Lets suppose that you have a simple class AB amplifier choose any you like the DX amplifier or a P3A
lets suppose that i am running it from a rowbust regulated power supply for rails
( to regulate the first stages is a technology that i am familiar with so comments about that might help but its not the target )
Will a regulated power supply work better or a capacitance multiplier for amplifier will work better ?
can you comment what's in it for me ?
what will I gain if my rails are fully regulated ?
In my world a P3A runs from 600W trafo and 60K ufd for 2 chanels
So a regulated PSU or a capacitance multiplier have to give me something more than my standard PSU
lets suppose that i am running it from a rowbust regulated power supply for rails
( to regulate the first stages is a technology that i am familiar with so comments about that might help but its not the target )
Will a regulated power supply work better or a capacitance multiplier for amplifier will work better ?
can you comment what's in it for me ?
what will I gain if my rails are fully regulated ?
In my world a P3A runs from 600W trafo and 60K ufd for 2 chanels
So a regulated PSU or a capacitance multiplier have to give me something more than my standard PSU
I think that voltage regulation has 2 significant benefits. As required, it allows the amp. to operate safely, right up to its max. peak power rating and that same regulating action also cleans up most noise coming from the mains and PSU. A capacitance multiplier only cleans up the noise, trading off the regulation for better efficiency.
In my experience though, regulation is a wasteful safety feature for a DIY power amplifier, where the typical over-sized output stages design could be powered via a simpler cap. multiplier, doing a similar filtering job with less wasted power. There is also a question mark over the dynamic impedance of regulators and their effect on audio quality but I haven't seem any convincing evidence - just personal comment and simulated comparisons. Maybe others here have some credible references. Your 600W trafo is certainly overkill and plenty of regulation already though. Why in the world would you need that monster......perhaps it was just what was available?
JLH (John Linsley Hood) and Rod Elliott discuss the pros and cons in easily understood terms, for anyone wanting to know more. The class A (TCAAS) site for JLH designs is also hosted and indexed at Rod's ESP site: https://www.sound-au.com/
In my experience though, regulation is a wasteful safety feature for a DIY power amplifier, where the typical over-sized output stages design could be powered via a simpler cap. multiplier, doing a similar filtering job with less wasted power. There is also a question mark over the dynamic impedance of regulators and their effect on audio quality but I haven't seem any convincing evidence - just personal comment and simulated comparisons. Maybe others here have some credible references. Your 600W trafo is certainly overkill and plenty of regulation already though. Why in the world would you need that monster......perhaps it was just what was available?
JLH (John Linsley Hood) and Rod Elliott discuss the pros and cons in easily understood terms, for anyone wanting to know more. The class A (TCAAS) site for JLH designs is also hosted and indexed at Rod's ESP site: https://www.sound-au.com/
Last edited:
Two advantages to regulated rails. 1 Allows simple resistance feed to input transistor & VAS, no CCS transistors resistors & ****. Cutts parts count, wire count, and number of wires to install if building point to point (I did AX6 ptp). Yeah, CCS is $.08 in parts. PCBs for CCS designs come 99.9999% from a country I chose not to buy from. Each wire point to point is 10-20 minutes.
2. Allows to run closer to SOA of output transistors if unregulated voltage is too high. My ST120 had 72 volt rail, too much for 2n3055. Lots of blown output transistors in that model as OEM. The supply of domestically made 44 to 50 v full load no center tap transformers is much more prevalent than the 36-40 dual winding transformers for modern designs. 99.9999% of toroid transformers come from one country I don't buy from. TO3 transistors can be bought from a country with no nukes or aircraft carriers, and heat sinks can be bought military or industrial surplus.
2. Allows to run closer to SOA of output transistors if unregulated voltage is too high. My ST120 had 72 volt rail, too much for 2n3055. Lots of blown output transistors in that model as OEM. The supply of domestically made 44 to 50 v full load no center tap transformers is much more prevalent than the 36-40 dual winding transformers for modern designs. 99.9999% of toroid transformers come from one country I don't buy from. TO3 transistors can be bought from a country with no nukes or aircraft carriers, and heat sinks can be bought military or industrial surplus.
Last edited:
Any eventual gain from the regulated PS would mostly depend on the specific amplifier properties like PSRR, output impedance, operating class and dynamic load imposed on the PS, and so on.
There will be no measured change in distortion with regulated rails. I’m not aware of any test case that has confirmed change in amplifier distortion with usual measurement techniques, which rely on stationary signal and averaged result.
For an amplifier with high PSRR, even high rails ripple won’t result with any noticeable hum. Channels crosstalk through common single PS, caused by rails modulation with load current, will be practically nonexistent as well.
However, for an amplifier with low PSRR situation is different and in favor for some sort of regulation or ripple reduction (regulator or capacitance multiplier if not CLC). Capacitance multiplier is not great with transient load ( Why is good to have large capacitor at the capacitance multiplier output ), so regulator has an advantage on condition that its transient response is good.
Capacitance multiplier doesn’t have any efficiency advantage, as usual voltage drop across Cap Mx is 3V, while serial LDO regulator can work with less. Single common regulated PS can be used with perfect results for both channels, even on zero PSRR amplifier, as rails modulation and harmonics of good regulator under load will be in the uV range, while heavy load transients could be in the single digit mV range.
Any discussion on improved or ruined sound by regulated PS is a big can of worms. We can rely only on easily misguided subjective impressions, so we better don’t.
For me, regulated PS is one part of puzzle called good sound reproduction, that is in the right place and I can concentrate on other parts, as this one is 'perfect'.
The power supply of the amplifier operates on a pulsating audio frequency current. Stabilized too - it's not direct current. Making a good power supply is harder than making an amplifier.
I thought cap multipliers were more intended to work at a fixed current or at least a limited range? The ones I'm familiar with would show significant modulation in a 100w AB amp? Happy to see designs that would not!
Building an amp that results in fine performance is relatively easy.
Mainly, the power transformer's regulation stats should allow mimimal sag - suitable output transistors can handle that without audible issues - Reasonably speaking.
As for input stages, regulation at that stage has been, and is, effective in eliminating audible distortions.
This is just proven sensible advice backed by commercial manufacturers designs - not overblown and obsessive discussions from website blogs.
Mainly, the power transformer's regulation stats should allow mimimal sag - suitable output transistors can handle that without audible issues - Reasonably speaking.
As for input stages, regulation at that stage has been, and is, effective in eliminating audible distortions.
This is just proven sensible advice backed by commercial manufacturers designs - not overblown and obsessive discussions from website blogs.
Douglas Selfs' "Audio Power Amplifier Design Handbook" has an excellent chapter on the subject. In short, he does not recommend rail regulation in power amps as quoted here;
"The use of regulated supplies is definitely unnecessary, and I would recommend strongly against their use. At best, you have doubled the amount of high-power circuitry to be bought, built, and tested.
At worst, you could have intractable HF stability problems, peculiar slew-limiting, and some expensive device failures."
I have tried both linear regulators and simple unregulated supplies and I have to agree. Get the book, it's well worth it.
"The use of regulated supplies is definitely unnecessary, and I would recommend strongly against their use. At best, you have doubled the amount of high-power circuitry to be bought, built, and tested.
At worst, you could have intractable HF stability problems, peculiar slew-limiting, and some expensive device failures."
I have tried both linear regulators and simple unregulated supplies and I have to agree. Get the book, it's well worth it.
Most of it is understood agree also that designing a simple amplifier is far easier ( now days ) than designing a regulated psu to feed the amp with the correct current at the correct speed under all conditions .
to make it more simple
even with 600W trafo and 60K ufd for both ch there will be one point where the amplifier will modulate the PSU
there will be some other points that rails will sugg under transients especially when speaker impedance drop to some ammount
so if regulated rails solve the above two issues then yes worth looking at them
if the contribution on the problems is not of any importance then the all thing is pointless
Kindest regards
Sakis
PS
in the specific application hum and noise was never the issue .
cross talk also either way i make dual mono in the same enclosure
and to achive the gretatest if done yet again it will be dual mono ( two regulated PSU's )
to make it more simple
even with 600W trafo and 60K ufd for both ch there will be one point where the amplifier will modulate the PSU
there will be some other points that rails will sugg under transients especially when speaker impedance drop to some ammount
so if regulated rails solve the above two issues then yes worth looking at them
if the contribution on the problems is not of any importance then the all thing is pointless
Kindest regards
Sakis
PS
in the specific application hum and noise was never the issue .
cross talk also either way i make dual mono in the same enclosure
and to achive the gretatest if done yet again it will be dual mono ( two regulated PSU's )
Haven’t you omitted an important sentence from the Douglas Self’s recommendation?
“The generic amplifier designs examined in this book have excellent supply-rail rejection, and
so a simple unregulated supply is perfectly adequate.”
This omission presents his recommendation as an absolute one for all amplifiers and all conditions.
Yes, I agree that regulated PS are not always an advantage.
According to my measurements, this is absolutely unnecessary. In example, regulated PS rails modulation at 10 W/8Ω is about 70 uV.
The Acoustical Manufacturing Company (QUAD) used regulated supplies on an early transistor amplifier for SOAR protection. They never worried about SOAR protection when they still used valves, but after switching to transistors, they got complaints from the BBC and solved those with a regulated supply. Their later models have the protection circuitry included in the amplifier design and have no regulated supply.
Yes, I did omit that because the OP stated the thread is about class AB amps, which are in the book.
Good catch though, and thank you for keeping things clear.