diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Analogue Source (http://www.diyaudio.com/forums/analogue-source/)
-   -   JGs Nobrainer and Nobrainer Discrete (http://www.diyaudio.com/forums/analogue-source/195017-jg-s-nobrainer-nobrainer-discrete.html)

Joachim Gerhard 20th August 2011 08:52 PM

JGs Nobrainer and Nobrainer Discrete
 
1 Attachment(s)
I open this thread to discuss my Nobrainer circuit only. In my MPP thread i have tried out all kind of approaches and the question that comes more often then others " What is the best circuit ? ". I have tried to answer that question many times so i will only repeat that it depends a lot on the implementation. The same ( good ) topology can sound very different depending on active and passive part selection, building method ( 3D, Perfboard, PCB ), grounding and of cause PSU. Some say that they can even hear contacts, solder, mechanical coupling or decoupling, power cords, interconnects and other exotic accessory like Bybees and the like. So the MPP thread is very complete and detailed but also kind of confusing because of the many roads i have taken. This thread has has another philosophy. I will try to focus on only one circuit the Nobrainer.
As far as i can tell it is the most popular phono stage i designed. On the FF2011 alone where 3 versions of it, Georgs on a PCB designed by Holger Barske,
another guy had made a 3D version and Holger had my original "Igel" ( porcupine ) in his system. i think the reasons are mostly simplicity and low cost but the circuit sounds very well too. It also performs well technically.
It is dead silent although i use 0.04 input transistors, it is a transimpedance circuit with a high PRAT factor ( Power, Rhythm, Attack, Timing ) that a transcondactance stage hardly can match ( its possible but hard to do ) and it solves the major problem in a common base stage, the DC offset into the cartridge. Some people say that a transconducting ( or summing ) stage over
damps the cartridge including JC. That can happen when the cartridge impedance is rather high like in a high output MC. Shortening the inductance can then damp the treble. I have not made that experience, besides the input impedance of the Nobrainer can be adjusted over a certain range. Normally i set it at around 3.5mA and with that value it gave good results with my Lyra carts ( ca. 6 Ohm ), with the EMTs and Denon 103s ( 20 to 40 ) 9 and even in system with a 20 year old High Output Audio Technica MC. We played that cartridge on FF2011 on a system from our russian friends and the Nobrainer did not sound much worse then a very expensive siberian tube phonostage with an Auditorium 23 transformer. I was surprised. I will check where i find the specs of the Audio Technica. Andre`said that he found it after 20 years in the waist bin. It played just fine, i would even say spectacular on a restored Lenco L75 so please do not tell my great vinyl must be ultra. The russians, not stupid, had placed the table on an enormous substructure though. I will post a photo. Ok, so some think its over damped with a TI ( transimpedance ) stage, other have said like Tominari and Candeas taht it acts like a damper in the suspension of a car, preventing rattle by shorting the moving coil and acting like an eddy current brake. Anyway, i and people i trust like the Nobrainer as is. The last 3 days i have also being working on a discrete version. So there is some new wine in old bottles. So here is the original
circuit. More to come.......

Joachim Gerhard 21st August 2011 01:53 AM

2 Attachment(s)
Here is Georgs Nobrainer on a PCB from Holger Barske.
See also the russian system i liked so much at Frickelfest 2011. See the stand for the Lenco. Cheep and effective.

Joachim Gerhard 21st August 2011 03:34 PM

1 Attachment(s)
I am now out of the concept phase for the Nobrainer Discrete. There will be two versions,
one runs on 12V batteries and the other runs on plus-minus 24V from a PSU that i designed.
I ended up with a circuit that is conceptually as simple as possible. When you look at it you will see that the signal goes only through two BJT transistors and one Fet. In fact in the final circuit there will be two resistors also in the signal chain but the circuit also works without them. At first i tried to improve on the original Nobrainer input stage and one of the inspirations is the ATS MC preamp by Rabeylolles. When it came out in the Autumn 1986 issue of L`Audiophile it was revolutionary and visionary. No other phono input stage fascinated me as much over the years and i also needed years to really understand it. The principles he uses i would call multi use and entanglement. In principle it is the parallel symmetric version of the simple Hiraga current mirror. The differences are besides the symmetry that he supplies the mirror with current sources and adds folded cascodes to make DC coupling possible. Bias to the folded cascodes is from a symmetric current source that is referenced to ground so allows DC offset trim. The current sources that supply the mirror make also sure that the folded cascode sees a high impedance so open loop gain is high. He then feds back the output signal to the current mirror. As a real twist he feds the bases of the folded cascodes with a level shifted version of the Ube generators in the mirror. You can find the circuit and others on the Claudio Bonavolta - Photography - Web Design website. So i did my take on the ATS. See my NOCMUS circuit. That stands for No Capacitor Multiple Use. I use only two current sources instead of the ATSs 4. I simply level shift the current with a diode and feed it into the bases of the folded cascode. A buffered copy of the input signal feeds this point too. The input transistors are cross cascoded with a driven cascode and the second transistors of the input diamond feed the emitters of the folded cascode. Guess what ? It worked although only over a narrow dynamic range until deformation of the input signal set in. I could put more work into this put i put it aside for the moment. The ATS has one disadvantage that bothers me. By using the Ube generators there is always a 3dB noise penalty. Especially the way he designed it with feedback. His stage has the noise of a 130 Ohm resistor and that is rather poor after modern standards. I am thinking more about a noise impedance that is 5 to 10 times lower.

Joachim Gerhard 21st August 2011 04:25 PM

1 Attachment(s)
The ATS is drawn in a strange way so i have redrawn it to make it more digestible to me.

Joachim Gerhard 22nd August 2011 12:00 AM

1 Attachment(s)
Here is the basic idea for the Nobrainer Discrete,

Joachim Gerhard 22nd August 2011 05:49 PM

1 Attachment(s)
Here is a more detailed circuit of the Nobrainer Discrete. I have not build it yet so details may change like the servo and the values of the RIAA components. I also have an idea for a line input with volume control so the buffer can be used for CD too. I will also present a list of transistors that can be substituted, Fets and bipolars because this circuit works with a wide variety of active parts and you may not have the ones in the schematic. The 2N4401, 2N4403 combination are the classic low noise transistors from the 60th, used for example by JC in the original Mark Levinson JC1 MC pre-pre. They still work in this position. They have an Rbbof 40 Ohm and medium Hfe.

Joachim Gerhard 5th September 2011 03:11 AM

1 Attachment(s)
Here is another way to set up the input stage. This has higher input impedance in mind.

Joachim Gerhard 28th October 2011 03:26 PM

1 Attachment(s)
The Nobrainer in Georg Strackes system ( Klangmeister ) at the Krefeld AAA show.

Joachim Gerhard 4th November 2011 05:28 AM

I promised Holger Barske to publish a PSU design for the Nobrainer. The design is based on a circuit that Holger made for the Dispre. I spotted a mistake and made some additions that improve performance : I added common mode chokes, i changes the resistor that supplies the TL431 to 220 Ohm. Holgers original value of 10kOhm will not work. The TL431 needs at least 1 - 2mA to work correctly and the 10kOhm will give only 0,65V / 10000 = 65uA. Maybe this was a typo. I added 220 uF to the adjust Pin for less noise. This is a very simple PSU that nevertheless works and sounds very well.
Circuit comes with the next post.

Joachim Gerhard 4th November 2011 05:37 AM

2 Attachment(s)
Here is the circuit divided into pre regulator and regulator. I would put the pre regulator in a separate box. You can add elcaps before the common mode coils.


All times are GMT. The time now is 08:15 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2