This is an input stage of a line preamp. I do not want to design a phono input that way.
My plan is to go balanced out of the phono and then into that circuit and then into my unbalanced linestage. So it is just to convert the balanced line into unbalanced.
i have my turntable quite far from my preamp (5m). In the moment i use a 75 Ohm coax video cable and it works fine. I whould like to find out if a balnaced connection over that
quite long lenghth brings any benefit over what i have now, I fear not because i have no problem with noise whatsoever. Some peolple claim audible benefits of balanced connections and i am just curious.
The discreet INA i have shown in my last posts brings audible benefits. To my ear it sounds more dynamic on both end of the scale. Soft things sound softer without blurring of detail and loud things sound louder. That brings the unconvenience that i have to adjust the volume quite often. This circuit is 3dB more noisy then a unbalnced stage with the same amout of active input elements but sounds more quiet when the music is playing.
On the speculative side i have a totally different idea for a phono input.
I whould like to measure the current and the voltage balanced and mix that together in a summing stage. I made a quick scetch in a moment of meditation. When i find it i will post it. I have no idea if that works but sometime even in a mistake is a moment of truth.
My plan is to go balanced out of the phono and then into that circuit and then into my unbalanced linestage. So it is just to convert the balanced line into unbalanced.
i have my turntable quite far from my preamp (5m). In the moment i use a 75 Ohm coax video cable and it works fine. I whould like to find out if a balnaced connection over that
quite long lenghth brings any benefit over what i have now, I fear not because i have no problem with noise whatsoever. Some peolple claim audible benefits of balanced connections and i am just curious.
The discreet INA i have shown in my last posts brings audible benefits. To my ear it sounds more dynamic on both end of the scale. Soft things sound softer without blurring of detail and loud things sound louder. That brings the unconvenience that i have to adjust the volume quite often. This circuit is 3dB more noisy then a unbalnced stage with the same amout of active input elements but sounds more quiet when the music is playing.
On the speculative side i have a totally different idea for a phono input.
I whould like to measure the current and the voltage balanced and mix that together in a summing stage. I made a quick scetch in a moment of meditation. When i find it i will post it. I have no idea if that works but sometime even in a mistake is a moment of truth.
I got Douglas Self´s new book "Small Signal Audio Design" and found two fascinating circuits that could be usefull.
One lowers noise at the input of an MM stage by lowing the contribution of the 47kOhm loading resistor. See the schematic and the explanation i put there.
The other solves my Balanced - In , Unbalanced - Out problem.
The superbal stucture i have shown is 14.2 dB noisier then an unbalaced solution and put quite heavy loading on the line.
Here i show Self´s Low Noise Unity-Gain Balanced Input Stage that has only a 0.1dB disadvantage over unbalnced in case you use quad NE5532 as buffers and quad LM4562 as diff. amps. With dual AD797 and quad diff. amp AD797 you even get a noise advantage of -1.6dB over the unbalnced situation. Noise output is -120.6 dBu now.
I am also reading Burghard Vogel " The sound of silence" to get a better theoretical understanding of noise. I am perfectly happy with the practical results i got and did not find anything so far that i do not know already but i will be better equipped to answer more detailed theoretical questions in the future.
In my privat time i am now reading "Anathem" from Neal Stephenson.
One lowers noise at the input of an MM stage by lowing the contribution of the 47kOhm loading resistor. See the schematic and the explanation i put there.
The other solves my Balanced - In , Unbalanced - Out problem.
The superbal stucture i have shown is 14.2 dB noisier then an unbalaced solution and put quite heavy loading on the line.
Here i show Self´s Low Noise Unity-Gain Balanced Input Stage that has only a 0.1dB disadvantage over unbalnced in case you use quad NE5532 as buffers and quad LM4562 as diff. amps. With dual AD797 and quad diff. amp AD797 you even get a noise advantage of -1.6dB over the unbalnced situation. Noise output is -120.6 dBu now.
I am also reading Burghard Vogel " The sound of silence" to get a better theoretical understanding of noise. I am perfectly happy with the practical results i got and did not find anything so far that i do not know already but i will be better equipped to answer more detailed theoretical questions in the future.
In my privat time i am now reading "Anathem" from Neal Stephenson.
Joachim,
You called me a couple of days ago. I called your office yesterday to reply but you were out.
I'm here at work today and will be here untill 18:00.
Call me: +41 52 6203581
Regards, Allen (Vacuum State)
You called me a couple of days ago. I called your office yesterday to reply but you were out.
I'm here at work today and will be here untill 18:00.
Call me: +41 52 6203581
Regards, Allen (Vacuum State)
The op-amp current noise will get you certainly in the second example. In the first a typical NE5534 (that has been added) is .4pA/rt-Hz and the 47K is .58pA/rt-Hz not to mention the other added noise sources. The 1.5dB improvement (if this is correct) is only at higher frequencies anyway. In first example you would be much better off buffering with JFET source follower at Idss. BTW the "cold" resistor trick is pretty old.
Yes, i heard some hours ago that Dick Burween came up with the "cold" resistor idea 40 years ago. For me it is new and this is certainly true for a lot of engineers here in europe. Maybe we are simply more stupid. Certainly with the exception of the NE5534 we are not famous to have contibuted to analog IC design much. The focus here was more on discreet design and many circuits where of the "good enough to be not audible" variety. We simply did not have a "High End" here until the late 70th. A lot of good work was done in professional circuits and when you hear a direct broadcast from Die alte Oper in Frankfurt you know what i mean. In the states this "trick" seems to be known well but why is nobody using it ?
Thanks for the idea with the Fet buffers. I have build circuits like that and they perform very well. I did a source follower buffer with LSK389C ( top source folower, down CCS Idss) that performed extremely well. Fed from a low source resistance ( 50 Ohm) frequency response was down -3dB at 6.5MHz and distortion was around 0.008 % second harmonic @ 2V.
The AD797 certainly is a special specimen and when i look at Samuel Groners work it is not beaten even today. I use it in many of my commercial phonostages with flattering results.
Thanks for the idea with the Fet buffers. I have build circuits like that and they perform very well. I did a source follower buffer with LSK389C ( top source folower, down CCS Idss) that performed extremely well. Fed from a low source resistance ( 50 Ohm) frequency response was down -3dB at 6.5MHz and distortion was around 0.008 % second harmonic @ 2V.
The AD797 certainly is a special specimen and when i look at Samuel Groners work it is not beaten even today. I use it in many of my commercial phonostages with flattering results.
Hi Scott ! Could that work any better? The double Fet i am using has 10 - 20mA Idss (LSK389C) so it could swing enough voltage into that load. I am showing only halve of the circuit for simplicity. The OP´s are AD797.
http://www.linearsystems.com/datasheets/LSK389.pdf
http://www.linearsystems.com/datasheets/LSK389.pdf
Joachim, please understand, that before and maybe during WW2, Germany was ahead of virtually every country in mechanical engineering and electronics. However, after WW2 and because the 'cold war' forced the USA to invest in the newest and best electronics that was possible, we jumped ahead for a few decades. In fact, there was a 'brain drain' of many of the best engineers in Europe, during the 60's. Americans are intellectually lazy as a group, but when motivated, they learn fast. We were ahead from perhaps 1960-1975. Then the Japanese 'cleaned our clocks' by making superior semiconductors, that we all use today. Now, Germany is back into action. It is not a personal thing.
I think we made very good capacitors in the 60th and 70th. I still have to find a better cap then the Röderstein KP1822.
Today we are still making good raw material. I talked to RTI that make caps for us and they buy the foils in Germany. Brian Daily has just specked an oil filled Teflon ! cap for bypass in our speakers.
Today we are still making good raw material. I talked to RTI that make caps for us and they buy the foils in Germany. Brian Daily has just specked an oil filled Teflon ! cap for bypass in our speakers.
That would have a much better chance of working, the signal levels are so low that op-amp followers are just a waste of devices.
Radar? How about the development of the B17 and the P51, 117 days start to finish! Are you some kind of self hating American (please apply liberal smilies here). 🙂
As I said, before WW2, our science and engineering in the USA was mostly geared to make more cars, etc, more cheaply. Radar was invented by the British, I am pretty sure, and for a darn good reason. This was more 'basic research' than engineering, at least at first. The British scientific establishment has always been among the very best. However, engineering has been sometimes lacking. Look at 'Darkness by Lucas', etc
When it came to electronics, especially solid state, we Americans were way ahead of the game, even in the 1960's. We were designing complementary symmetry pp throughout power amps, before the British were even into dual supplies, for the most part. I have more evidence, but so what? The best book I have on fet design is British. The best idea I ever got from an article was British. And so it goes.
When it came to electronics, especially solid state, we Americans were way ahead of the game, even in the 1960's. We were designing complementary symmetry pp throughout power amps, before the British were even into dual supplies, for the most part. I have more evidence, but so what? The best book I have on fet design is British. The best idea I ever got from an article was British. And so it goes.
I worked with Malcolm Hawksford in Essex. That changed my life.
I was also a big fan of Peter Walker. Never liked his electronics much ( ideas where briliant, but execusion so so) but owned the ESL57 for 7 years and even build an HQD system with Levinson crossover and Threshold Stasis amps and NS10 preamp.
I was also a big fan of Peter Walker. Never liked his electronics much ( ideas where briliant, but execusion so so) but owned the ESL57 for 7 years and even build an HQD system with Levinson crossover and Threshold Stasis amps and NS10 preamp.
I heard that it needs 20 years in germany until an invention, for example made at a university, is put into a commercial product.
Right you are, both radar and JFETs were invented in Germany decades before they were useful. The inventors usually get more credit than the ones that bring it to practice.