High(er)-voltage, low noise v-reg?

[Khron]

Greetings...

I'm currently searching for some low-noise voltage regulators, to provide some low-ish current, medium-high voltage for the input/VAS stages in some solid-state discrete class AB amplifiers.

The LC Audio design looks ok, but it can only go so high as far as voltage is concerned (limited by the opamp's max supply rails). I'd need something in the order of 35-50V, 1A(max).

Monolithic designs (ie. LM317/337) kinda lack the "low-noise" requirement i'm looking for...

I'll be eagerly awaiting some indications / ideas...

 

[jackinnj]

Greetings...

I'm currently searching for some low-noise voltage regulators, to provide some low-ish current, medium-high voltage for the input/VAS stages in some solid-state discrete class AB amplifiers.

The LC Audio design looks ok, but it can only go so high as far as voltage is concerned (limited by the opamp's max supply rails). I'd need something in the order of 35-50V, 1A(max).

Monolithic designs (ie. LM317/337) kinda lack the "low-noise" requirement i'm looking for...

I'll be eagerly awaiting some indications / ideas...

If you bypass the adjust pin you knock the noise down pretty dramatically.

Secondly, you can use a cleanup circuit after the discrete regulator -- http://www.wenzel.com/documents/finesse.html

 

[Khron]

Regarding that finesse regulator... For the negative rail, how should i approach it? Use the transistors' complements and make a "mirror image" of the schematic? Or do other modifications need to be done?

 

[jackinnj]

you might be lucky enough that your transformer has two secondaries (a lot of toroids do) -- in which case 2 identical regulators would work.

 

[peranders]

Monolithic designs (ie. LM317/337) kinda lack the "low-noise" requirement i'm looking for...

How low noise are you looking for then? It could be an idea to set a target and then see what options you'll have.

 

[richie00boy]

I was also wondering how low noise a power amp front end needs to be? I find an optimised 317/337 PSU fine and very quiet. It's going to be vastly superior to running the front end off the main amp supply anyway, even without bypassed adj pin.

 

[banana]

you might be lucky enough that your transformer has two secondaries (a lot of toroids do) -- in which case 2 identical regulators would work.

I agree it'll work. But I always think that PNP regulator for -ve rail is a better idea, though I don't have a good reason for that.

Maybe it's because the +ve pin cannot source current, or it's the feedback referencing to rail instead of ground, or maybe it's nothing wrong...

Anyone have more information on this?

 

[Khron]

Well, frankly, i haven't set any numerical goals, i was just "testing the waters", to see what other/better options were out there

 

[Khron]

I played around in Eagle and ended up with something like in the file i attached.

If i did any major blunders (as in "very likely noisy ground" or any ground loops), please bring them to attention

 

[tinitus]

Secondly, you can use a cleanup circuit after the discrete regulator -- Finesse Voltage Regulator Noise!

still valid ?

if trafo voltage is just a few volt too much for the regulator, would it be ok to use zener diode before the regulator ?

edit, a search on pre regulators shows the use of additional transistor together with the zener
but I guess thats for when input voltage is much higher than just a few volts too much ?