Best low noise regulator?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi salas,

I'm sure that your SLS (Salas-Lamb-Shunt) impressed relatives and friends...;)
I'm finishing the layout for the quadraple shunts and because experimenting with zeners on the live circuit will be PIA, I'm wondering if there is a simple test circuit that could put together for that purpose. Any suggestions ?
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Hello to you too.

You can simply use a 20V-24V voltage source (your psu plus side will do), and just with a protection resistor you can check the exact Zener voltage of each one you have.

P.S. Hey, you have promised pics when ready. Don't forget. :D
 

Attachments

  • zmatch.gif
    zmatch.gif
    1.6 KB · Views: 1,342
Hi Salas,

I have read this thread and some others where you explain your design and how it sounds and I have one question. . . .

When you explain the clear & clean sound from using your regulators I am interested to know what other measures you had taken up stream to clean up the incoming supply ?

For instance . . .

Do you use mains filters ?

Do you use Schottky diodes or similar ?

Do you use a snubber on the secondary to prevent ringing when the diodes switch off ?

Do you use any chokes to filter out HF & RF noise ?

Do you use fancy filter caps for maximum filtration ?

Have you ever tried using two of your regulators in series and if you did, did you hear an improvement ?

As you may have guessed I am trying to work out if your regulator is so effective that you do not need to take any other measures to get best possible results.

In the past I have found any measures I have taken to reduce noise have always brought audible improvements but I have not yet tried any "super regulators" but have had some very good results with C R C RL C filters and all the stuff I mentioned above.

many thanks for all of your contributions

mike mears
 
yes - interesting idea

If this is indeed true, it would confirm my belief that much of the noise of 3 pin regulators is passed though from noise upstream, rather than generated within the chips.

wonder if this idea would considered an ultimate solution . . . at this stage I want to focus on the very best I can achieve.

thanks

mike
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Nothing fancy really. I use mainly UF4007, rarely better, plus a 10-33R power resistor before the main filter cap if I have some volts to spare. A guy in the Riaa thread had mains input filter and CLC after rectification but his chokes were not adequate for the current, losing almost all of their inductance, so he got correct Hammond ones and he only reported a very slight subjective difference for the better. All that in sensitive MC phono stages with rudimentary PSRR.
No mains filters or better than MUR diodes I have seen implemented and listened to. I would use a CLC just to avoid possible RF and rectification junk. My main capacitors choices are good value for money ones like Panasonic FC or BHC slit foil, or Elna For Audio. No snubbers I have tried. No two regs in series also.
The main reason I did not go far with pre filtering is that the sound was right, and I could not feel real differences between main filter capacitors ranging from 2200 to 10000uF of various good makes. So I thought it may be immune enough. I did not perceive harshness. On the other hand, Ricardo over the Riaa thread who is a dedicated tweaker, can hear the main caps and changes for the better. So from what the community shares, I would recommend a CLC and trusted caps in diy circles. Ricardo talks about Nichicon KG and Jensen 4 pole. Also soft recovery diodes that people know and trust in DIY threads. Prefer softer over faster. What I know it changes the tone of the sonics is the output capacitor and there I like Silmic II or some NOS BG NP Red if you can get. Also short and thick twisted wire between the shunt and the audio circuit that receives power, is audible, plus a 0.1uF stacked ceramic or fast PP with short legs on B+ to ground at the audio circuit, can help if there is some RF around.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
mikelm said:
. . . at this stage I want to focus on the very best I can achieve.

thanks

mike

You gotta build some tried ones and listen then. Many people are reporting good sonics with mine as well as with some others. Its not complex, put together a P2P one, tweak some parts, give it 48 hours play and then you decide. A cheap easy to make reg like this can be used in a less critical application later, if you chose some other finally. Just change a Zener if the voltage needs will differ.
 
Salas,

Thanks for your feedback and ideas.

I will indeed try some of these ideas over the next few weeks and will report my conclusions here.

I am not expecting a drastic change because all the passive measures I mention above also create a very sweet and clear sound, but I keep an open mind and am interested to see what I notice.

For my contribution I offer this . . .

The one biggest improvement I have ever noticed on power supply design for least cost & effort is a 0.1 - 1.0uF cap in series with a 47 - 100 ohm resistor directly across each secondary coil of the transformer.

This has the effect of reducing the frequency, and damping the ringing that always happens in the transformer coils when the diodes suddenly stop the current flow. If the values are fine tuned for each transformer the ringing can be damped to nothing - but any values between those mentioned above still has an extremely good effect.

Some people just use a cap and no resistor and this does help the sound very much but it only reduces frequency of ringing and does not give any damping so with no resistor the magnitude of ringing can get worse causing some extra heat in the transformer.

I hope some of you enjoy the effect of this easy mod

cheers

mike
 
diyAudio Chief Moderator
Joined 2002
Paid Member
I use a 1uF cap in that position in my KT88 amp. I guess I have to see for RC before rectification for the shunt in my Riaa, where things are more sensitive and more audible. Thanks for that.

Also there is a dedicated thread that Ikoflexer opened about my shunt, so if you make it, its better to report there, so the info will be concentrated better for the community.

Regards
 
I do not have a sk170 model so resistor values will be different on CCS to give 3mA.

The mosfets are zvp3306a but I will probably build with zvp3310a.

Output impedance is about 7m ohms to about 1meg hz ( without the 10ohm resistor )

Stability is mostly governed by value of o/p capacitor and I will aim to avoid any ringing even on fastest transient.

The 10 ohm resistor is optional ( at least in spice ) but is a neat way to reduce open loop gain and increase stability if that is needed.

replacing the zener / LED with a resistor was just an idea that I will test - not sure yet how this will impact noise so I will experiment in this area.

I will also experiment to see how many of the upstream noise reducing measures I list above are needed with this design.

cheers

mike
 

iko

Ex-Moderator
Joined 2008
Just in case you needed them.

Changing the betap parameter will change the Idss.

.subckt 2sk170 1 2 3 params: betap=33m
j1 1 2 3 jx
.model jx NJF(Beta={betap} Rs=4.151 Rd=4.151 Betatce=-.5 Lambda=1.923m Vto=-.5024 Vtotc=-2.5m Cgd=20p M=.3805 Pb=.4746 Fc=.5 Cgs=25.48p Isr=84.77p Nr=2 Is=8.477p N=1 Xti=3 Alpha=10u Vk=100 Kf=111.3E-18 Af=1)
.ends

.subckt 2sj74 1 2 3 params: betap=30m
j1 1 2 3 jx
.model jx PJF(Beta={betap} Rs=7.748 Rd=7.748 Betatce=-.5 Lambda=4.464m Vto=-.5428 Vtotc=-2.5m Cgd=85.67p M=.3246 Pb=.3905 Fc=.5 Cgs=78.27p Isr=129.8p Nr=2 Is=12.98p N=1 Xti=3 Alpha=10u Vk=100 Kf=26.64E-18 Af=1)
.ends
 
Teddy Pardo found that what went in front of an ALW Jung super regulator (ie the pre-regulator) had a very significant effect on the performance (significant audible effect).
I am not sure that what he has done is entirely relevant to this design, I just want to make sure we cover all bases (it is an interesting read at least) http://www.pinkfishmedia.net/forum/showthread.php?t=29501

So I don't think we should neglect this area, considering the great work that has gone into this so far (unless this design is immune to front end performance).
 
I made a veroboard p2p version of the circuit from posting #67

I made the input CCS without cascode, though.
On my 2mV/div scope I can hardly tell if the channel is open at all. But I do get all from small to very hefty oscillations when placing the cap across R5 depending on its value and the chosen opamps. Slower ones like the op27 might put out a wrong voltage in this case. Apart from the cap issue the thing seems pretty stable, and a fast opamp like an OPA637 or a LT1028 works fine.

Rüdiger
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.