happen:
1. the noise pulsates in amplitude with the horizontal frequency
This can be seen on the power rails and in other points of the circuit (eg VR1099 and VR1039).
2. the beam is always off the screen horizontally.

[pbarton]

The A and B timebase knob is connected to a long enclosed cam switch on the A7 board, which is very difficult to access.
If the A and B timebase switch contacts do not close completely, this leaves the following amplification circuits with an aerial,  which can cause parasitic oscillations.
Usually this only happens on a few timebase settings (not on all the timebase settings).
To test this theory you need to insert a stout plastic ruler behind the A7 PCB and gently force the A7 PCB towards the cam switch.
The objective being to bring the cam switch closer to the PCB, on which it is mounted.
Hopefully the cam switch contacts will close and the oscillation will disappear? Of course I could be completely wrong.

[francozappa]

Thanks for the tip, in my case the noise occurs in all the positions of the TIME/DIV knob, it seems unlikely that all the contacts are open, also because the time base works, in the sense that by rotating the knob the scan speed changes.
I think I have to concentrate on the fact that pressing the MIX+AUTO keys the noise pulsate in amplitude with the horizontal frequency and the beam leaves the screen. This means that the potential difference applied at rest on the horizontal plates is different from zero. But this depends on what .....

[BreakingOhmsLaw]

Can you show us a screenshot of that behaviour?