Connect with us

My brand new oscilloscope

Discussion in 'Troubleshooting and Repair' started by (*steve*), Nov 20, 2018.

  1. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Ok, so clear heatshrinking kinda makes it less necessary to mark the anode end.

    1545887180778703.jpg
     
    bushtech likes this.
  2. bushtech

    bushtech

    885
    139
    Sep 13, 2016
    But I like the the black and red ends. Much less likely to have a blonde moment when installing.
     
  3. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Installed! (I'm a blonde, but it's ok. I always have a headset on playing "breathe in... breathe out...")

    154589082485478.jpg

    I have a couple of bulbs to replace, then I'll double check everything I've done and power it up!!
     
  4. bushtech

    bushtech

    885
    139
    Sep 13, 2016
    lol. Here's hoping that your magic smoke retention powers work
     
    (*steve*) likes this.
  5. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Always one more step...

    Here is a failed graticule illumination lamp.

    1546049247012937.jpg

    I'm planning to replace it with a LED.

    Currently there is a 100Ω pot across a 13.5VAC winding with the wiper connected to two small 14V globes.

    I'm planning to half wave rectify the AC from the pot, filter it and use that to drive a pair of LEDs, each having a 680Ω resistor in series with them.

    Another globe can be replaced without disassembling the front panel, so I won't do it now.

    1546049845349561.jpg
     
  6. 73's de Edd

    73's de Edd

    2,644
    1,074
    Aug 21, 2015
    Sir Steve . . . . .

    And just to think, that back in the late 60- early 70's one of my engineering crew was desperate for rent money coming up and was trying to sell both his Hoolet Plastered, state of the art of miniaturization DVM and his Telequipment scope.
    His was being the dual channel D54 version of your S54.

    Thank goodness that I passed on them . . . . with the HP being a nightmare on its custom chip and on ceramic board construction.

    The Telequipment . . .you are finding all out about now.
    Seems like back in that timeframe Telequipment was being submarketed by TEK as a cheap alternative.
    With the actual commonality, being that they both used RED and BLACK control knobs.

    On your diode string of two end 1N4007's and resistor . . . . . was the shorter case diode, just to adjust overall length optimization ?

    In my buildup of series string components like that . . .I use the smallest size of modellers hollow brass tubing .
    Flux tin it and insert the tinned wire leads within the tubing . . .reflow solder . . .then you have a straight linear joining.

    HP Breakthrough . . . at that time . . . .
    [​IMG]

    ADD-ON . . . .

    WOW . . . . is that what looks to be a #46 screw in bulb . .6.3V @ 300 ma . .. that is having its TIT soldered to a wire. Then . . . is it dependent upon screwing into that open coil spring "threading" for about 5 turns.
    With that then leaving those twists on the wire length attached to the bulb ?


    73's de Edd . . . . .

    Now I wonder . ? . ? . ? . . . . if I made some wine out of raisins . . . would I have to wait for it to age ?
     
    Last edited: Dec 29, 2018
    Richard9025 likes this.
  7. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Those are GP02-40-E3/54's. They're somewhat higher rated than 1N4007's :)

    Those replacement rectifiers are probably OK up to 8kV -- The originals were rated at 3.5kV. I used 2 diodes because:
    1. Selenium rectifiers are more tolerant of transient high reverse voltage than silicon diodes
    2. The replacements came in a pack of 5, and I only needed 2.
    3. They physically fit.
    4. They're cheap.
    5. I don't want to have to fix this again :)
    The other diode is a zener to mimic the higher forward drop of the selenium rectifiers. And obviously the resistor because selenium resistors have a higher dynamic resistance too.

    In the parts list it is shown as a 14V bulb (with an internal part number that doesn't mean much to me (150-0095-00).

    upload_2018-12-29_11-9-58.png

    And damn if they're not available on ebay!

    Unfortunately they are variously listed (on ebay and Tektronix "spares" sites) as 12V or 5V... Bu the schematic looks like this:

    upload_2018-12-29_11-12-59.png

    I've decided to replace them with orange LEDs rather than white. They are used to edge-light an acrylic sheet with the graticule engraved and coloured orange. At low voltages these bulbs would be more orange than white anyway.

    The springs are free to move and simply press into the back of a "cup". You can undo the bulb by holding the spring, and twisting the bulb a little before allowing the bulb to rotate (along with the spring) back to it's original orientation, then rinse and repeat until you can easily turn the spring. As photographed, the bulb was almost pulled out to the full length of the wire!

    Oh, and that other bulb is a neon bulb. It seems to be used as both a power on indicator and a voltage regulator.

    upload_2018-12-29_11-27-4.png
     

    Attached Files:

  8. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    There's not a lot of room to replace those lamps with LEDs, so I might make a strip of surface mount LEDs... later.

    Another issue is that after reading the in-circuit values of many of these carbon composition resistors, about half read higher than the rated value. The worst I found was a 330K resistor reading over 1M. I really don't want to replace all of those resistors! And... I'd have to remove the board.

    Oh well. If I have to, then I have to. I may be able to get away with just removing the CRT.

    Should I power it up as it is now? I'm leaning toward yes. It's been on before, so I'm unlikely to do any more damage.

    Also, it's clear that someone thought that cleaning the pots was the cure for everything. The inside is almost completely covered in oil (WD50?). I might clean the board with IPA before I turn it on again. I'm not sure it's a real problem, but it's certainly annoying.
     
  9. 73's de Edd

    73's de Edd

    2,644
    1,074
    Aug 21, 2015
    WD-50 . . . . . IS . . . . . being the excess use of WD-40. . . . .however, with being in that state now, it certainly is being a nice / effective fly repellent !

    In the parts list it is shown as a 14V bulb
    and with that 0.75W spec, with its 53/4ma current consumption . . . and if HAVING to use an incandescent lamp . . that would be your power saving choice of lamp.
     
    Last edited: Dec 29, 2018
    Richard9025 likes this.
  10. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    I'll go with that being my meaning :)

    The front cover us back on...

    1546074192297740.jpg

    Did I mention that a previous owner has disconnected the earth?

    15460744886115.jpg

    At least they made it obvious.

    I know what this means, and I'll be careful :)
     
  11. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Partial success!

    1546075234310799.jpg

    The vertical position doesn't change with the vertical gain any more.

    The width of the trace doesn't vary with time.

    But the "brilliance" is fixed, and I can't get a better focus than shown (but I can do a lot worse).

    Also the cal signal is 50Hz, which is line frequency. And the amplitude is low.

    Here is a signal from something I have handy...

    1546076762517772.jpg

    The time is off by about 20% but the vertical gain looks close -- hard to tell from this waveform since my meter reads RMS values. The signal is 0.55V RMS, and that corresponds to about 1.5V probably, and the vertical amplifier is set to 0.1V/div using a 10x probe.

    Surprisingly there's enough adjustment range in the "set speed" control to bring the time base closer to what it should be.
     
  12. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    For anyone repairing a scope, this document might come in handy.

    It's from Tektronix, but it's somewhat agnoistic to the brand or type of scope you have. Other than some particular Tektronix quirks, much of it applies to any analog scope.
     
    bushtech likes this.
  13. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    The first step is to check the power supply. If it's doing something wrong then lots of stuff can be affected.

    Lots of high voltages, and my meter read up to 1000VAC and 500VDC :-(

    Not to worry, not everything needs to be measured with respect to ground. But placing the meter's ground lead into the -1000V rail does make one somewhat cautious...

    Here's the results:

    154613323443154.jpg

    The unit is set for 250VAC and my Variac is set for 250VAC. The AC voltages in the secondary read a little low but not enough for me to worry about.

    All the rails I can measure look pretty good. I measured the voltages across C409, C411, and C412 individually. With the meter's ground lead at point 41 I could measure C409, then I moved the ground lead to between C411 & C412 to measure them. Unfortunately (but maybe fortuitously) I left the ground lead in this position while measuring the grid and focus voltages.

    The first odd thing I discovered was that the voltage at point 52 (grid voltage) did not vary. That's not too surprising because the "brilliance" control is totally ineffective. This was also the point I noted the ground lead was still between the two capacitors. A quick calculation showed the voltage at point 52 was really close to the -1000V rail, far closer than I would expect (and that was odd).

    On moving the ground lead to the -1000V rail, I discovered that the voltage across D301 was 0V! I also noted that there was a lot more muck around this pot than the others. (Perhaps this was a WD60 amount!!! :-D). It seems that someone had tried LOTS of cleaning of this pot when the problem looks to be a shorted zener diode (or maybe a shorted cap -- less likely).

    I'm pretty sure I have some high voltage zeners, but the placement of this component is going to make replacement really hard. Perhaps this goes on top of the board like the caps too.

    And I just happen to have some 100V 3W zeners. The original is a 330mW. I may be about to add some safety margin :)

    Oh, and a good thing about this scope is that the caps discharge by themselves in about a minute.

    Another thing of happened while I was checking voltages. At one point the trace suddenly shrunk to about half the width of the display. I may have a dry joint to find, it fixed itself up again a few minutes later.
     
    Last edited: Dec 30, 2018
  14. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    I knew these would come in handy one day.

    1546135609661626.jpg

    What? me? a magpie?
     
  15. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Woo hoo!

    1546136566208926.jpg

    And now the focus works correctly too.

    Need to fix the cal output next.
     
    bushtech likes this.
  16. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Doh!

    upload_2018-12-30_19-29-11.png

    I guess I don't.

    Do you know how long I spent looking for the calibration oscillator in the circuit diagram?

    Maybe I should run through the calibration procedure and see if everything can be brought into spec.

    Or maybe I should try to find what caused the horizontal width to reduce. Prod around the board with a stick?
     
  17. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Here is the cal signal.

    1546171835223704.jpg

    The trigger seems to be a bit "dicky".

    And the cal signal isn't 500mV, it's closer to 50mV

    1546172351052553.jpg
     
  18. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    Maybe 50mV is what it's supposed to be:

    upload_2018-12-30_20-33-18.png
     
  19. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,174
    2,689
    Jan 21, 2010
    One of the buttons on this scope self destructed.

    IMG_20190526_201101_compress4.jpg

    Above is the exploded button along with a good button and a 3D printed replacement.

    IMG_20190526_201139_compress27.jpg

    As soon as we have some black filament I'll print them in black, but the red is growing on me.

    IMG_20190526_201144_compress36.jpg

    They're a bit loose at the moment, so I need to reprint them with a slightly smaller hole., But I'm impressed for my first 3D printed thing.
     
  20. 73's de Edd

    73's de Edd

    2,644
    1,074
    Aug 21, 2015
    Sir Steve . . . . .

    I was just observing the highly erose surface of the printed knob and since you were going to print again, I was going to suggest that you make it larger in diameter in orderr to be able to rotary WET and dry sand it to a smoother surface.
    However due to either a choice of too large of a diameter /gauge of plastic filament OR inadequate heat level for attaining a FULL,fusing and flow together, I don't think that would be fully successfull.
    Probably, being with a result much akin to trying to sand a honeycomb or wasp nest smooth.
    If being my puppy to recreate, I would be using clear resin and hardener casting liquid as is used in fibreglass work on boats and Corvettes bodywork.
    Black and red tinting or powder added and a mold of an original knob being made of wax or silicone rubber molding liquid.

    Looking at the fracture manner and pattern I was expecting either bakelite or styrene family of the original knob, but that now looks like the styrene family of materiel being used on the original knobs, 10 years further back and the bakelite was more likely to have been used.

    73's de Edd . . . . .


    The difference between art and science is that if some absolute piece of shit WORKS as art, you DON'T have to explain why, BUT, that's DEFINITELY not being so in science.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-