Why 7805 is better than LM1085?

[CraigBuckingham]

Elso, you could swap D1 and C3 and put a resistor in between to filter the noise from the voltage reference.

Originally posted by Elso Kwak
Hi All,
Attached is a further improved powersupply circuit.
The red LED and the PNP transistor provide a 2mA constant current source for the LM329. The red LED is used as a 1.6V reference providing a 1V drop across the 470 Ohm resistor in the emittor lead.
This circuit also works with four green LED's, in a series string, in place of the LM329 but sound is better with the circuit as depicted.

I also did some experiments with LT1021-5 and MAX 6250. These are expensive low-noise 5V references. I amplified the 5V to 6.9V with a OP27 and connected this to the base of Q1. Sound is definitely different but I am not sure it is better. It has a kind of inflated quality just as the sound of commercials on our national television. Also more fatigueing. I keep lowering the volume, not a good sign...

[Elso Kwak]

Originally posted by CraigBuckingham
Elso, you could swap D1 and C3 and put a resistor in between to filter the noise from the voltage reference.

Hi Craig,
Noise is not a issue.

[CraigBuckingham]

Elso,

It may not be critical for the circuit you have designed but from a point of elegance, the addition of a resistor and swapping D1 and C3 could significantly reduce the 75nV/sq.root-Hz reference noise.

I mean you have got the capacitors there in the first place, why brute force filter the reference when you can add a litle more finesse with a resistor and be more gentle to your reference.

As I said, low noise may not be your ultimate goal but it is there for the taking and may be beneficial for other designs.

Regards, Craig.

Originally posted by Elso Kwak

Hi Craig,
Noise is not a issue.

[CraigBuckingham]

Hi Terry,

Were they my Audio Amateurs that I had in your library? Interesting articles when I first read them in 90,91,92. When did you read them 95,96?

That discrete reg low open loop gain reg you built? Wasn't that the one based on my original shunt arangement which I said wouldn't sound as good but would measure better? It was the CFB or Silaki pair wasn't it.

Any how we both found my original shunt was the better one hey.

I share your frustration sometimes when I read these groups but hey we have been there and done that.

You said it was a very steep learning curve for you when we first colaborated but I find you are giving Jan more recognition than myself. Who was it that did the AD811 regs? Who did not know about capcitors when I first met them?

Can I have some recognition too please?

Thanks, Craig.

BTW, is that mountain bike ride still on ;-)

Originally posted by Terry Demol
[B]

I had studied the Jumg regs performance years ago when
they were published in Audio Amateur. I directly compared
a discrete low open loop gain reg I designed myself and
it was lower OP Z than the graphs on AA above a few hundred
kHz. I can't remember the exact measurements, it was years ago,
but they were definately MUCH wider BW than the Jung regs.
The OP Z of my regs were in the order of 0.05ohms out to
1 mHz without any OP cap in place.

WRT to flatness, it is only an indication that the reg has
very wide BW and good settling time. If you look at opamps,
the ones with best settling time are always low open loop gain,
less internal stages and very fast.


[B]

Just as I readily accepted that YOU heard a difference when
reading about listening tests in AA. What makes you right and
me and others wrong? What makes your listening tests unbiased?
I tested my regs with some NON TECHNICAL listeners who didn't
even know what feedback was and observed their opinions,
totally free of placebo. So, whereas my judgement could have
been biased, they only wanted the best sound for their DAC...
and I gave them whichever iteration they preferred... often
my reaction was "why does this sound better". Often there
are more questions than answers. At least I am honest about it!



It seems you feel most people here use open loop + trannies
for I-V due to a lack of OPA implemetational knowledge, deluding
ourselves that we have made a sonic improvement? Do you
think we are so stupid? Do you not think that many of us
have tried the venerable AD811 and countless others with
countless power supply iterations and THEN decided to move
on to more fruitful ground from shear frustration.

You may find it interesting that one of the biggest innovators in
digital, Apogee Digital, used zero feedback discrete I-V on
the DA1000. This was the one repackaged by Mark
Levinson for Cello (same I-V). Wadia also has their own style of
ZFB I-V on their ultra expensive DACs.

Then there's Mr Edmund Meitner, know him.... he also seems to
be a proponent of ZFB DAC AND ADC interface on his incredibly
highly regarded and measuring converters.

Jan, I have a lot of respect for your work and read very
carefully all the articles on the super regs. They were very well
done, represented a huge amount of work and I learnt a lot from
them. I applaud your effort!

But could you please do some of us the favour of
acknowledgment of our existence... there are also some other
people here who have much to contribute
to this so very challenging of pursuits.

Terry

[Terry Demol]

Originally posted by CraigBuckingham
Hi Terry,

Were they my Audio Amateurs that I had in your library? Interesting articles when I first read them in 90,91,92. When did you read them 95,96?

That discrete reg low open loop gain reg you built? Wasn't that the one based on my original shunt arangement which I said wouldn't sound as good but would measure better? It was the CFB or Silaki pair wasn't it.

Any how we both found my original shunt was the better one hey.

I share your frustration sometimes when I read these groups but hey we have been there and done that.

You said it was a very steep learning curve for you when we first colaborated but I find you are giving Jan more recognition than myself. Who was it that did the AD811 regs? Who did not know about capcitors when I first met them?

Can I have some recognition too please?

Well I'll give you recognition for political/diplomatic incorrectness!
You sound like my 6 YO son.

Look, let's take this off forum and do everyone here a big
favour. The idea is to keep personal attacks/details low and
information up.

BTW, is that mountain bike ride still on ;-)

MTB rides are always on! But you'll have to a) wait till
I finish building a 2nd bike or b) buy one for yourself.
Either way, after this post, don't expect an easy one.

Terry

[mlloyd1]

Yeah, we had a short chat about the IC here

Oops, I just realized I never scanned the data sheet. It was indeed writtenup in EWW though. I don't remember the issue.

mlloyd1

Originally posted by jam
Elso,

I believe that Wadia uses a circuit known as a current conveyor and it is manufactured for them as an IC. Many years ago there was an article about current conveyors in Wireless World I will have to try to look for it. Meridian might have used the same topology too.

Jam

[CraigBuckingham]

Originally posted by jean-paul
Hello Per,

Snipped all the previous stuff

Controversial thing: in digital OSCON behind the regulator is very good. In analog they sound worse than normal electrolytics.
Does somebody experience the same with OSCON ?

jean-paul,

I agree with your assessment of the Sanyo Oscon for analog.

I can't stand the things but I have friends that swear by them. I cannot believe they like them as we almost always agree on all our other observations.

Call me disillusioned.

Craig.

[patwen]

thanks for all the help, I was busy and unable to post my final findings untill now.

I have gone back to the 7805, apperently (enlightened by T), in a digital design, it is better to filter out all the high freq noise (using caps and inductors) instead of trying to regulatred them using a ultra fast wide band width regulator.

However, even after extensive mod, my Rotel 991 is still not as good as my old Arcam Alpha 5 (TDA1541A) in NOS mode. Also I have made a sin(x)/x compensation filter which improve the sound of the NOS TDA1541A quite a bit. Details in the Sin(x)/x post.

[Elso Kwak]

Originally posted by Elso Kwak

Hi,
I just got a email pointing me at a stupid drawing error. As the shaded part of the electrolytics is the minus side these have to be connected to ground of course.
I apologise.

Attached (hopefully) the corrected schematic.

[PMiczek]

Was there something earlier in this thread about additional caps in parallel with C10, or am I confused? What kind of regulation do you expect with this circuit as shown?

And, thank you very much for posting it.

PM