The Kuartlotron - keantoken's simple error-correction superbuffer

Hello everyone!

for my phono preamp I did a Borbely-Sulzer supply with soft start and it works fine but is a bit too complicated I think for a little editing as Kuartlotron.

Regard's!
 

Attachments

  • Borbely - Sulzer (Top).png
    Borbely - Sulzer (Top).png
    86.8 KB · Views: 514
Please give more details.Where we can find schematic and pcb to built this?
I gathered everything I found on the subject and I designed a circuit that adjusts the output voltage and includes a soft start with more or less delay.
I have drawn two circuits (with a transformer on the circuit board and another without transformer).
This power supply is very stable and quiet but is rather complex.
I can offer my circuits but I would have drawn a diagram with some explanations for some points.
 
If you are asking me:
5. Why virtual GND when center tapped transformers are standard ?
6. Why TL431 when it can be better ? I spotted a 15V Zener diode. We can do better than that and also more compact. SMD ultra low noise regs + cheap 2 x 8V transformer....
7. I don't see much decoupling in the schematic. Doesn't it need decoupling caps ?

* 2 x 8V transformer would give way more voltage with this load. It has to be tested but with LDO regs I think it will work out OK. A 2 x 9V will work for sure.

5: Center tap doesn't improve anything AFAIK.

6: If you actually look at the TI datasheet for the TL431 it's quite good as far as I can tell. The zener is just to reduce the temperature for the TL431, in that position it does not add any noise.

7: The schematic is just an overview, I drew it quickly.
 
Last edited:
Disabled Account
Joined 2002
5: Center tap doesn't improve anything AFAIK

It does not deteriorate things either. Isn't a real GND better (lower impedance...) in your opinion ? As said, dual tapped transformers are standard just as dual supplies. Yeah progress is unstoppable ;)

I don't use TL431 at all. The datasheet from TI is clear but in my experience many have to much noise and most of the time I need more current. Admitted, this was tested long ago. Maybe they improved them, I don't know.
 
Last edited:
It does not deteriorate things either. Isn't a real GND better (lower impedance...) in your opinion ? As said, dual tapped transformers are standard just as dual supplies. Yeah progress is unstoppable ;)

I don't use TL431 at all. The datasheet from TI is clear but in my experience many have to much noise and most of the time I need more current. Admitted, this was tested long ago. Maybe they improved them, I don't know.

But that means you're using it to supply? Using the shunt current for power?

What about using it as referance?

What's a good 3 leg replacement?
 
Nevermind, I remembered the TL431 incorrectly. The output impedance is larger than would be for a KM. It is still useful, but needs boosted to get low output impedance. I can combine it with the KM to get a very accurate voltage regulator.

A center tap is just a midpoint. In power amps it is necessary because there may be very uneven current draw from the rails and the center tap can absorb the difference. There are no other magical qualities of center taps that I know of. I do know they they allow a third path of current from the rectifiers which can cause uneven rectifier currents and thus intermodulation with the rectifier pulses.
 
Salas' regulator is very good, it would not be a bad choice at all! The Jung/Didden supereg is also very good and if you wanted the ultimate performance then it might be a good idea to build both and do a listening text. I think the supereg gets lower output impedance, based on the Linear Audio article of some time ago.
 
Here's what I'm thinking.

If you were really worried I suppose you could do two 1uf electrolytic caps?

With the rejection of noise, and low impedance of the KM setup, I'm not sure this is the place to worry about noise so much as the power source to it. Although I won't argue other regulators mentioned here have some outstanding graphs...

Speaking from experience with the use of the KM in this configuration, it undeniable works well even if I don't have graphs of it in a shootout of regulators.

Yet I'm really curious about what JP might mention.
 
I have built several of the Salas DCB1 buffers, and the lower the impedance of the supply the better they sound. Salas says this himself and that's why he recommends "hotrodding". The best proof is in the listening, if someone tries it and says the regulator does not make a difference then I'd tend to believe them, but so far no one has tried it. I can only say that it makes a worthwhile difference with the DCB1, and it might be the same case with this buffer.
 
Disabled Account
Joined 2002
First thing in audio to make things better is to use good if not excellent power supplies. Simple. Also a simple schematic, good power supplies do not need to be more complex than average ones. Salas shunt supplies are fine (but they're not small or cheap). So are the newer LDO regs with astonishing low noise numbers. If you are into low component count, low board space and low noise the newer LDO's might be just what the doctor ordered but then no K multiplier necessary....Uh oh...

I suggest to make some decisions and set up first. You will notice everybody wants something different which can slow down things. Like me posting in this thread ;)
 
Last edited:
First thing in audio to make things better is to use good power supplies. Simple.

We've chatted a little bit on this before. My way would involve a PSU before the KM that's larger than the whole buffer... starting from the transformer(s).

I believe it'd be a separate device (or two), but perhaps people would be interested. It'd basically be a dual rail vs. of the Antipole. At the same time some boards for the transformer with filter would be nice too. (I've got something, but it's never been made)

There's nothing "cheap" about what I use. And yet if you want to try and match any of it commercially your pockets better be expansive.

JP, what LDO's specifically? Can you name one please.

For the record the KM I've been using needed regulation due to power use, but sound very good without regulation just underpowered.
 
Last edited:
Disabled Account
Joined 2002
Hi, I already have a design but mentioning the regs will result in many pages of comment that will pollute this thread. I have some experience with Group Buys. Best is to cooperate and present a design if a real working device is the goal. If people have choice the design cycle will be eternal. Regarding your PSU, I am only into simple but effective one board designs with just the transformer external (when possible). Sorry.
 
Last edited:
I think this project is best kept as something you feed DC into. I said it earlier, people love to do different PSU designs. Who cares if it's an Xbox PSU? People will be enjoying the project.

In this light, Keantoken's regulated version makes perfect sense to me. If you want 20v, you just need to feed it around 25-28v. However I'd suggest making it 24v compatible for batteries. I'd assume 16-18v operating would be good?
 
Disabled Account
Joined 2002
Every buffer works best when fed with DC. As said before: many people, many ideas, no PCB :)

Adding pads for regs makes it a design that can be used immediately. Check the project drawer of many DIYers and check what unfinished projects are in there and why. After a few GB's I have come to the conclusion that ease of building, no wiring and simplicity are valued by many that want a device to be finished in a case ready for everyday use. Just a small % wants to fiddle around with different PSU's etc. connected to a breadboard with crocodile clips. You can't say that I don't have a strong opinion in this :D
 
Last edited: