• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Photo Gallery

Sansui AU-70 clone
1714836646930.png
 
  • Like
Reactions: 9 users
A rebuild of https://www.diyaudio.com/community/threads/photo-gallery.71300/page-547#post-6969646
since that was 'destroyed' during shipping so someone could have a listen. The transformers used to be tie wrapped to a wooden plank that was glued. The weight of the transformers caused them to rip loose during transport Long story short. I used bolts for the ot since my brother is going to take it overseas in his suitcase. Also decided to use a tube rectifier and all but the first cap are dc link or pp.
PXL_20240504_161318409-EDIT.jpg

One bolt missing...so that's a little to-do.
PXL_20240504_155015759-EDIT.jpg
PXL_20240504_161344740.jpg
 
Last edited:
  • Like
Reactions: 7 users
Tricky and laborious to build, but the sound quality is very, very rewarding. AU-70 was an outstanding amplifier.

1715027344069.png

  1. 12AX7 preamplifier
  2. VU meter driver
  3. 12.6VDC: heating preamplifier (1) tubes and also DC power supply for VU meter (2) and input selector (7)
  4. 6AN8 driver/phase inverter
  5. 7189 power amplifier
  6. power supply
  7. input selector
 
  • Like
Reactions: 6 users
Forgive me Tim deP, but I redesigned your amp TVA10 to Ultralinear....EL34, with 12au7 srpp input, 12at7 ltp phase slitter and 12au7 cathode followers, this amp did not use global negative feedback...
 

Attachments

  • IMG_20240516_090110 (1).jpg
    IMG_20240516_090110 (1).jpg
    537.7 KB · Views: 82
  • IMG_20240516_090110.jpg
    IMG_20240516_090110.jpg
    537.7 KB · Views: 77
  • IMG_20240516_090124.jpg
    IMG_20240516_090124.jpg
    465.6 KB · Views: 77
  • IMG_20240516_090132.jpg
    IMG_20240516_090132.jpg
    484.1 KB · Views: 81
  • VID_20240515_165959.mp4
    18.5 MB
  • Like
Reactions: 1 users
After 2 years of prototyping, this amp is nearing completion. It is a pure class A self splitting push-pull amp, with a esp32 based regulator that ensures both tubes are biased exactly equal (less than 40uA). Tubematching is optional.
Driver is 6E5P but minor redesign to 6E6P is something I want to do, because of the pinout and the possibility to try different driver tubes. (6E6P, D3A, E180F and others all have more or less the same pinout). Power tubes are EL84, NOS. Power output is a measured 8W RMS just before clipping. Distortion figures at 1W are about 0,3% mostly 2nd and 3th. No global feedback is applied.
IMG_4164.jpg

7025198C-3E45-4FB5-BD75-E4E33798F95C.jpg

6AF6FBA6-E3E4-4B55-9ED3-93AF70C185BC.jpg

IMG_4173.jpg


I designed it to be an alternative for my 300B SE amplifier, because the 300B tube are becoming too expensive, and they lasted only 2-3 years in my setup. The amplifier uses silicon everywhere (CCS, regulators, Vref, DAC,ADC ...), so my initial name for the project was "Apostate". Because of the CCS's used the powersupply is almost completely isolated from the signal, and there is only one capacitor per channel in the signalpath. (I used a russian Teflon capacitor)
The amplifier sounds amazing, never had a more stable 3D stereo image.
After living with the proof of concept prototype till january this year I designed the new enclosure to match my Thorens TD160 turntable, that I build in 2023.
IMG_4163.jpg


IMG_4236.jpg

IMG_3619.jpg
 
Last edited:
  • Like
Reactions: 9 users
Beautiful work.

Self inverting?
Is there a CCS in the cathodes of the EL84 pair?

Post a schematic if you care to share.
Thanks!
Yes, self inverting, and there is a CCS in the cathodes.
Schematics: (from old eagle version I still use)
The latest version of the amp is slightly different, but I didnt had to change the PCB's for that. The hardest thing was to get a rock-stable and noisefree -2,5v as reference for the bias controller; the regulating voltage is directly applied to the grid of the second el84, and should be able to range from -2,5v to +2,5v. The voltage splitter in the -18v line is now a 1k2 resistor and a TL431 instead of the 1k2 /240R. Vref is now very stable -2,49V, but noisy (I short that noise to ground with a capacitor.) The opamps in the bias controller are for detecting audio signal, if there is audio signal, regulating is postponed till there is no audio signal. (between songs) The 1K balancing resistors on the Bias controller and the "realground" connection are removed from the PCB. Oh and dont mind the 1n5400 diodes everywhere, i took them only for their shape/size. In reality it is a mix of shottky's, and high voltage ultrafast diodes.
amp.JPG

Hv supply.JPG


bias control supply.JPG
bias control.JPG