Sansui Output transfomers from a SM-30

Discussion in 'Tube Audio' started by battradio, Oct 7, 2017.

  1. primosounds

    primosounds SE KT120 w/ 6J5G drivers. LCLC ps + 2 x 5ar4 Subscriber

    Messages:
    2,574
    Location:
    Terra, 21st century CE
    I wonder why the extra gain stage for this amp? It seems like this circuit would probably work with harder to drive or low mu tubes.
     
    Last edited: Oct 11, 2017

     

    Please register to disable this ad.

  2. kward

    kward AK Subscriber Subscriber

    Messages:
    3,351
    Location:
    Utah
    It looks like a Williamson topology. A clue may be the output stage--it's biased at 35 volts. That's in the ball park of where you would bias a 6L6 or KT66 output stage.
     
    battradio likes this.
  3. battradio

    battradio Electron trainer

    Messages:
    5,211
    Location:
    Near ST. Louis MO
    The HF-V60 is one i'm leaning to and will build it with some pioneer iron i acquired from Tinkerbelle to see ( hear) how it sounds with 6v6's .

    Also will make some adapter so 6BQ5's can be used in the same amp and maye 12AU7/6FQ7 adapters also

    Got the SM-30 transfomers in last evening will get some photos and measurements on it later tonight . The outputs are 32 , 16 and 8 ohms , the primaries are split and measure , 203 and 190 each .

    Whatever i do with them they will make a very nice looking amp and sounding amp .
     
    Last edited: Oct 12, 2017
  4. primosounds

    primosounds SE KT120 w/ 6J5G drivers. LCLC ps + 2 x 5ar4 Subscriber

    Messages:
    2,574
    Location:
    Terra, 21st century CE
    Are those transformers potted or just covered to look like they are potted.
     
  5. battradio

    battradio Electron trainer

    Messages:
    5,211
    Location:
    Near ST. Louis MO
    They are potted , i will be potting one pair of the pioneer iron in peerless cans i salvage that i will be using with the HF-v60 circuit

    [​IMG][​IMG][​IMG]
    [​IMG]

    Going to paint the peerless cans and find a pioneer decal to put on them so someone in the future doesn't mistake the for a real peerless transfomer .
     
  6. BinaryMike

    BinaryMike Pelagic EE Subscriber

    High-gain triode input stages place a relatively large capacitive load on the preamp due to Miller effect. I would use a cascoded triode or pentode input stage unless I was very sure that only low-output-impedance preamps would be used. The 6SN7 input stage is a reasonable compromise because of its low gain.
     
    battradio likes this.

     

    Please register to disable this ad.

  7. battradio

    battradio Electron trainer

    Messages:
    5,211
    Location:
    Near ST. Louis MO
    The next line stage i will be building will have a pair LS 151 as output transformers , a 16 K ohm to 500 ohm transfomers , i have 500 ohm input transfomers on a couple of my amps .

    My Grant DAC -11 has a 6DJ8 in it as the output buffer and my other preamp is a copy of a Marantz 7 with a cathode follower
     
    Last edited: Oct 11, 2017
    Pio1980 likes this.
  8. battradio

    battradio Electron trainer

    Messages:
    5,211
    Location:
    Near ST. Louis MO
    double post
     
    Last edited: Oct 15, 2017
  9. battradio

    battradio Electron trainer

    Messages:
    5,211
    Location:
    Near ST. Louis MO
    i measured the inductance of the primary to plate to plate at 100 hz and the inductance is about 34.5 henry for each transfomer . The pairs two pairs of of pioneers AT-6123 6BQ5 output transfomer have 42 henry plate to plate
    Will measure the transfomers in my Scott 222C next .
     
  10. AmadeusMozart

    AmadeusMozart Active Member

    Messages:
    106
    Audio Note P2-PP 20 Watt 6L6.gif Audio Note 4.gif audio_note - P1 PP question mark.gif
    I'm building a clone of the Audio Note kit 4 - picked up the PCB on ePray and did surgery on that. Mains transformers and output transformers are oversized and I will be able to change the end stage from the 6V6 to the 6L6 but output impedance will change and I may have to adjust the feedback loop. I was always hesistant to make changes in a feedback loop until I read a handy trick: get some tuning capacitor, feed the amp with a square wave and observe the output. Then adjust with the tuning capacitor till you get the best square wave at the output and subsequently measure the capacitance of the tuning capacitor and replace with fixed values. The other thing to check is the overall phase shift and that can be done with a dual trace oscilloscope - even a cheap USB one will do that.

    Am attaching the circuit diagram of the Audio Note kit 4 and of the Audio Note P2-PP (third version - in it the 6SN7 was replaced with the ECC83/12AX7) which has the 6L6 and is 20 Watts. I also upload another schematic which I think is for the Audio Note P1-PP but am not sure although the schematic makes it quite believable. (give away are the floating paraphase resistor values and the general configuration of the finals. It is interesting to see that they have brought back the 47K trimpot in the floating paraphase inverter.) The 6V6 has about 10 dB feedback, the gain is not excessive. Sensitivity is approx 250mV for full output which may be usefull when driving from an iPod etc.
     
    Last edited: Oct 17, 2017
  11. BinaryMike

    BinaryMike Pelagic EE Subscriber

    You will find that radio tuning capacitors don't have enough range in many cases, depending on feedback divider resistance values, so you might want to add a capacitance decade box or substitution box to your shopping list. Also, be aware that other elements of the amplifier may need tuning for acceptable squarewave fidelity --- particularly voltage amplifier bandwidth, rolloff shape, and output transformer damping. And expect interaction. This whole area tends to be about as sticky as tube amplifier engineering gets.
     

     

    Please register to disable this ad.

  12. battradio

    battradio Electron trainer

    Messages:
    5,211
    Location:
    Near ST. Louis MO
    I would qualify that statement with some variable capacitors don't have the needed range . I have silver plated variable capacitors over 1600 PF that i use to make crystal radios , they are four section and 430 pf per section so the can be wired in series or parallel to get the capacitance you need . Most caps in AA-5 radios have a 365 PF tuning cap and a 100 to 180 pf section for the oscillator which would give about 465 pf to tune mosts amps .


    http://peeblesoriginals.com/catalog/15.php
     
  13. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    25,650
    Location:
    Angel Station, Alabama
    Compression mica trimmer varicaps with the required range will do.
     
  14. AmadeusMozart

    AmadeusMozart Active Member

    Messages:
    106
    DSC00912.JPG DSC00809 small.jpg DSC00861 small.jpg DSC00864 small.jpg

    Amplifier is up and running and I am happy. Have made some changes, main ones are: 12AX7 on 6.3 VAC for the filaments (rather than DC) and have given the 6V6 slightly more current: 510 Ohm resistors rather than 560 ohm. I have cut some PCB tracks and placed the 6V6 grid stoppers as close as possible to the sockets using 3/4W SMD resistors. Hum is approx 78dB down and noise in excess of 80dB and output is 9.8W RMS. Peak output is approx 12W. Attached some construction pictures - top plate is not yet screwed down due to some changes I want to test (and am waiting for parts from China to arrive to do so). (Having a top plate makes it easy to adjust and test. Due to arthritis the chassis was made by someone else.)

    I had a heck of a time finding some decent octal tube sockets that satisfied my quality requirements. In the end I found some NOS Cinch Octal and Noval sockets that are just perfect.

    The amplifier in the background is a SE 6BQ5 that I made about four years ago.
     
    Last edited: Nov 20, 2017
  15. AmadeusMozart

    AmadeusMozart Active Member

    Messages:
    106
    If I had found the Sansui schematics before I had bought the PCB and had ordered the chassis then I would have build the Sansui one. I am still tempted to order another top plate with holes cut for eight octal sockets and build the HF-V60: octal tubes normally have a longer life expectancy than noval (and am not overly fond of the 12AX7 - the 6SN7 is very linear and according to some about the best driver to be found).
    AM
     
    Last edited: Nov 20, 2017
  16. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    25,650
    Location:
    Angel Station, Alabama
    You'd use a 6SL7 or 12SL7, or a 6SC7. A 6SN7 is like a 12AU7.
     

     

    Please register to disable this ad.

  17. AmadeusMozart

    AmadeusMozart Active Member

    Messages:
    106
    Thanks for the reply, I prefer to stick close to orginal designs unless there is a compelling reason to make changes. The early Sansui designs have quite a good reputation, quality on later Sansui gear leaves lots to be desired.
     
  18. Pio1980

    Pio1980 AK Member Subscriber

    Messages:
    25,650
    Location:
    Angel Station, Alabama
    I have an SM-80 and early 1000 that somewhat resembles it that both use the elegant Laurent /Dynaco 6AN8 driver circuit into P/P 25E5 beam tubes.
    Since they aren't in regular use here, I cant say much other than they have the virtues of this simple driver ckt combined with Sansui output transformers.
     
  19. primosounds

    primosounds SE KT120 w/ 6J5G drivers. LCLC ps + 2 x 5ar4 Subscriber

    Messages:
    2,574
    Location:
    Terra, 21st century CE
    I copied an Andio Note PP EL84 amp circuit. It had a 6BL8 for input/splitter duty. I am not sure if that is the one you will be using but the circuit that i used had too much gain. It needed tons of -FB to tame it , and frankly i just did not like listening to negative FB. I scrapped that circuit and switched to the Scott 299 circuit which was more in tune with my needs.
     
    Pio1980 likes this.
  20. Shadowdog

    Shadowdog Super Member

    Messages:
    2,752
    Location:
    Flin Flon, Manitoba
    What about the "Musical Machine" 6V6 schem. that could be used with the easier to drive EL84/6bq5 and the latest version had CCS It ran the tubes in Triode mode and had no GNFB.

    It looked like a very simple differential feedback circuit and everyone that built one loved it! You however needed according to the designer 10-11K primary output transformers for the best

    sonics (with 6V6) and you only got 6W/CH. You could parallel tubes for more power and a fellow did that on AA using Dynaco Z-565 trans . He changed the driver/splitter to 6CG7 spoke highly of it.
     
    Last edited: Nov 20, 2017

Share This Page