Shiny Eyes by Radu C. Tarta
This is a fantastic 4P1L push pull incarnation done by Radu. I hope you enjoy reading this post as much as I did. Cheers, Ale
Introduction
“Shiny Eyes” Push-Pull tube amplifier is the result of 6 months of research, experimentation and lots of music listening. It started in the summer of 2012 when life opportunity re-opened the door of an old passion: electron tubes. However, this time my goals were directed towards audio. I understood from the beginning that after 30 years of interruption I have to dedicate a great amount of time and energy to catch up with technology and ideas. I want to thank my wife Paula for her emotional support and understanding. I am also grateful to the fanatics from the DIYaudio community for their contribution to the world of tube audio experimentation. Personal thanks to Kevin Carter of K&K Audio for his advises and help, to Ale Moglia of Bartola Valves for his inspiring work and to Rod Coleman for his excellent filament regulators.
Initial tests
Here are few images taken during design, experimentation and final building: Early stages…
Final build and design notes
Design goals:
- pleasure our senses
- not trying to measure perfect, but make it sound warm and transparent
- zero global feedback
- 2 stages design, 4П1Л DHT triode strapped for final valves
- minimum number of capacitors in the signal path
- quality components in a reasonable budget
Schematic description:
- played with different final stage tubes(6P3, KT77, 6C41, 4П1Л) and driver stage tubes(6S3P, 6C45P, 6H6,6E5P,6H30P) and loved the sound of 4П1Л and 6H30П-EB.
- 4П1Л filaments powered from a separate dual bobbin transformer, 11V raw supply and Rod Coleman’s regulators ( one for each channel)
- 6H30П-EB filament DC powered
- lots of Lundahl iron: LL1649 main, LL1638 Choke in special configuration for improved common mode rejection, LL1663 as OT, LL1660/10mA Alt “V” as IT
- bridge rectifier using UF5408 ultrafast rectifier and filtering capacitors are Clarity Caps TC series 110uF
- Delay HT 60 seconds based on 555 IC
- Final stage parallel 4П1Л (4X 4П1Л) filament biased, operating point at 245 V, 35mA, -21V ( tubes were selected to match as close as possible
- Driver stage 6H30П-EB balanced and cryogenic treated(tried also the standard version but liked the sound of the cryo better(more depth)), cathode biased, operating point at 120V, 25mA, -5.5V
- Elna Cerafine for cathode caps and Takman for resistors in the signal path except the final stage cathode (filament) resistor which is wire wound chassis mount
- Wires in the signal path are Teflon silver plated ( just had them and try to give it a shot, but I didn’t hear any noticeable difference in sound. However, I like to work with Teflon isolation)
- ground wiring is a combination of star and ground follow signal
THD measurements
- Gain 17dB (7.3X)
- Maximum power before clipping: 7W
- When 1Vrms input signal, 6.7W power output @ 0.2% THD
N.B. Measurements were done repeatedly under the same conditions, however high accuracy and precision was limited by measuring tools used. To measure THD I used ARTA software and USB external sound card EMU0204 and a attenuator interface( THD of sound card and attenuator at -17dB position was 0.025%) to protect the sound card. To measure and visualize output signal a Rigol DS1102E digital oscilloscope was used.
Listening impressions
- vocal sounds amazing, transparent and warm
- instrumental sounds well defined, full of depth and great separation
- sound is amazing for jazz, classical and vocal
My setup is: Marantz CD5001 player and Klipsch Synergy B3 speakers
Final Note
This amplifier will come soon at www.simplepleasuretubeamp.com:
I am working on a documentation for my amp, and I am hoping on performing more accurate measurements. However, this amp’s sound can better tell the truth.
Hope that you enjoy seeing a finalized amp with the amazingly linear Russian 4П1Л.
Best regards,
Radu Tarta
Congratulations Rada.
One thing that caught my eye in your measurements is the residual harmonic at 180Hz (3rd harmonic of 60Hz supply). If this is not an artifact coming from your sound card it might increase intermodulation when you use some power.
45,
I guess it is an artifact, but not from the sound card. Sound card locks clean and noise is down bellow 115 dB. The attenuator is increasing distortion and captures main power noise. I also tried to use only the sound card at low levels (up to 1V) and just by connecting the signal ground from the sound card to the ground of the amplifier it will show the same picture. When I use Dr. Jordan’s software it will just show noise at 50-400 Hz. I wish I have other options for measurements. I don’t know anybody around here who can help me with more professional measurements and listening. Thanks for noticing.
Best,
Radu
Hi Radu,
Not sure if I understood your testing procedure here. If you are concerned about your soundcard interface you should try building the Pete Millett’s interface. It’s fantastic and will convert your PC in a powerful lab instrument, without the need of any expensive equipment.
I will look at the FFT with:
1) amplifier input grounded
2) soundcard input grounded
3) amplifier with earth lifted
4) amplifier with low output signal (no attenuator)
You may find from the above where the H3 from mains is coming from. To me it looks like the mains hum is amplified by your amp and therefore you are seeing the H3 product from 60Hz…
Hope this helps.
Ale
Hi Ale,
Actually I am planning to built Pete Millett’s interface, thanks for recommending it.
Thanks for guidance. I already did #2 and looks clean. I’ll get back when I’ll figure out what’s going on.
Best,
Radu
Hi Ale,
I did some measurements again. My conclusion is: The presence of 60Hz and odd harmonics mainly 3rd at 180Hz are not caused by amplified mains hum by my amp. They are present even with the amp disconnected and unplugged, and I believe they are picked up somehow from my workbench.
If you look at Pete Millett’s interface web page, he is showing in his FFTs the same issue.
However, a strong conclusion cannot be reach as the results are ambiguous. Just bought PPete Millett’s kit from ebay and I am looking forward to see if it is any different with his interface. Hope for best,
Radu