Tektronix 465 horizontal focus issue

[sean0118]

Okay, I have another scope to repair, this one is a Tek 465. I'll upload some more photos when I get time.


It generally seems to be working, except the horizontal is completely out of focus. On a time base of 1ms/div you can't even tell it's a square wave.

I'm just wondering if this is a known issue? Otherwise I suspect it might be the main filter capacitors, but I thought that would effect vertical focus as well?

The +15V, +5V and -8V are only meant to have less than 2mV ripple, and the +55V less than 4mV, I'm not sure about the 110V? I'm finding it difficult to measure the ripple accurately, but it looks like the 100Hz ripple on the +5V and -8V might be exceeding the spec?











Power Supply ripple:
This is using my Tek 7603 which still needs it's horizontal gain calibrated (that's why the ripple isn't aligned to the 10ms/div)

-8V rail (5mV/div):


+5V rail (5mV/div):


+55V rail (5mV/div):


+15V rail (5mV/div):


+110V rail (5mV/div):

[tggzzz]

If it was horizontal focus, that would be visible when you display a stationary dot in XY mode.

What does a sine or triangle waveform look like?

Is the triggering working correctly?

[David Hess]

The power supplies do not look bad.  That looks like a triggering problem.

1. Try testing it using the external trigger input.  It can be connected in parallel to the same source using another probe.

2. Do the trigger calibration procedure in the manual.

[sean0118]

If it was horizontal focus, that would be visible when you display a stationary dot in XY mode.

What does a sine or triangle waveform look like?

Is the triggering working correctly?

Thanks, I forgot these scopes had an X-Y mode. Looks like the vertical is also out of focus, which makes a lot of sense rather than it just being the horizontal. I had the focus knob trimmed most of the way anti-clockwise which masked the lack of focus in the vertical.

I can still connect a sig gen for the sine or triangle wave if required, but will need some setting up so will leave it until tomorrow.

Triggering seems to be working, but I haven't looked at it that closely.



X-Y Mode (with focus knob centered):

[sean0118]

The power supplies do not look bad.  That looks like a triggering problem.

1. Try testing it using the external trigger input.  It can be connected in parallel to the same source using another probe.

2. Do the trigger calibration procedure in the manual.

Thanks, I can see how it looks like a triggering issue from the initial photos, but I think it was triggered on the waveform. If I turned the intensity up I could see the rising / falling edges were really fat. I will still try using the external trigger though.

[tggzzz]

That dot is far too big (and probably too bright).

Go through the service manual procedures for astigmatism.

If those don't work, you check the resistors associated with the focus control; they can drift over time. If you replace them, ensure the resistors have an appropriate voltage rating.

[David Hess]

The 465 uses tunnel diode triggering and it is not uncommon for their trigger circuits to drift out of calibration.  It is also possible that a tunnel diode has failed but be very careful about messing with them because they are easy to damage.

[Lunasix]

Hi,
I've had a Tek465 with a focus/blanking problem. The spot was too bright, too big and adjustement had little effect. It was a small diode which was ok when tested with multimeter but was conductive like a Zener above some voltage. I don't remember where it was, I don't have anymore the scope and the schematics.

[bd139]

Agree with tggzzz here 100%. Check the whole focus and tube bias makes sense as per service manual.

If you stick it XY, stick BW limit on and turn the brightness down and try and get the dot visible using minimal brightness and check astig and focus it should be possible to get it absolutely tiny.

[bdunham7]

X-Y Mode (with focus knob centered):

What does the dot look like if you make your best effort to focus it?  And can you improve upon that with the astigmatism adjustment? (small screwdriver, to the right of the focus knob)

[T3sl4co1l]

Also does it vary in any obvious way with sweep rate (if it's ripple on a supply you haven't tested, it will show modulation at lucky frequencies; may be hard to get a beat if it's high frequency, but mains is most likely, anyways?).

Tim

[sean0118]

Thanks all, that was the best I could get by trimming the front panel focus knob, trimming it further would give me a horizontal or vertical line.

But I hadn't tried trimming the astig, I gave it a go and it helped a lot. I think it might be pretty close to normal now? Trace is maybe still a bit too fat. That was just using the front panel astig adjustment, I'll go through the calibration procedure and see if I can get it better.

I may have been jumping to conclusions assuming it was broken. Although initially the beam finder button was getting stuck in some intermediate state causing all sorts of weird issues. I fixed that with some DeOxit though. 





1KHz cal:

[sean0118]

If you stick it XY, stick BW limit on and turn the brightness down and try and get the dot visible using minimal brightness and check astig and focus it should be possible to get it absolutely tiny.

Ok thanks, I followed this and the result is below, I guess it is just a calibration issue then. I think I may have left the probe connected in my previous post, which added noise, classic rookie mistake. 

It's even a bit better than the photo shows.

[bd139]

Yeah that looks about right. At least you didn’t have to repair it 

The 465 vertical amp is a bit noisy due to the wide bandwidth compared to a lot of scopes. If you pull the bandwidth limit switch out to 20Mhz it’ll probably look sharper.

475 is even worse! (I have one)

[med6753]

Agreed with bd, looks good. And you're lucky. I have a 465B that had a CRT with apparent physical shock damage which knocked the gun assembly out of alignment which resulted in poor focus. I had to replace the CRT.

[David Hess]

So it was just the astigmatism.  That is good news although I wonder how it drifted so far out of calibration.

The 465 vertical amp is a bit noisy due to the wide bandwidth compared to a lot of scopes. If you pull the bandwidth limit switch out to 20Mhz it’ll probably look sharper.

475 is even worse! (I have one)

That is just a result of the wider bandwidth displaying greater noise.  Amplifiers with wider bandwidth have both higher total noise because of increased bandwidth, and higher spot noise and flicker noise if they rely on higher bandwidth transistors of the same type.

[sean0118]

So it was just the astigmatism.  That is good news although I wonder how it drifted so far out of calibration.

Yeah I'm not too sure, I don't really know it's history but I suspect someone tried tweaking the astig adjustment. Unless mechanical shock can cause it? There's a hole punched in the plastic on it's end, just below the fan, may have been dropped at some point.

It seems to work pretty well now though, I just bumped up the vertical gain a little bit and adjusted the CH2 step attenuation balance. I'll adjust the horizontal gain as well eventually, but I'm going to borrow a calibrator for that.

Something weird I just noticed though, I think there's a DC offet when using the ADD function. If I set CH1 & CH2 both to GND coupling and centre the traces, when I select ADD the resultant trace is offset nearly 2 divisions from centre. I've got the case back on now, but I might have a closer look at that when I cal the horizontal amp.

 

[bd139]

That’s not unusual. I suspect that the last person who did the calibration didn’t do it properly. It’s fairly easy to sort it but you have to do all the steps rather than try and tweak that on its own as there are dependencies between calibration changes.

[sean0118]

I suspect that the last person who did the calibration didn’t do it properly.

That might have been me. I couldn't follow the proper procedure for the vert amp adjustment because I don't have a 20mV ref, I'll borrow one of those and do it again. 

[bd139]

Haha  .

If you’re short just make up a fairly low impedance divider with a couple of resistors and measure it with a DMM or something. That’s what I did if I remember last time.

[tggzzz]

I suspect that the last person who did the calibration didn’t do it properly.

That might have been me. I couldn't follow the proper procedure for the vert amp adjustment because I don't have a 20mV ref, I'll borrow one of those and do it again. 

I'd simply make such a reference, by taking any available reference and adding a resistive divider. It won't need to be high frequency, and the 1Mohm//20pF scope input won't load it too much. Given that the scope is only (IIRC) 2% vertically, no great precision is required.

Get the service manual, follow the procedures while thinking about what they are trying to achieve. That will mean you can use equipment available to you rather than the specific equipment mentioned in the manual.

[Gyro]

I may have been jumping to conclusions assuming it was broken. Although initially the beam finder button was getting stuck in some intermediate state causing all sorts of weird issues. I fixed that with some DeOxit though. 

Just a note - don't get DeOxit anywhere near the Y channel input  attenuator contacts. The attenuators are on a special substrate and it's strictly IPA only for those, the cleaning procedure (if needed) is in the manual, it involves sliding IPA soaked paper between the closed contacts using just horizontal movement. Vertical movement (lifting the contact springs) will result in reduce contact pressure.

Impedances are high - I once traced a noisy input channel to a stray eyelash on the attenuator PCB.

[David Hess]

Something weird I just noticed though, I think there's a DC offet when using the ADD function. If I set CH1 & CH2 both to GND coupling and centre the traces, when I select ADD the resultant trace is offset nearly 2 divisions from centre. I've got the case back on now, but I might have a closer look at that when I cal the horizontal amp.

There is no calibration adjustment for offset in ADD mode on the 475.  You have to use the position controls after ADD mode to zero the trace.

The reason for the offset is that the CRT amplifier has some offset so centering the trace actually requires the vertical input to have an offset, and when ADD mode is used, the offset from both channels gets added together.

[bd139]

I the two channels are centered during cal properly, the offset is virtually zero when you hit add. I bet there's a large offset if you hit INV as an example.

[David Hess]

I the two channels are centered during cal properly, the offset is virtually zero when you hit add. I bet there's a large offset if you hit INV as an example.

No, because the CRT amplifier requires some offset to be centered, so each individual channel is actually offset to make a centered display, which is what happens when you adjust the vertical offset control.  You are actually setting it to some point away from zero volts to center the display.

Each channel has balance adjustments for the variable control and the 2 mV/div setting.  Channel two also has a balance adjustment for invert.

To correct the balance for ADD, the vertical CRT amplifier would require a balance control after the channel switch to adjust the trace centering with zero signal and they did not include one.  This is a common feature with their mainframe oscilloscopes because their preamplifier sections are interchangeable.  Offhand I do not know of any portable oscilloscopes that have it.