The simplistic Salas low voltage shunt regulator

[Salas]

Originally Posted by HiFiNutNut

So could I try the ones measured at 3.5mA, for example?

Will force higher value resistors that will make your film dream into cinemascope. If you start to feel it opens up but starts to lose slam, up the CCS current.

[HiFiNutNut]

Originally Posted by Salas

Don't go thinking with OPA pssr because in a 627 pre first thing V1 builder Marinos got, was bass dynamics in his posts early in another thread and in this one. And we talk VS large batteries.

I had a battery powered preamp 10 years ago for a few years. I don't think it was as good as my simple opa627 buffer.

I have read that battery power is not necessarily low impedance and its AC performance won't be as good as a well design regulator.

When I try equipments, I always find that the higher the PSSR the better the sound. I guess that the problem with all opamps is that they are good at low frequencies but bad at higher frequencies.

My build of v1.1 with film caps should behave very well above 200Hz at least, even using the current value of 2k4 resistor and 1.1uF capacitor. So for me it is a solved problem. If the lower Idss could lift the resistor to about 5k, I would be happy to use 2 x 1.1uF with a single regulator to drive the entire circuit. If not, I will use two separate regulators to drive the high-mid and bass-sub separately.

[Salas]

Don't get your output cap very low though IMHO. It does not terminate as it should in rise time for driving current to the shunt mosfet. What is the main benefit you like for your films across vref? Less grain and lumpiness? And then you will compare boutique ones? Never ending story... Batteries must be bypassed with thousands of uf, still their impedance rises with frequency a lot though. They also can present weird chemical noise. In general, it gets a truckload of special high amperage batteries in parallel to compete. Not elegant at least. He had it much better than with 317-337 already with the batteries as he said, so the comment goes to ''it will better be a shunt in your lows too'', if the guys experience has anything to do with your line circuits.

[ikoflexer]

You can lower the current through Vref by inserting a resistor between the gate and source of the jfet. The gate stays connected to R5. Then you can increase R7 and Rtrim, if you want that. I run in v2 about 250uA through Vref.

[Salas]

It does work, but the internal impedance will go up, so the root noise of the jfet too. A small thing, but I may remind.

[HiFiNutNut]

Originally Posted by Salas

Don't get your output cap very low though IMHO. It does not terminate as it should in rise time for driving current to the shunt mosfet.

The output cap is and will be electrolytic. I have never thought of film cap in that place. I don't even think I will bypass the electrolytic with film.

Originally Posted by Salas

What is the main benefit you like for your films across vref?

Replace your 470uF with a 1+uF film cap and hear it.

[HiFiNutNut]

Ikoflexer,

When you did the simulations with estonishing low impedance figures in the first couple of pages of this thread, did you include some ESR into the Vref bypass capacitor?

Regards,
Bill

[Salas]

Originally Posted by HiFiNutNut

Replace your 470uF with a 1+uF film cap and hear it.

I have done all those things. That is why I am asking what you hear so to see if it correlates. I just give ''behaving'' practical and economic enough values that work more or less balanced in many systems and are easy for layout. In my personal phono I use 10uF Mundorf plain supreme.

[HiFiNutNut]

Originally Posted by ikoflexer

You can lower the current through Vref by inserting a resistor between the gate and source of the jfet. The gate stays connected to R5. Then you can increase R7 and Rtrim, if you want that. I run in v2 about 250uA through Vref.

250uA through Vref? You must be talking about the Vref in the CCS, not in the shunt.

With a resistor between g and s, the noise increase will be so little to justify any concerns, right?

[ikoflexer]

Originally Posted by HiFiNutNut

Ikoflexer,
When you did the simulations with estonishing low impedance figures in the first couple of pages of this thread, did you include some ESR into the Vref bypass capacitor?
Regards,
Bill

In all simulations I try to pretend a somewhat careless implementation; For caps I usually do ESR = 1R and ESL = 15uH. Pretty extreme eh?