DIYHiFi.org

Somebody asked me about a post I made years ago at the cesspool regarding HV regulators for tubed circuits. The best ones I ever heard were simple followers. One example is the "capacitance multiplier" type of circuit found in the original ARC SP-3a. As I recall, it had a string of Zener diodes to create a reference voltage and then used a high-voltage bipolar transistor in a TO-3 case as the follower. I can't remember if it had a bypass cap on the Zeners, but the schematic diagram is easily available on the web.

This person also asked specifically about the old NYAL Super-It phono stage. I had one of those. It had an HV regulator that had a reference of some type, a floating op-amp in a socketed 8-pin DIP, and a MOSFET pass element. This one was particularly easy to modify. Simply pull the op-amp out of the socket and insert a resistor (I can't remember the value I used, probably 1000 ohms or so) between the input and output pins of where the op-amp used to be. This ties the voltage reference (again, probably a Zener, but I can't remember) to the gate of the MOSFET.

There are variations you can make -- different references besides Zeners, different types of followers, what kind of bypass cap (if any) to use across the zener or at the output. Experiment and have fun.

Charles,
We have been doing CCS fed, shunt regs for tube gear for over 30 years now - and all I can say is that they are sonically worth all the extra trouble they may be. As well as using them in all our commercial gear, we sell them as a kit on our site and the client feedback is all good, often almost overwhelming in their praise!

As a mod to gear with series regs, a mjor sonic improvement results. An ARC SP-8 was my last one.

We of course use the same CCS fed shunt topolgy in our SS gear as well - with, we believe, the same positive results.

Even our simple CD player opamp replacement buffers utilise this form of regulation - ity's too good to leave out to save a $$ or two.

Regards, Allen (Vacuum State)

Hi.

Yes, any regulator needs a reference voltage to keep the I/P voltage regulated. There are various ways of obtaining the required reference voltage, like zener diodes in the old times to OP amps today.

IMO, whatever reference devices used, it is its sonic effect that concerns me most.
I believe simplier the better. I borrowed the very simple basic methodology suggested by
John Broskie - shunt regulator useing a TIP-50 type NPN 500V1A with its emiiter grounded via a LED. It's the LED to provide the reference voltage of 1.8V-2.2V.

For ALL my tube amps, from power (+440V) to phonostages (440V - 230V), I still use this very basic TIP-50 shunt+LED regulator fed by tube rectifiers. Sonically, it is pretty quiet & neutral without sounding sandy.

c-J

I found the lowest ripple and zero wandering (common problem with shunt regs) with broskies LT1085+10M45S series reg PS-2. It is so simple but measures excellent.

I'm not using fancy caps or oil cans yet so I can't comment on the sound (little harsh in bass with the stock electrolytics.)

Folks,

Not to offend, but I find all this "best regulator" talk amusing, but ultimately futile.

There is no single best solution.

After all, a Solution must answer a specific equation with a number specific variable values.

And if the equation and variables change, then the solution changes.

So, I would always ask "What are we trying to achieve?" and then seek a solution that answers the requirements.

For example, all shunt regulators I know and have been able to come up with use looped feedback.The best I find loop around a single device. But with single devices there are drawbacks...

Equally, I find that all references are very noisy (compared to the regulator circuit). Maybe sometimes we do not need a "reference"?

And I am always incredibly amused by authors of the latest SuperHyperMegaCalifragilistic Regulator advertising the incredible DC precision of their creations, when what makes the "sound" is all AC and transient behaver of the circuit and often 10% regulation at DC are perfectly sufficient for the circuit supplied...

Ciao T

You should have been here yesterday, to join our "best sounding op-amp" fiasco.

Jocko

Jocko,



Yes, that looked like fun.

Instead I spend my time hanging around JFK waiting for my airline to open the counter so I could go on standby for a 14 Hour transpacific flight which I actually got on.

So you guy's had way more fun than I had...

Of course, the best sounding op-amp is the one that is not there...

Ciao T

Mine was an unusual design where a set stable voltage was critical, it had to maintain this voltage as the wall mains voltage varied. It is a hybrid amp where the the DC servo'd SS output buffer was coupled to the tube stage with caps that would let a wandering voltage reg pass very low HZ AC. This would end up with a few mV's DC on the output which is bad for headphones. Broskie's PS-1 was the only HV PS I could find that had low ripple and no wandering. But I would like to try others if they are out there.

Konnichiwa Thorsten

Although I am trying to avoid silicon I am now starting to warm up to the idea of having a "self referencing" regulator which sits a small percentgae below the B+ so that the B+ can fluctuate and we still have smoothing of the ripple.

Sayonara AM

another example