New Guy seeking Oscilloscope advice

[alsetalokin4017]

Looks like the cat pissed in there somehow. I've never seen corrosion like that unless something bad spilled on the parts. It looks like there may be some inter-leg conductivity happening.

I think I'd replace all three of those ICs (and inspect their sockets and maybe replace sockets as well) and also that cracked capacitor, before going much further.  Even though those parts are in the vertical circuitry... shorted pins could cause problems in other subcircuits too.

[tautech]

Gee, if this is really going to be a repair effort, should I just make a new thread in the correct place, or I guess, could it be moved?

As the OP you can move it to Repair yourself.
Very bottom of this page, bottom left: Move Topic.
Click that and follow your nose.

[tggzzz]

So I made it home and took off the cover. Most of the center (where the CRT is) is covered by the safety shield, which I didn't touch. I took a look at the parts I could see (sides and bottom) while the manual was printing, and the only thing I saw that was immediately obvious was corrosion on some of the IC legs. Looking at the manual, this is the Vertical Channel Switching section of the vertical preamp board, and  two of those are Quad NAND-gates and the other is a dual Flip-flop. No idea what they do, though. Other than that, the coating on a ceramic cap had a tiny crack, but it's right at the leg.

That corrosion looks unpleasant - but contained to the IC itself. It doesn't appear to have spread to the PCB or even the socket. If that's the case, at worst the ICs would need to be replaced, but I don't know whether that is necessary. Cleaning them with IPA might reveal the actual extent of the corrosion. Be careful not to allow any corrosion products to fall into any of the switches!

An interestion observation is that those ICs are in the ch1/ch2 trigger select circuit. As you will know from reading the manual (hint hint ), the trigger is the thing that kicks off a sweep. If you look at the manual, you will see that the ext trig input avoids those ICs and kicks off the sweep directly. So, try connecting a signal to the "ext trig" input, selecting "ext trig", and fiddling with the slope/level to see if you can kick off a sweep and/or get the "triggered" light to illuminate.

Replacing a socketed IC is trivial, provided you have a replacement. Now 74xx devices are rare, but, without checking the timing, I would try using a 74LSxx or 74xCTxx device. Be aware that 74xx devices have a 5V power supply - but in this case the +ve (pin 14 (or 16)) is not at +5V and the -ve (pin 7 (or ) is not at 0V. So do not connect a scopes probe's shield to the -ve ! Either use a multimeter, or do some mental arithmetic!

But before jumping to the conclusion that the ICs are faulty, it would be worth isolating the fault to that part of the circuit.

I followed tggzzz's advice and put the scope in X-Y mode, and lo and behold, with the right settings, the trace rotated 90 degrees and was horizontal(pic attached). With a little improvisation (I only have one probe!) I set the calibrator signal to both pins at once and got an actual 2-D shape, as well.

If the deflection is roughly equal in both directions, then the horizontal amp is OK and the problem is in the sweep circuits.

There are a lot of switches and ribbon cable connectors around the sweep circuit, and they might or might not be the cause of the problem. Some of the switches are easily accessible and others are not. Most have exposed contacts and can have grit or oxidation problems. If you really think the switches are the cause of the problem, then it is best to carefully ascertain what is still working so that you can avoid fiddling with working components.

Having said that, your audible "pop" doesn't indicate a switch problem; a capacitor is still a likely cause. Look for tantalum or electrolytic caps which are damaged.

[tggzzz]

Looks like the cat pissed in there somehow. I've never seen corrosion like that unless something bad spilled on the parts. It looks like there may be some inter-leg conductivity happening.

I think I'd replace all three of those ICs (and inspect their sockets and maybe replace sockets as well) and also that cracked capacitor, before going much further.  Even though those parts are in the vertical circuitry... shorted pins could cause problems in other subcircuits too.

The ICs are in the ch1/ch2 trigger select, and can be avoided by using the external trigger.

C437 is a PSU decoupler for those ICs, and thus probably isn't the cause of a complete sweep failure.

[The_Boots]

Noob moment alert: how do I use the external trigger? Do I just hook up any old device to the external trigger plug and choose external from the trigger menu? What kind of signal do I need to feed it?

[tggzzz]

Noob moment alert: how do I use the external trigger? Do I just hook up any old device to the external trigger plug and choose external from the trigger menu? What kind of signal do I need to feed it?

RTFM. It is effectively a third channel. Use the calibration output or anything else convenient.

[rhb]

Get a Harbor Freight toothbrush size stainless steel brush and brush the IC legs.  Place a piece of wood between the legs to to support them as you brush each side. Then tin them with solder.  You *really* don't want to replace old 7400 series logic if you don't have to.   A lot of it borders on unobtanium.  Replacing the sockets is probably a good idea.  But a diamond nail file ground down to match an IC pin might be less work.  There is also Deoxit.  I've never used it, but I've heard a lot of very favorable comments.  So I'm going to try it on the transistor sockets in my Tek and  Dumont scopes.

[Peabody]

So I made it home and took off the cover. Most of the center (where the CRT is) is covered by the safety shield, which I didn't touch. I took a look at the parts I could see (sides and bottom) while the manual was printing, and the only thing I saw that was immediately obvious was corrosion on some of the IC legs. Looking at the manual, this is the Vertical Channel Switching section of the vertical preamp board, and  two of those are Quad NAND-gates and the other is a dual Flip-flop. No idea what they do, though. Other than that, the coating on a ceramic cap had a tiny crack, but it's right at the leg.

I have three used 7400's  and four used 74LS74's that I'd be happy to send you.  But you know, since the chips are socketed, you could just take them out, clean up the pins and test them on your breadboard.  Truth tables from the datasheets.

My my guess is these are unlikely to be the cause of the pop, but you never know.  I would look again for blown caps and such.

[Peabody]

I found this week-old video on Youtube.  Possibly similar issues.  It's promised to be the first of a series on repairing his 475.

[The_Boots]

So I spent an hour or so last night trying to get used to reading the diagrams and learning what boards did what, and testing voltages. Most of the voltages came back within about 5%, which I figure is close enough, right? All voltages were taken with no input signal.
The only voltage level that was way off was the unregulated 50V which read 70 Volts. That's not 50V, but since it's unregulated, I wasn't sure that was a really big deal.
There was also some substantial corrosion on the "B Trigger Generator" IC pins 5 and 6. That's an input preamp.
Still haven't found any proper smoking guns, or... Smoked guns?
I'll keep looking. Going to start reading through that Tek Troubleshooting guide, too... Seems like good stuff.

[Cliff Matthews]

I wonder if this vintage Tek stuff had any Tin-whisker issues... maybe get out the magnifying glass and see.

[rhb]

Corrosion, cracked solder joints and electrolytic caps are the biggest problems with vintage gear like the Tek.  The trace pitch is such that tin whiskers seem an unlikely source of problems relative to those and general dirt.  I had a step in the calibrator response when using the delay on mine that I tracked down to a delay line termination resistor.  When I put the iron to  one lead of the resistor it fell out of the board!

My 465 is all socketed transistors.  In a humid environment, corrosion at the socket-lead interface is a big headache.  Unfortunately I see no easy way to address that because of the plastic sockets used in my unit.  The OP's unit appears to be either soldered or uses individual pin sockets.

A note to The_Boots.  Be very careful about cleaning switch contacts,  especially in the vertical attenuator section.  RTFM for the section of the instrument you're working on.  Also I forgot to mention, get a 12-18" section of wooden broom handle and whack the frame of the scope while watching the CRT with the scope in free running sweep mode to test for intermittent connections.  I learned this from an old Tek service guy I took my 465 to for advice.  I was quite alarmed at first, but after fixing a slew of broken solder joints it makes complete sense.  Just make sure you hit the frame and not some other part.  Also slap the sides of the scope with both hands.  This won't find an intermittent, but it will tell you you're looking for one.  Once you narrow it down to a small area an application of liquid flux and a soldering iron to all the joints is the solution.  If one joint is bad, likely several more in that area are also.

[The_Boots]

Checking for physical connections makes sense, even if it doesn't help for the sweep issues.

There's a separate issue that I kind of back-burnered, but I probably should mention it as well:
The channel 1 voltage seems very low. While channel 2 shows voltage like I would expect, channel 1 is always about 1/10 at the same V/Div. I can sometimes get it to jump just by brushing against the V/Div knob, so it seemed likely that it was a bad connection somewhere. Since channel 2 still had correct voltage, I didn't think it could be related to the sweep issue and I could tackle it once I got the sweep worked out. Of course, I'm nowhere near experienced enough to know for sure.

[tggzzz]

So I spent an hour or so last night trying to get used to reading the diagrams and learning what boards did what, and testing voltages. Most of the voltages came back within about 5%, which I figure is close enough, right? All voltages were taken with no input signal.

Table 5.5 gives the DC tolerances and the ripple tolerances. Having said that, the sweep won't be affected by voltages being a bit out. If you look at the circuit diagrams, all "precision" analogue electronics comes from the regulated supplies.

Be aware that once you change the DC PSU voltages, just about everything in the scope may need tweaking - so that is a "let sleeping dogs lie" area.

The only voltage level that was way off was the unregulated 50V which read 70 Volts. That's not 50V, but since it's unregulated, I wasn't sure that was a really big deal.
There was also some substantial corrosion on the "B Trigger Generator" IC pins 5 and 6. That's an input preamp.
Still haven't found any proper smoking guns, or... Smoked guns?
I'll keep looking. Going to start reading through that Tek Troubleshooting guide, too... Seems like good stuff.

Read the circuit description for the 475. Look at the block diagram to see how the signals flow - that will help you ignore much of the detailed circuit description.

[rhb]

You've probably got a dirty attenuator switch.  I don't recall the Tek recommended procedure for cleaning them.  But I did read, *after* doing it, that you should *not* spray tuner cleaner on the attenuator section.  When I read that I immediately washed them many times with 99% isopropyl alcohol.  That's the reason for my RTFM comment.

A safe way to clean switch contacts is to close the switch on a piece of good paper and pull the paper out. Potentially tricky to do with a wafer switch, but teasing the contacts apart with a toothpick or  bamboo fondue skewer will work if you're careful to not bend the contacts too much. You can safely push the contacts together when you pull out the paper.

Once it is working, you're going to want to check and adjust it.  I bought one of Leo Bodnar's 40 pS rise time pulsers, so I now have two Tek 106 pulsers which are surplus to my needs.

[tautech]

So I spent an hour or so last night trying to get used to reading the diagrams and learning what boards did what, and testing voltages. Most of the voltages came back within about 5%, which I figure is close enough, right? All voltages were taken with no input signal.

Table 5.5 gives the DC tolerances and the ripple tolerances. Having said that, the sweep won't be affected by voltages being a bit out. If you look at the circuit diagrams, all "precision" analogue electronics comes from the regulated supplies.

Be aware that once you change the DC PSU voltages, just about everything in the scope may need tweaking - so that is a "let sleeping dogs lie" area.

Good advice to a point.

Minor inaccuracies can often be corrected by bringing a PSU voltage/s back into spec but for now just the sweep needs be fixed.

[tggzzz]

So I spent an hour or so last night trying to get used to reading the diagrams and learning what boards did what, and testing voltages. Most of the voltages came back within about 5%, which I figure is close enough, right? All voltages were taken with no input signal.

Table 5.5 gives the DC tolerances and the ripple tolerances. Having said that, the sweep won't be affected by voltages being a bit out. If you look at the circuit diagrams, all "precision" analogue electronics comes from the regulated supplies.

Be aware that once you change the DC PSU voltages, just about everything in the scope may need tweaking - so that is a "let sleeping dogs lie" area.

Good advice to a point.

Minor inaccuracies can often be corrected by bringing a PSU voltage/s back into spec but for now just the sweep needs be fixed.

Just so, as I've written in other posts, and implied in the rest of that post.

[Ray Gianelli]

I lot of folks gravitate toward the Tek 400 series scopes for various reasons, i.e. no NLA IC's for one.  OTOH, you can get a newer Tek 2000 series for around $100.  I picked up a 2445A for $105.  Yeah, it's got plenty of ASIC's in it, but for $100 it's practically disposable.  It's a much newer scope and therefore won't have as many of the age related issues you might find in an older scope.

I know you've already bought your 475 but it's a complex piece for a novice to troubleshoot.

[rhb]

It depends a lot upon the "novice".  The quality of the Tek manual is just breathtaking.  After 20+ years I'd forgotten until I pulled it down to look at the horizontal section.  Great care went into documenting the whys and wherefores of each stage in the instrument.  It's a fantastic education in the realities of high performance analog design.

[tggzzz]

I lot of folks gravitate toward the Tek 400 series scopes for various reasons, i.e. no NLA IC's for one.  OTOH, you can get a newer Tek 2000 series for around $100.  I picked up a 2445A for $105.  Yeah, it's got plenty of ASIC's in it, but for $100 it's practically disposable.  It's a much newer scope and therefore won't have as many of the age related issues you might find in an older scope.

I know you've already bought your 475 but it's a complex piece for a novice to troubleshoot.

In my experience, having repaired 465,475,485,2445B,2465, it is swings and roundabouts.

[JohnPen]

I am not sure if this applies to Tek 475s but in the Tek 453 they use tunnel diodes for triggering the timebase.  Also Nuvistors (Valves) were used for the timebase A and B in the 453,  these do not apply to the 475.  After some searching and probing for 'no sweep' in timebase A of my 453 I discovered that the Nuvistor had failed but more importantly the tunnel diode had also failed.  I modified timebase A to duplicate the FET circuit as used in the TEK 453A but still had 'no sweep until I replaced the tunnel diode.  It appeared that even free running the timebase would not work properly unless a working tunnel diode was present.  My timebase B circuit also did not trigger but this time it was only the tunnel diode that had failed.

It would seem likely that the 'pop' you heard was, as others have suggested, the failure of a capacitor but it probably disconnected itself as well as the most of the scope is still working OK.  However that sort of transient behaviour could well have taken out a sensitive semiconductor component in the triggering circuitry which may be the cause of your problem.  Best of luck.

[tggzzz]

I am not sure if this applies to Tek 475s but in the Tek 453 they use tunnel diodes for triggering the timebase.  Also Nuvistors (Valves) were used for the timebase A and B in the 453,  these do not apply to the 475.  After some searching and probing for 'no sweep' in timebase A of my 453 I discovered that the Nuvistor had failed but more importantly the tunnel diode had also failed.  I modified timebase A to duplicate the FET circuit as used in the TEK 453A but still had 'no sweep until I replaced the tunnel diode.  It appeared that even free running the timebase would not work properly unless a working tunnel diode was present.  My timebase B circuit also did not trigger but this time it was only the tunnel diode that had failed.

Except for the nuvistor, that applies to the 475.

To proceed down that path, the OP needs to see where signals are entering/leaving each of the major functional blocks shown in the block diagram.

It would seem likely that the 'pop' you heard was, as others have suggested, the failure of a capacitor but it probably disconnected itself as well as the most of the scope is still working OK.  However that sort of transient behaviour could well have taken out a sensitive semiconductor component in the triggering circuitry which may be the cause of your problem.  Best of luck.

It can do - and sometimes it takes out a resistor as well I've seen that in a 485 on the "hidden" 13V rail, and in a 2465 where the Rifa firework took out a resistor and a 1cm diameter patch of prepreg

[JohnPen]

At the time I had nothing fast enough to see the trigger output of the tunnel diode so after considerable checking I bit the bullet and purchased some tunnel diodes from somewhere in Russia.  Replacing the tunnel diode on the A timebase cured the problem but I still couldn't see the trigger pulse.

[Gyro]

Be very careful messing about with the tunnel diodes - in fact I would have advised not touching them at all if it wasn't too late (Not too late for the OP). They are delicate and sensitive parts. It's unlikely that the ones from Russia would have matching characteristics.

http://w140.com/tekwiki/wiki/Tunnel_diodes

[JohnPen]

I basically selected one from the attached list of Russian Tunnel diodes list.  I went for the one with the closest current rating to the TD 4.7ma as 'sort of specified' in the 453 parts list.  The ones I used were identified as 1l304A and these seemed to work fine fortunately.  Certainly triggering appeared to work properly and reliably for both A and B timebases.   I don't know if these tunnel diodes are available now as this repair was done about 4 years ago.  I still have 4 out of the 6 I purchased as spares if anyone is desperate.

Edit: Oops I forgot the attachment.

Edit 2:  Found my original receipt from 2014 so  the actual identity of the TD is Quantity     
   
GI304A Ge Amplifier Tunnel Diode military USSR Lot of 6 pcs