Tektronix 2465B oscilloscope teardown

[David Hess]

If I use auto scale, or any of the auto measurements, it momentarily blanks the trace (I assume this is normal), but when it's done, the trace is either very dim, or invisible.  I can get the trace back by turning up the intensity, but the same thing happens again if I repeat the auto operation.  After enough times, it needs to be near maximum to get a visible trace.  Reducing the intensity control to minimum, and turning it up restores its normal range.  I assume this isn't normal.

This is completely normal and how the 4 channel 2247A series work as well.

After the oscilloscope sets the intensity to nominal, it requires you to turn the intensity control through that setting before it will allow manual control of the intensity again.  This prevents the zero or offset of the manual control from shifting which could be a little confusing.

Modern oscilloscopes use an infinite turn control so this would not be an issue for them because the zero or offset is always relative anyway but Tektronix used a 300 degree potentiometer for digital control; rotary encoders were very expensive back when these oscilloscopes were made.

[SoundTech-LG]

go through this alot...  but it acts exactly as you describe. It's just kind of strange. The 2467B also has the added annoying "feature" of blanking or reducing the intensity to nearly nothing as a protection to the CRT from burns, every 90 seconds or so. Irritating. This really only helps those who don't use the intensity knob to lower the level when leaving a high intensity trace on the screen. The rest of us suffer.

[GerryBags]

Hi all, just received a very clean 2465B with a perennial problem: Start-up proceeds right up until the ADD led comes on and it stops. A/B TRIG brings about normal front-panel operation but no display or trace on the CRT, but it is glowing. So I have a look at A5 around the DAC area and look what I find! Green pins, and one corroded torx bolt are the only real evidence of electrolyte release, but thankfully I took pictures, and I'll lay ten-to-one that that resistor there has opened, or one near-by.

One question I have: does anyone know when these SMD control boards first started failing? This machine was last cal'd (by Tek, no less, it was owned by Sony Europe) in '09, when this particular A5 board was already 19 years old. Considering SMD hadn't been around that long in 1990 that's really not bad going.

With any luck that will be the only issue (apart from the dreaded cracked Rifas of course) this baby has!

[GerryBags]

This may be a redundant question, but would you guys change the Dallas NVRAM while you had the board off. I'm really hoping someone is going to say: "No, Gerry, you don't need to learn how to read and program cal data just yet." Anyone?

[GerryBags]

Anyone know what it means when a 2465B bleeps its front lights? Fan's running and it starts up like normal until it starts that, the blinking of it's front light i mean.

Start with the basics and work from there. Make absolutely certain all your supply voltages are within spec as far as voltage and ripple. J119 on the main board is a convenient location to check them. I don't have the manual in front of me at the moment so I can't tell you what page it's on.

So then, the blinking points to nothing specific then, therefore the long way it is. Thank you that is exactly what i wanted to know, the 70s were easier to work on 

https://www.eevblog.com/forum/repair/tektronix-2465b-repair/

i think I might have found a clue for you, Neo. Having just built a dim-bulb tester recently I was keen to use it to power up the 2465B I mention above. It didn't like it. The bulbs brightened, fans on, the leds on the 'scope all came on, then all went off, fan off, then it tries to do it again. It starts as normal (barring the missing display) when plugged directly into the mains. I'm thinking that the cold resistance of the filaments in series with the inrush current limiters, which are NTC thermistors, is stopping the switching psu from starting, and that may be the issue that is affecting your 'scope.

On my 2467B the larger thermistor (RT1010) had drifted high and the resistor in parallell (R1010) with it had opened. If your R1010 has drifted high rather than opened it could be having a similar effect to my dim-bulb tester and not allowing enough current through to get the switcher started.

[tggzzz]

Anyone know what it means when a 2465B bleeps its front lights? Fan's running and it starts up like normal until it starts that, the blinking of it's front light i mean.

Start with the basics and work from there. Make absolutely certain all your supply voltages are within spec as far as voltage and ripple. J119 on the main board is a convenient location to check them. I don't have the manual in front of me at the moment so I can't tell you what page it's on.

So then, the blinking points to nothing specific then, therefore the long way it is. Thank you that is exactly what i wanted to know, the 70s were easier to work on 

https://www.eevblog.com/forum/repair/tektronix-2465b-repair/

i think I might have found a clue for you, Neo. Having just built a dim-bulb tester recently I was keen to use it to power up the 2465B I mention above. It didn't like it. The bulbs brightened, fans on, the leds on the 'scope all came on, then all went off, fan off, then it tries to do it again. It starts as normal (barring the missing display) when plugged directly into the mains. I'm thinking that the cold resistance of the filaments in series with the inrush current limiters, which are NTC thermistors, is stopping the switching psu from starting, and that may be the issue that is affecting your 'scope.

On my 2467B the larger thermistor (RT1010) had drifted high and the resistor in parallell (R1010) with it had opened. If your R1010 has drifted high rather than opened it could be having a similar effect to my dim-bulb tester and not allowing enough current through to get the switcher started.

An SMPS, unlike a selenium rectifier, will attempt to deliver constant output power. If the input voltage is abnormally low, the input current will be abnormally high - and that can cause "problems".

[andy2000]

I tested it on a late 2465B and a late 2445B, both set the intensity to the same intensity as adjusting the intensity knob to the mid-position. On one this is fine, on the other it is to dark. But both are equal to the mid-position if adjustet by hand.

Re-adjusting grid bias and the other intensity adjustments (see the service manual) so that mid-position of the intensity pot gives a "normal" intensity will help.

It's good to know it's normal.  I'll readjust it to provide usable intensity at mid range.  It seems strange that it couldn't just remember the previous intensity value and recall it when it's done. 

[andy2000]

Hi all, just received a very clean 2465B with a perennial problem: Start-up proceeds right up until the ADD led comes on and it stops. A/B TRIG brings about normal front-panel operation but no display or trace on the CRT, but it is glowing. So I have a look at A5 around the DAC area and look what I find! Green pins, and one corroded torx bolt are the only real evidence of electrolyte release, but thankfully I took pictures, and I'll lay ten-to-one that that resistor there has opened, or one near-by.

One question I have: does anyone know when these SMD control boards first started failing? This machine was last cal'd (by Tek, no less, it was owned by Sony Europe) in '09, when this particular A5 board was already 19 years old. Considering SMD hadn't been around that long in 1990 that's really not bad going.

With any luck that will be the only issue (apart from the dreaded cracked Rifas of course) this baby has!

I started seeing numerous problems with leaking SMD caps in consumer devices as long ago as the mid 90's.  I'm not sure when these started to leak, but I would guess they had been leaking for many years before symptoms started to show up.  I doubt they even opened your scope when they calibrated it if it met specs when they tested it. 

I don't think I've ever seen an SMD electrolytic made before the late 90's that wasn't bad.  My suspicion is that the seals in the early SMD caps couldn't take the heat from the reflow oven.

You'll have to look carefully for open traces.  I had to remove the DAC to inspect underneath, and verify the continuity of all traces in the affected areas with an ohm meter.  I ended up finding about 5 open, or nearly open traces.

Definitely backup your Dallas NVRAM ASAP.  Mine failed a few days after I started working on it.  Unfortunately, I can't find the backup I thought I made when I first bought the scope a couple of years ago! 

[tggzzz]

I tested it on a late 2465B and a late 2445B, both set the intensity to the same intensity as adjusting the intensity knob to the mid-position. On one this is fine, on the other it is to dark. But both are equal to the mid-position if adjustet by hand.

Re-adjusting grid bias and the other intensity adjustments (see the service manual) so that mid-position of the intensity pot gives a "normal" intensity will help.

It's good to know it's normal.  I'll readjust it to provide usable intensity at mid range.  It seems strange that it couldn't just remember the previous intensity value and recall it when it's done.

The previous value might be invalid, since the signal and hence timebase may have significantly changed.

[cheeseit]

This may be a redundant question, but would you guys change the Dallas NVRAM while you had the board off. I'm really hoping someone is going to say: "No, Gerry, you don't need to learn how to read and program cal data just yet." Anyone?

Most definitely. At the very least get a backup of your cal data, either by using the exerciser and record the data or by reading its content. That will be far easier than going through calibration if you lose your data. You'll find all the info you'd need on how to do it in this very thread.

[GerryBags]

I ordered the little TL866 mini programmer about half an hour ago and have just finished watching Dave's review.  this thread has already come in handy a couple of times, so a big thanks to all the people who've contributed to it!

[tautech]

This may be a redundant question, but would you guys change the Dallas NVRAM while you had the board off. I'm really hoping someone is going to say: "No, Gerry, you don't need to learn how to read and program cal data just yet." Anyone?

Most definitely. At the very least get a backup of your cal data, either by using the exerciser and record the data or by reading its content. That will be far easier than going through calibration if you lose your data.

I was always to chicken to get one of these great scopes (and part of me would still like one) for the risk of losing the Cal data but the above is excellent advice. 
Make some record, any record is better than none so when the Dallas battery dies you're not left totally high and dry.

[BravoV]

One question I have: does anyone know when these SMD control boards first started failing? This machine was last cal'd (by Tek, no less, it was owned by Sony Europe) in '09, when this particular A5 board was already 19 years old. Considering SMD hadn't been around that long in 1990 that's really not bad going.

There is post few pages back, similar to your problem, that the leaked liquid from the electrolytic cap also ruined the PCB traces and pins under and surrounding components  , here is the affected board photo and the description, just beware.

[BravoV]

I was always to chicken to get one of these great scopes (and part of me would still like one) for the risk of losing the Cal data but the above is excellent advice. 

Still waiting you to join the club Rob. 

[tautech]

I was always to chicken to get one of these great scopes (and part of me would still like one) for the risk of losing the Cal data but the above is excellent advice. 

Still waiting you to join the club Rob. 

Once upon a time (fairy tale starting  ) I did relish fixing scopes but the demands of what I do now takes away all that spare time. 
TBH I probably never will as there's only a few months now to wait for something new in an entirely different league.
If I'm real lucky I'll get one to beta test but probably not. 
Anyways, I've started saving for it.......

[BravoV]

All,
  My scope is currently fully assembled, so I could not look at the inverter board. However, I did consult my 2445B manual (also covers 2455B) and saw the same component size mismatch as shown in previous posts. I also have a copy of the service manuals for the 2445 and the 2465. Those have the diagrams with the physical size correct, but positioned slightly different. According to all three manuals I have, the parts list C1114 and C1115 as the same value. When looking at the 2445 board layout, the physical size of both C1114 and C1115 are the same, with C1132 being physically smaller. According to what I see in my manuals, C1132 is supposed to be 10uf@160v, C1115 is supposed to be 250uf@20v.

Mitch

It may be that only the "B" versions are affected judging from your layouts.

As stated before, the BOM and schematics are correct. Here are annotated uncorrected and corrected versions of fig 10-13 from service manual 070-6863-01

Uncorrected:


Corrected:

Re quoting Howardlong's post #475 (HERE) above as I think this is important one, but with this forum's local attachment as these images are currently hosted at Photobucket.com, my browser can't view it.

Re-attaching the photos here below and also edited Howardlong's post above with the images from the attachments below.

Just worry if someday photobucket.com change their business, as like other problems we've encountered in this forum at images hosted externally that are gone forever.

[GerryBags]

One question I have: does anyone know when these SMD control boards first started failing? This machine was last cal'd (by Tek, no less, it was owned by Sony Europe) in '09, when this particular A5 board was already 19 years old. Considering SMD hadn't been around that long in 1990 that's really not bad going.

There is post few pages back, similar to your problem, that the leaked liquid from the electrolytic cap also ruined the PCB traces and pins under and surrounding components  , here is the affected board photo and the description, just beware.

Yes, that was what made me go back and look at the green pins on mine. I can only see a couple of other affected components nearby, none of the ICs have any corrosion on their pins but I will check the continuity of all the traces in that area. Mine looks like a much milder case, fortunately.

I saw a YT video last night of a guy fixing up a TDS540 with no display, and it was caused by exactly the same thing: corroded pins on components and ICs near SMD electrolytics that look normal but have green pins. It looks like heat is causing the caps to break their seals while still working, as the caps aren't dead short after removal but the capacitance drifts with applied voltage. Electrolyte vapor is going to affect the highest current pins near it in that scenario I think (please correct me if not) which seems to fit with the observed damage.

I've ordered some of that Quik-Chip SMD removal alloy, 2 gold pin sockets and 2 Maxim DS1225Y-200IND+, as I may as well do the A5 on my 2467B at the same time. I thought I'd finished re-capping the PSU on that.... until I saw one little, cracked, RIFA on A3 right between the honking great transistor on its own heatsink and transformer T1050 that I'd put off due to its gittish location. I could have sworn I bought Y2 2200pF caps but could not find them anywhere, so it is still in bits waiting for parts.... the 2465B will need the same caps changing out anyway. I had some caps that were close in value, also safety caps, but as these're doing EMI filter duty on the Inverter I thought the value was going to be more critical than, say, bypass caps. Am I right? Better safe than sorry?

[cheeseit]

At first I thought my A5 didn't look too bad but upon close examination I ended up removing quite a few components, followed by careful cleaning of both components, traces and pads before putting all the components back again. I used lots of alcohol, solderwick and a glass fiber brush for cleaning. Attached image shows what needed cleaning and I believe I reflowed a bunch of components around the areas as well.

Edit: oh, and vias too, don't forget those.

[GerryBags]

Seeing that the top one of the two connectors inside your red boundary brings in the power from A5, checking that thoroughly will be vital, thanks! i hadn't even twigged that that might be an issue but, of course, it is in the airflow path from those caps and has plenty of potential for electrolysis.

I'm thinking that power-down might be the point when the most damage is caused. My theory goes: While the 'scope is running the SMD caps heat up and start off-gassing their electrolyte, but the fan will be pulling air from the main board on the bottom and chucking it through the port at the top of A5 where the ribbon cable comes through. Once the power is turned off, the fan stops straight away, but the caps will still be off-gassing and the PSU filter caps will ensure there is still charge on some of the pins. I wonder, if the fan ran on for a few minutes after power off, whether this would have mitigated the issue? Obviously the caps were still going to fail at some point, but they might not have caused so much... collateral damage when they did.

[BravoV]

GerryBags, appreciate if you can share & post the close up photos of the damages if any, or any other new findings on that cap problem here, hopefully not.

So many victims of these damned caps.   

[GerryBags]

Bravo-V, here's a crop of one I took on first opening. I'll make sure to document the repair, as I'm bound to make plenty of mistakes which others can learn from - although I'm going to try my damnedest not to. I'll post the pics here.

[cheeseit]

I think more than those with arrows need attention, like the cap to the left of the upper tantalum. And possibly those rows of three vias in the upper left, some of them looks corroded.

[BravoV]

I think more than those with arrows need attention, like the cap to the left of the upper tantalum. And possibly those rows of three vias in the upper left, some of them looks corroded.

Agree, any nearby vias to the leaking caps also could act as pipe for leaking/sipping the liquid to the other side, and also the black tantalum, I would de-solder it just to inspect the trace under it, its way too close to that cap above it.

[GerryBags]

Thanks guys, I'm ratcheting up my sensitivity to the green-ness now, I'd rather exceed the affected area with my repair than have to revisit it.  I will also investigate the DAC adjust pot (R2010) very thoroughly. I've read blog posts from a couple of people who have had issues with that, maybe as it's raised off the board.

Will boiling flux keep the oxides out of the solder if those exposed pins with slight corrosion are reflowed, do you think?

[Miti]

This may be a redundant question, but would you guys change the Dallas NVRAM while you had the board off. I'm really hoping someone is going to say: "No, Gerry, you don't need to learn how to read and program cal data just yet." Anyone?

Yes, I did. I saved a copy years ago but just last month I changed it. I tried to use an FM16W08 from ebay but it didn't work for me so I ordered a DS1225AD. Programmed it with my TL866 and all is well.