Connect with us

Home made valve oscilloscope. Version 2.0

Discussion in 'General Electronics Discussion' started by Miguel Lopez, Sep 10, 2014.

Scroll to continue with content
  1. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    Hello everybody

    A couple of years ago, I began to build an oscilloscope from scratch. I had a Soviet C1-94, but the CRT failed, as well as several parts of the circuit that I had to repair from time to time. Finally the C1-94 got out of service and i wasn't able to fix it, mainly due to the CRT dying.

    As by the time I was interested on vacuum electronics, and I had build my first valve amplifier, I thought that it would be a good idea to try to build my own valve-based oscilloscope. I used a cabinet from an old Soviet C1-99 scope that never worked. After weeks of study and trial and error test, I decided to used the circuit of the Telequipment S51 for the Y amplifier; and the circuit from the Telequipment Serviscope Minor for the X sweep generator.

    This decision was in part due to their simplicity and in part due to the availability of components. I used the point-to-point technic to solder the components. For the EHT, first I used a high frequency multiplier, but this induced a lot on noise on the trace, so I opted to use a mains transformer with a high voltage winding and a bridge rectifier.The CRT was a Soviet 17LOI2 (100x120mm)

    I built the scope after almost two years of work. Of course, I just worked on it from time to time, but still it was a great amount of working hours that I dedicated to this scope. It finally worked well, at least for my needs. (Up to 50kHz and up to 200V)

    Well, once that scope was finished, it worked OK for a couple of months (+/-). I was in the glory. Then some day, it smoked and smelled burnt varnish. I checked it out and I found the EHT transformer was burnt. As to wind this transformer was very hard and time consuming (about 22000 turns + 5500turns) I decided that it was not a good idea to wind it again and expect for it to get burnt again.....well, I don't like to be a Sisyphus.

    So the purpose of this thread is to share with my Odissey refurbishing my own oscilloscope. As now I have some background on vacuum electronics, this Version 2.0 of my scope, did not last too much, as it was the case with Version 1.0.

    I decided to use another variant and I thought on the voltage multiplier that I had used at the beginning. As this voltage multiplier injected noise to the scope signal (it used the same HT PSU as the whole circuit), this time I need a diferent PSU. To avoid even to use the same power transformer, I rectified the mains directly and supplied the heaters of an ECL82 directly from mains using a capacitor in series.

    The ECL82 triode was the oscillator and the pentode was the power output for the ferrite transformer. This circuit worked fine but it injected noise to the signal too. So, I thought that I had to obtain the EHT without oscillators, but without EHT transformer too. So I bet on for a 60Hz voltage multiplier.

    As building a 4kV multiplier from 60Hz required (for me) a big amount of components, I decided to use PCB to build these multipliers. Then, as at this point I had "betrayed" the Gods of point-to-point layout, I decided to betray at a bigger degree, so this time I used PCB for everything. Here some of the changes that I did:

    1.-) Generation of EHT is throught 60Hz voltage multiplier. Using 1N4007 diodes and 10uF/450V caps, supplied from a 120V winding in the mains transformer. +4kV, -1,5kV and -600V are now generated with these multipliers.
    2.-) PCB for everything.
    3.) All supplies are regulated (except those for EHT).
    4.) According to above mentioned:
    - HT regulated at 290V-
    - HT2 for input stage, regulated at 150V-
    - Heaters: regulated at 18,9V CD (three valves in series on each section of the circuit. Series supply to decrease the total current.
    - Heaters supply elevated about 35V- from -HT
    - CRT heater: Regulated at 6,3V-.
    - To further noise reduction, the X and Y circuits are shielded from each other, with a new steel made chassis.
    - Ranges sellector are now attached to a PCB with the required component, to avoid the use of wiring and so reducing possibility to induced noise.
    - Vacuum diodes from the sweep generator are now replaced by FR107 S-S diodes. (this is a betrayal to the Gods of Vacuum Electronics too, I'm a heretic)
    - The circuitry for CRT supply is also mounted on a PCB and all required potentiometers for adjustment of CRT characteristics are now directly wired to the PCB.

    In general, the circuits for X and Y amplifier are the same as in Version 1.0, but this time they are mounted on PCB. Now I will be sharing some pictures of the new version.
     
  2. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
    Sounds really cool, Miguel! I'm keen to see the photos!
     
  3. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    All rigth, let's go on

    The first problem to solve was induced noise. To solve this I built a steel chassis, with shielded compartment for the diferent sections. Y amplifier, X Sweep, and Transformer-rectifier-filter are in the upper part of the chassis, shielded from each other by a steel wall.
    PSUs and multipliers are bellow the chassis. This chassis is made using 1mm thick steel sheet. The parts are riveted. Here a picture of the chassis.
    [​IMG]
    http://www.imagebam.com/image/b8dd0d350659973
     
    Last edited: Sep 11, 2014
  4. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    Last edited: Sep 11, 2014
    davenn likes this.
  5. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    This is the basic circuit that I used for the HT PSU. I did some changes. Instead of LM317, I used LM7815, and instead of LM329, I used a 15V Zener diode with a 1N4007 in series, for temperature compensating.

    Resistors are adjusted for my purposes. The circuit is supplied with 340 to 355V DC and it delivers 290V DC regulated. Two MOSFETs are connected to the output. one for the Y amplifier, and the other for the sweep generator. A voltage divider with suitable resistors and another MOSFET at their junction, provides the 150V which are also required for the Y amplifier.


    [​IMG]
    http://www.imagebam.com/image/e3df6c350675235

    This circuit is from National Semiconductor Linear Brief 47 (1995)


    So, the Y amplifier, as said, is that of the Telequipment S51 oscilloscope, with minor changes. Here the circuit with the changes in red.

    [​IMG]
    http://www.imagebam.com/image/bf919a350660473
     
    Last edited: Sep 11, 2014
  6. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    The mains transformer is mounted at the last compartment of the chassis; while the other two are intended for the Y amplifier and X sweep generator. The transformer is a Soviet TCA-70 core with factory-made primary and several home-made secondaries.
    The three big capacitors seen behind the transformer are 220uF/500V and they are the filter at the rectifier output. 100 ohm resistors connect the capacitors to each other. Bridge rectifier is made with 1N4007.
    The black cables that you can see on the first and second compartment are for the Y and X shift potentiometers.


    [​IMG]
    http://www.imagebam.com/image/5adbf2350686851
     
  7. davenn

    davenn Moderator

    13,424
    1,803
    Sep 5, 2009
  8. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,269
    2,718
    Jan 21, 2010
    Miguel, I should show you some of the construction work I'm proud of. It almost makes me cry in shame after looking at yours.

    I'm interested in why you don't have a reverse biased diode across the 317, or indeed the entire high voltage precision regulator to protect the base-emitter junctions from a sudden loss of the high voltage input?

    I suspect the load may not be highly capacitive, but even your small output capacitance could cause damage to these junctions.

    I assume that it's for radiated noise reasons that you don't employ a switching pre-regulator.
     
  9. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    Hi all.

    Remember that I'm using a cabinet from a Soviet C1-99 oscilloscope, and that the mu-metal cone shielding the CRT is from a C1-83 Soviet oscilloscope. I did not build the cabinet. I just built the chassis, the PCB and the front and rear panel (we will get there)

    @Steve
    I didn't used the diode cause it is not on the original schematic. Now, I see that you're right and a reverse voltage could damage the regulator. I will include the diode. Thanks. I thought that the Zener would be enough for this, but I understand that this Zeners do not stand too much current. To include the diode is not a big deal, so I will.

    By the way, I found the original circuit among a lot of Application Notes sent to me by some guy who lives Mid way between Beijing and the Ronne Ice Shelf. If you see him someday, please thank him for me.

    I do not use a switching regulator by several reasons.
    - Noise surely is one of them, but not the most important.
    - Lack of parts
    - Lack of knowledge.
    - Simplicity.
     
  10. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
  11. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,269
    2,718
    Jan 21, 2010
    I'll pass the message along. I'm sure he'll be glad that some of that proved to be useful :)

    I guess #2 is the one that's the killer.

    For anyone that doesn't want to get into a political argument, please skip over the following paragraphs.

    But I guess it's all your fault. Cuba is obviously a far more dangerous country than Russia. Even now that the US has again stepped up sanctions on Russia, more trade is allowed between the US and Russia than the US and Cuba. (I hope the sarcasm translates)

    I found it interesting to contrast US sanctions with the official Australian Government briefing statement on Cuba.

    Incidentally, my wife mentioned last night that Cuba would be a nice place to visit. I wouldn't disagree.
     
  12. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    We (the Cubans) are friends of everybody who wants to be a friend.

    Well, the Version 1.0 of the scope was a mess regarding the CRT wiring. That is something I corrected this time, using a PCB. I placed all of the CRT circuit on a PCB located besides the mu-metal cone. I placed directly on it the brightness, focus and astigmatism potentiometers that were point-to-point wired on the previous version. The four potentiometer to control several aspects of the CRT: i.e: curvature, amplitude, etc are now placed on the same PCB and the wiring to the CRT socket is more neat.

    [​IMG]

    [​IMG]

    [​IMG]
     
  13. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    The next step is the X amplifier and sweep generator. As said before I used the Telequipment Serviscope Minor circuit with minor changes.

    [​IMG]

    The X sweep generator is to the left. Sorry, I took no pictures of the X section alone.
    [​IMG]
     
    Last edited: Sep 19, 2014
  14. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    The sweep generator is triggered by a flip-flop circuit which delivers a pulse synchronized with the input signal. Without signal (DC mode) the flip-flop oscillate at 70Hz (+/-) so starting the sweep on this condition too. The tube is an ECC81 twin triode.

    [​IMG]
    http://www.imagebam.com/image/051685352336183
     
  15. davenn

    davenn Moderator

    13,424
    1,803
    Sep 5, 2009
    Miguel

    am enjoying watching your project take shape :)

    Dave
     
  16. Fish4Fun

    Fish4Fun So long, and Thanks for all the Fish!

    450
    103
    Aug 27, 2013
    Miguel,

    WOW. I am in awe....What an absolutely AMAZING project! Living in the US, I and so many others are so incredibly spoiled by the plethora of virtually "free" hi-tech equipment that we generally take it for granted that building something like a tube-based analog oscilloscope would not be a good use of time....even though what you are building would rival anything available anywhere in the world when I was first learning about electronics.... to see the meticulous care and the prodigious effort you have invested in this project is a real "eye-opener" for me! And it warms my heart.

    Thank You! Thank you for sharing this magnificent and unique project! Thank You for reminding me that intelligence, integrity and craftsmanship are words used to describe people not products.

    I look very forward to watching your progress!

    Fish
     
  17. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    Thank you very much for your kind words. As somebody would say: "When need must".

    Now a picture of the scope with both the Y and X circuits on their respective places. Finally I removed one of the valves of the PCB cause X output doesn't need to be simmetric.

    [​IMG]
     
  18. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    With PSU, CRT, X sweep generator and Y amplifier on their respective places, there's nothing else to do but adjust both time and voltage ranges. On Version 1.0 I used point-to-point wiring and there was a lot of wires eager to get induced noise on them, so this time in order to reduce that mess, I used PCB too for the range selection.

    The better way I found, it was to attach the PCB directle to the selector pins. (see the picture). The Y ranges (voltage) requires just one PCB as I just need to change resistors for the voltage divider. It is not frequency compensated cause I do not intend this scope to be a standard, maybe for a Version 3.0...........

    The X range selector requires two PCBs as I need to change resistors and capacitors, so you can see in the picture the way I found to do this. Very simple, one PCB on each section of the selector, and the use of small caps to fit between the two PCBs.

    Step selectors with attached PCBs
    [​IMG]
    http://www.imagebam.com/image/e75df7350683733

    Step selectors mounted on the Scope front panel.
    [​IMG]
    http://www.imagebam.com/image/85e382350683734
     
    Last edited: Sep 23, 2014
  19. Miguel Lopez

    Miguel Lopez

    251
    62
    Jan 25, 2012
    Here is the front panel. I used the same as in Version 1.0. The markings are made of a photograph, designed in Photoshop and then covered by a transparent plastic piece.
    Ranges in voltage are [V/div]:
    50m, 0.1 0.2 0.5 1.0 2.0 5.0 10 20 50 100 and 200; plus two adittional fixed ranges of 300 and 500V/div. I intend to use this scope to work with vacuum circuits, so I need to measure high voltage.

    Ranges in time are
    [mS/div] 5 2 1 0,5 0,2 0,1
    [uS/div] 50 20 10 5 2 1

    The screen is covered by a blue piece from the original C1-99 scope.

    Sine wave
    [​IMG]


    TTL pulses
    [​IMG]
     
    Ian likes this.
  20. HellasTechn

    HellasTechn

    1,472
    210
    Apr 14, 2013
    That is what i call a person with knowledge on electronics !!! well done. i wish i even had half the knowledge you have Miguel.
     
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

-