TEKTRONIX 475 oscilloscope power problem

Discussion in 'Technical Repair' started by reptooyep, Sep 29, 2016.

  1. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    Hi everyone,
    I have a power failure on my TEK 475 oscilloscope.
    I worked all day along on this and i found the failure but i don't know wher does it comes from.
    I checked all the power supply :

    110v = ok
    50v = ok
    5v = ok
    +15v = 0v
    -15v = 0v
    -8v = +0.5v

    i noticed that the negatives power supplies's regulation comparators are powerd by the +15 supply. So, the failure comes from the +15.
    I have a shortcut between +15v and GND and i don't know where it can come from. I've taken off all the transistors one by one and all the visibles connectors but it's always short cut with GND.
    Exploring all the shematics did not help me to find a componment capable of short-cutting the GND.
    Does someone has a hint about this ?
     
  2. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    I found the solution on another forum for the missing supplies, it was a shorted tantallum cap.
    But now, i have no beam displayed .. Anothe failure.
    I'll go back to scheme.
    Anyone has an idea about how to start ?
     
  3. dl324

    Distinguished Member

    Mar 30, 2015
    3,242
    619
    I'd get the service manual and go through the power supply and CRT adjustment section.

    Do you have a high voltage probe and are you confident in your ability to work with high voltages?
     
  4. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
    The standard procedure for no display goes like this: (stop when/if you get something on the screen)
    Turn brightness to max, select external X input if available, otherwise use 1mS timebase.
    Check low voltage power supplies
    Check high voltage power supply - generally around 1000V - skip this if you don't have the equipment to do it safely.
    connect a voltmeter between the Y-plates on the CRT and adjust the Y-shift to get zero volts. If you can't do this the fault is in the Y circuit
    connect the voltmeter between the X-plates on the tube and adjust controls to get zero volts. If you can't do this the fault is in the X circuit
    CAREFULLY short circuit the grid and cathode of the CRT. If this gets something on screeen the fault is in the Z circuit.
    Still nothing on the screen may be the very high voltage or the CRT.
     
  5. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    Hi,
    I've unsoldered the horizontal and vertical plates and turn on : No bean !
    It means that there is no high voltage. I have nothing to test this.
    The HV fuse is OK.
    I'll go back to the schematic to find something
     
  6. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
    You probably won't see anything if the plates are disconnected.
    With the plates connected, measure the voltage between them and adjust X or Y shift as appropriate to get zero volts. If you can't get zero then that is your problem.
     
  7. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    I read somewhere that disconnecting the plates makes the plot going to the center of screen.
    In the HV part of the board, there is two HV bulbs that should light in HV presence. They don't.
    When i press the search plot button, nothing happens.
    I'll check for the horizontal and vertical tuning when i'll get a plot.
    What do you think ?
     
  8. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
  9. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    Thanks for your link. the schematic quality is very nice.
    I don't know how to tect the CRT without HT probe..
    Any idea ?
     
  10. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
    Do you an oscilloscope (one that works I mean)?
    If so check TP1318, collector of Q1318 for a nearly 30V P-P signal. If that is there then the HV generator is working but its gonna be difficult getting further than that.
    Some ideas: Look in the back of the CRT (CAUTION 2.5kV there so don't get your nose too close!) to see if you can see the heater glowing in there. Sometimes you can detect the very high voltage by putting your arm near the screen as you turn on the scope and the hairs on your arm detect the static.
     
  11. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    I've checked this test point. I can see a sine wave but it's about 2 volts.
    I tryed to change the transistors one after one without any success.
    All the supplies are OK.
    Maybe it comes from the Q1318 but i don't know where it is and if it's easy to change.
     
  12. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
    Is that 2V signal sitting on a DC level around 15V?
     
  13. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    I found the solution.
    Q1318 (BD19) was dead.
    I replaced it with a BD911 (TO220) which has about the same power and hfe.

    Now, it works but need few adjustements
    [​IMG]
    Does anyone knows how to enlarge the signal to fit all screen ?
    I did not find in the tech manual but english is not my language
    [​IMG]
     
  14. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
    R1155 on the horizontal amplifier board.
     
  15. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    I've touched to all the trimmers to see what happen and understand the tuning procedure but the r1155 is set to the maximum value.
    Could the problem comes from the CRT ?
    About 0.6 square is missing on each side.
    It could be from the orizontal plates signal too low ?
     
  16. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
    It could be because the high voltage is too high (reduces the sensitivity of the CRT and possibly implicated in the death of Q1318) and if it was the first suspect would be R1303A, 24.5MΩ, somewhere near Q1318.
    Do have a meter which can measure that resistor? I think you will need to unsolder one end of it to get a good measurement.
     
  17. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    Hi,
    I've tryed to measure it but my metter don't go higher than 20M. I can unsolder a side and add 200k to see what happend ?
    What i'm sure of is that it's much than 20M cause i can't measure it.
    How can this resistor effect the horizontal display ? I don't understand the relationship..
    You seem to know this scope very well. Thanks a lot for helping
     
  18. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
    I spent quite a few years repairing all sorts of 'scopes (and other instruments).
    That resistor is the sense resistor for the 2.5kV supply. It feeds back to the circuit that controls the HV oscillator, Q1318. That voltage also supplies a voltage tripler that feeds final accelerator in the CRT. The higher the voltage the faster the electrons travel through the CRT. The faster they are travelling, the harder they are to deflect, and so the 'picture' will get smaller.

    Adding 200k will further increase the resistance and the HV so don't do that. I am not sure what to advise as that resistor has to be able to withstand that 2.5kV and also we don't actually know that it is the problem. High value resistors do tend to age to higher values.

    If you can beg/borrow/steal a meter that can measure the resistor then do that.

    You could measure the voltage on Q1316 collector (or CR1316 cathode). The voltage here is marked on the schematic as +1.6V. If this is much lower than that it suggests that the HV is too high but it's far from conclusive.

    If there is space, you might make up a string of lower value resistors in series to get 24.5MΩ total but you MUST check the voltage ratings of the resistors that you use and compare with what the voltage across each resistor in the string would be and make sure there is a good safety margin. Also they should be in heatshrink or similar to insulate them from their surroundings.
     
  19. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,931
    382
    I just had an idea. If you can get a 20MΩ, preferably 1%, resistor, connect it in parallel with the 24.5MΩ one and the combination should read 11.0MΩ.
     
  20. reptooyep

    Thread Starter New Member

    Sep 29, 2016
    25
    0
    Hi,
    Today, i measured the collector of Q1318.
    I have 30vpp and 20v dc so, considering what you previously said, that seems normal. For the Q1316 collector, i have —3.2v dc and 0v ac.
     
Loading...