Super Regulator, collecting the facts

[peranders]

Originally posted by janneman
I am sure that is the way to go. Not only for regulators, also in other audio-related applications. For one thing, you get low THD not just at 1kHz but out to 100kHz.
However, I would hesitate to give the details about this next revolution in audio to Per.

I'm a little bit skeptic about using 3 kV/us devices in our world of RFI but otherwise there are some very good CFB and pseudo-CFB opamps out there. A minor disadvantage would be (maybe) that many of them is optimized for 5 volts or +- 5 volts.

[peranders]

Originally posted by Elso Kwak

Faster is better? I doubt it! Your QSXPS on your site is in fact a Jung like regulator with a Darlington pass transistor and a LM431 as the reference.

If you check very closely it is not a Jung regulator, more like Sulzer, but the regulator was designed many years ago and the "invention" is mine but more people than I have got the same idea. A typcial case of parallel thinking. Not a very unique idea I'll guess.

Elso, I share your concerns about faster = better

faster = more potential problems also, no real quality enhancment

[janneman]

Originally posted by Elso Kwak


Hi Jan and John,
Faster is better? I doubt it! But the current trend is like that. I tried some pretty fast opamps like AD817. Not my piece of cake.
Per-Anders what are the real motives to start this thread? Collecting facts? Then you have only to copy articles from the net and the Audio Amateur journal. Your QSXPS on your site is in fact a Jung like regulator with a Darlington pass transistor and a LM431 as the reference.

Elso,

I don't consider the AD817 as a large step forward. No, what I mean is those opamps with essentially flat open loop gain up to 100kHz or so. They have negligeable phase shift up to 50kHz or so, so compensation can be done way outside the audio range. Then you get 20kHz performance which you have now up to 1kHz. And no, I'm not going to give you any part numbers. I need some job security...

Have a nice weekend,

Jan Didden

[Elso Kwak]

Originally posted by janneman


Elso,

I don't consider the AD817 as a large step forward. No, what I mean is those opamps with essentially flat open loop gain up to 100kHz or so. They have negligeable phase shift up to 50kHz or so, so compensation can be done way outside the audio range. Then you get 20kHz performance which you have now up to 1kHz. And no, I'm not going to give you any part numbers. I need some job security...

Have a nice weekend,

Jan Didden

Sorry Jan, I don't understand your post. Except for the AD811 current feedback opamp, the AD817 is the fastest opamp I have in house.
Open loop gain is flat to about 10k. You want still faster??? You believe you will get this working without oscillation? Chapeau.
Have a nice weekend too

[ALW]

Peranders,

Discussions here about the best etc. are rarely ever useful, different people have different opinions based on their own experiences. It matters not one jot how much experience they have as it all has a context and underlying agenda.

As one designer* used to say: -

"A system is only as good as it's weakest part"

It doesn't matter how good the best bit is, the worst bit will be the dominant factor. In my experience the process of audio design is about dealing with those 'worst' bits and then finding another layer of them underneath. As Matt Jonhson of The The sang "...and the whole godamn thing starts all over again".

With the greatest of respect the reason you cannot hear those things you mention is because of your worst bits, and only you can locate and deal with them.

I promise you you will find it enlightening, challenging and that you are unlikely to want to give away the resulting information on a forum, 'cos it will be bloody hard work for you!

Andy.

*Julian Vereker MBE

[tbla]

I can't hear any difference.

per - no offence, but what speakers do you use.....?

[peranders]

I think indeed that some discussions can bring up facts, like the one we had about symmetrical vs. unsymmetrical amps (thanks andy_c)! The conclusion this time was that distortion spectra looked different but we couldn't agree which one was the best

I will probably give away everything (except for one thing) in the future like I do now. I'll see no reason why I would change my mind.

[peranders]

Originally posted by tbla


per - no offence, but what speakers do you use.....?

I used Sennheiser HD545 reference (headphones) and Martin Logan SL3 electrostatic speakers.

Maybe there is a difference which I haven't discovered yet with the chosen music material and signal source.

[tbla]

ok - the logans can show the diferences.......

what ears do you use....hehe joke - its friday

[Christer]

and let's get back to the issue.

Thanks to a forum member I have now got hold of and read
the TAA article series on the superregulator (and other
regulators). It is very interesting reading and it seems basically
all questions asked so far in this thread were already answered
there. However, I do wonder about one thing. These articles
still use the version with AD797/AD848 and with the opamp
powered from the raw DC side. I understand that more
recent versions use AD825 as the preferred op amp and also
powers it from the regulated side. In fact, I already before had
such a schematic, which seems to be copied from a printed article.
I thus conclude there must be at least one more recent article
than those I have (parts 1 - 4). If this is a continuation of the
article series in TAA I don't know. I would thus appreciate info on
this/these article(s), and would also, if you excuse my boldness, appreciate if someone could mail it/them to me.

Thanks in advance