Your outstanding design will certainly improve the sound of your equipment, whatever that is.
Set aside it does not use global negative feedback (a well known sound killer, due to the high order harmonics and PIM impact) it's excellent sound is also the result of the very low output impedance up in the MHz range and very good load and line regulations.
This sort of design has been used since bipolar transistors first came available in the 1950's. It is often considered too simple and imperfect, however it has a number of advantages that make it useful, even today.
Personally, I first used it to IMPROVE an existing commercial +/- 15V supply by using this very design, WITHOUT the Zener diode, (since DC regulation was already established) in the Levinson JC-2 to lower the power supply noise to the phono stage, which, while balanced, had modest power supply rejection.
That was 1973, however, you will find it in almost exact form (with the Zener) in many Parasound power amps of my design over the last 20 years. I do often swap the bipolar transistor for a power mosfet in my better designs, but the circuit is basically the same. It is also true for the CTC Blowtorch, the Parasound JC-2 preamp, and most preamps (without the zener) that I have designed over the decades.
Personally, I first used it to IMPROVE an existing commercial +/- 15V supply by using this very design, WITHOUT the Zener diode, (since DC regulation was already established) in the Levinson JC-2 to lower the power supply noise to the phono stage, which, while balanced, had modest power supply rejection.
That was 1973, however, you will find it in almost exact form (with the Zener) in many Parasound power amps of my design over the last 20 years. I do often swap the bipolar transistor for a power mosfet in my better designs, but the circuit is basically the same. It is also true for the CTC Blowtorch, the Parasound JC-2 preamp, and most preamps (without the zener) that I have designed over the decades.
Last edited:
Maybe I lost track, but are you guys still talking about Joshua's emitter follower or have you switched to a real regulator?
jd
Thank you very much for your comments, now I'm illuminated. However, my question referred to the differences between IC regulator vs. discrete one. Do you have any comment to that question, or you'd be satisfied by only mocking others?
Joshua, it can be debated whether you need to add the .1uf cap across the Zener. The Zener diode has its own internal capacitance and works well into the RF region. Everything else looks OK.
This sort of design (without Zener) is used with 'real' IC regulators and are normally known as capacitor multipliers. They can be very quiet. Adding the Zener, and resistor reduces the 'breathing' from AC power lines and its potential compromise to the circuit, IF you do not use a 'real' regulator in front of the cap multiplier.
It is best to keep the channels separated by using a cap multiplier for every channel, and even every gain block, in an audio preamp or power amp driver stage. This is because the output Z of the regulator of this kind, while being broad-band, is fairly high, even a few ohms, especially if you add short circuit protection in the usual way. The Levinson JC-2 suffered from this.
This sort of design (without Zener) is used with 'real' IC regulators and are normally known as capacitor multipliers. They can be very quiet. Adding the Zener, and resistor reduces the 'breathing' from AC power lines and its potential compromise to the circuit, IF you do not use a 'real' regulator in front of the cap multiplier.
It is best to keep the channels separated by using a cap multiplier for every channel, and even every gain block, in an audio preamp or power amp driver stage. This is because the output Z of the regulator of this kind, while being broad-band, is fairly high, even a few ohms, especially if you add short circuit protection in the usual way. The Levinson JC-2 suffered from this.
Last edited:
Thanks, John.
What are the differences between such a discreet regulator and an IC one?
Also, aren't Zener diodes noisy? If yes, over what frequency range?
What are the differences between such a discreet regulator and an IC one?
Also, aren't Zener diodes noisy? If yes, over what frequency range?
Now that B, E and C of the transistor are nicely decoupled, let me propose a NE68819
from NEC / CEL for Q1. It is not only low noise, but even one of the very few types that have
their 1/f noise guaranteed.
Gerhard
from NEC / CEL for Q1. It is not only low noise, but even one of the very few types that have
their 1/f noise guaranteed.
Gerhard
It is important to note here that the 1/f noise of the output device in THIS example is not so important, unless you are regulating very low current 20ma or less loads. Sometimes, however, that could be important.
could not resist taking the photo, gave the 54750 a lot of work today
2*54751A, 1*54752A + some that are cannibalized for samplers
🙂
Last edited:
I was to lazy to take a picture of mine, so the current picture is stock. I have one 54721A and 2 x 54712A.
I gave the link to help you see the difference. If you don't want to, that's not my problem. I'm done giving free consulting here.
jd
I think that we are talking about a simple linear regulator, as defined in the Wikipedia, as well as design app notes from 50 years ago. What is the problem? Too simple? Too quiet? Too fast?
- Status
- Not open for further replies.