| dipchip |
Hi everyone,
I have completed the layout for a 3-terminal Super Audio Regulator. (I Hope!)
The design was very generously given to me by ALW.
from ALW Audio
I modified his design slightly to make it fit into a 3 terminal
78xx and 79xx footprint. (But I still have the original option of
the sense pins, if you hardwire them)
This voltage regulator has a normal linear pre-regulator that could be removed if needed, followed by a Walt Jung super low noise voltage regulator.
Dale (harvardian) and I, are going to try to offer this board (and a negative version) as a kit.
I named it the APOX-JR+ and APOX-JR-
I posted pictures of the layouts, so I would really like some feedback before I finalize the board layouts.
You can see the designhere
Is anyone interested in a board like this?
Thanks,
Craig Beiferman |
|
|
| Fred Dieckmann |
Has anybody asked Walt Jung about using his design (and it sure looks like his design to me) for commercial purposes? A magazine is selling the boards for his regulator with his permission if i am not mistaken.
I know you have the mounting foot print but, do you really think that thing is going to fit in someone equipment as an after market mod? ITS HUGE
You might want to see what the competition is doing.......... |
|
|
| Fred Dieckmann |
A high current design from the same vendor. I even think those are Black Gate caps in fact.........
As you said on your website "Dimensions: Not ready yet"
I am afraid I will have to agree. |
|
|
| PMiczek |
| Fred, can post what vendor, or a link to a datasheet if there is one? Thanks. |
|
|
| peranders |
| quote: | Originally posted by Fred Dieckmann
Has anybody asked Walt Jung about using his design (and it sure looks like his design to me) for commercial purposes? A magazine is selling the boards for his regulator with his permission if i am not mistaken.
I know you have the mounting foot print but, do you really think that thing is going to fit in someone equipment as an after market mod? ITS HUGE
You might want to see what the competition is doing.......... |
Fred, do you have any info about the picture you have attached (small SMD regulator pcb)? |
|
|
| dipchip |
| quote: | | Has anybody asked Walt Jung about using his design |
Fred,
I guess I was assuming it was O.K. to use the design because it was posted on EDN
under the title DESIGN IDEAS.
He also wrote the article under the Analog Devices name.
So I am assuming his intention is to get people to use the analog devices chip (AD797) in the article, which I will do.
I'm also guessing if you asked Analog Devices who owned the design, they wouldn't say Walt.
Its only 1.75" wide by 1.5" tall. I don't think thats very big.
I blew up the layout for readability.
Thanks for your comments,
Craig Beiferman |
|
|
| janneman |
| quote: | Originally posted by Fred Dieckmann
Has anybody asked Walt Jung about using his design (and it sure looks like his design to me) for commercial purposes? A magazine is selling the boards for his regulator with his permission if i am not mistaken.
I know you have the mounting foot print but, do you really think that thing is going to fit in someone equipment as an after market mod? ITS HUGE
You might want to see what the competition is doing.......... |
Fred,
The " Jung" regulators come from a 4-part article in Audio Amateur magazine (now AudioXpress) 1-4 1995 by Walt, the undersigned and Gary Galo. It was based on an earlier article by Mike Sulzer in Audio Amatear 2/80, and my own article (A wideband power supply) in AA 1/87. The PCB was designed by me and is being sold by Old Colony Sound Lab (an affiliation of AudioXpress).
So, even if the PCB design is new, the schematic is the result of a colloborative effort of several people. I personally have no problem with this, but Walt might see it differently. I don't know how to reach him though (except through AudioXpress) as he seems to have retired from Analog Devices.
Jan Didden |
|
|
| ALW |
| quote: | | Has anybody asked Walt Jung about using his design (and it sure looks like his design to me) for commercial purposes? A magazine is selling the boards for his regulator with his permission if i am not mistaken. |
audioXpress sell PCB's, as far as I'm aware Walt does not benefit from this.
Walt is well aware of the other's using what I suspect is his basic design and from my own questions he has no problem with it being marketed commercially by others.
I know, I'm doing it and have asked after the guilt got to me ;)
I'm not sure the Audiocom reg's have the full Jung cct, but they are a wideband op-amp based unit. The tracking pre-reg's and other mods (bootstrapping) bring benefits that the commercial units are not implementing.
Some are still using the AD797 :whazzat:
Companies such as Trichord are also using similar schemes.
Andy.
P.S. Jan - Walt has retired from ADI |
|
|
| peranders |
If we exclude the opamp itself (very low noise type), what is so special really? What about the output stage? If a regular emitter follower is used, what enhancements do I get with the "Jung" output stage?
My low noise (16 µV) QSXPS PS is the same except for that the opamp is fed before the regulator with a huge capacitor (compared to the current consumption). If I change the opamp I will proably get lower noise. |
|
|
| ALW |
| quote: | | If we exclude the opamp itself (very low noise type), what is so special really? What about the output stage? If a regular emitter follower is used, what enhancements do I get with the "Jung" output stage? |
As always with audio, the final circuit sound has little to do with the topology used.
The benefit of the o/p stage is primarily that it allows a start-up mechanism for the bootstrapped regulator, and the subsequent (big) benefits of that.
Andy. |
|
|
| peranders |
| quote: | Originally posted by janneman
I don't know how to reach him though (except through AudioXpress) as he seems to have retired from Analog Devices. |
I have had contact with Mr. Jung recently (April) about FNDR filter. |
|
|
| janneman |
Well, ALW will probably beat me to this (Hi Andy!) but believe me, in this designb EVERY (almost) component is tweaked, the reference, the opamp, the caps, the feedback circuitry, and yes, the bootstrapping of the opamp supply. Each giving measurable improvements and, as I have heard from other users, audible differences.
On top of that, the PCB layout has a MAJOR impact on the performance. A couple of wrongly routed traces can negate that expensive opamp or cap completely.
This is not for the faint at heart. If you see designs still using the 797 they either don't use remote sensing or use excess compensation to keep it stable or take a gamble.
Do yourself a favor and buy the 1995 backissue series of AudioXpress. Well worth the 20 $ or so it costs.
Jan Didden |
|
|
| Norbert |
Hi,
the later called Jung Design has been used by Akihiko Kaneda in the seventies already within all of his DC-Amplifiers published at that time using 709 opamps.
Best regards,
Norbert |
|
|
| janneman |
| quote: | Originally posted by Norbert
Hi,
the later called Jung Design has been used by Akihiko Kaneda in the seventies already within all of his DC-Amplifiers published at that time using 709 opamps.
Best regards,
Norbert |
" There is nothing new under the sun, but there is a lot we haven't seen yet" -- Anon
Jan Didden |
|
|
| peranders |
| quote: | Originally posted by ALW
As always with audio, the final circuit sound has little to do with the topology used.
The benefit of the o/p stage is primarily that it allows a start-up mechanism for the bootstrapped regulator, and the subsequent (big) benefits of that |
Oki doki, I had some startup problems with my negative regulator but I suspect with a true rail-to-rail opamp would solve the problem. The opamp I used LM324 can have the inputs at ground but not at the power supply voltage. |
|
|
| dipchip |
peranders,
The real benefit comes from the bootstrapping the op-amp.
This means that the op-amp is powered directly from the regulated output voltage of the circuit.
Normally, if the op-amp were powered from the supply, any ripple on the op-amp supply would show up as noise on the output.
But now, the regulated output powers the op-amp,
and the regulated output is also the feedback to the op-amp.
So the noise is now much more common mode, and hence the
big 30 to 35 dB improvement over typical regulators. :yes:
Thanks,
Craig Beiferman |
|
|
| ALW |
Hi Jan,
I've been corresponding with Craig privately about those issues (e.g. PCB layout).
With uV error signals, the PCB is an essential part of the circuit!
I don't mine helping for the greater good of the DIY community though, and have wanted to do an SMD version ever since I did my board :)
I reckon there's a good market for a 3-terminal replacement that crams as much performance as we can muster into a suitable footprint.
Andy. |
|
|
| janneman |
| quote: | Originally posted by peranders
Oki doki, I had some startup problems with my negative regulator but I suspect with a true rail-to-rail opamp would solve the problem. The opamp I used LM324 can have the inputs at ground but not at the power supply voltage. |
It's not that Per. If the opamp supply is bootstrapped, it can not regulate the output voltage away from zero, because there is no supply for the opamp without the output voltage. It's a matter of the chicken and the egg, who gets there first.
Jan Didden |
|
|
| peranders |
| quote: | Originally posted by janneman
Well, ALW will probably beat me to this (Hi Andy!) but believe me, in this designb EVERY (almost) component is tweaked, the reference, the opamp, the caps, the feedback circuitry, and yes, the bootstrapping of the opamp supply. Each giving measurable improvements and, as I have heard from other users, audible differences.
On top of that, the PCB layout has a MAJOR impact on the performance. A couple of wrongly routed traces can negate that expensive opamp or cap completely. |
I think the Jung design is pretty much normal engineering science, no tweaking at all. There are better references for instance.
Also the pcb layout, normal Kirchoffs laws, and normal science, no voodoo like at >GHz. |
|
|
| dipchip |
This is what the board size really looks like.
(at least on my 1280x1024 17" monitor) |
|
|
| ALW |
A couple of points, whilst it is basic engineering science, it's also an audio product, hence listening comes in somewhere along the line ;)
I'd be curious as to what references you feel are better, the LM329 is inexpensive and has low noise relative to it's output voltage.
This allows less noise gain within the system, resulting in less noise in the system as a whole.
Aboslute accuracy of the reference is irrelevant here as is long-term stability.
There's few, if any, bandgap devices that would beat it, sonically, and not many sub-surface zener alternatives.
And there is LOADS of tweaking in there.
But to test your authoritative statement, please tell us of the SONIC effects of altering various circuit parts :)
Andy. |
|
|
| peranders |
| quote: | Originally posted by ALW
There's few, if any, bandgap devices that would beat it, sonically, and not many sub-surface zener alternatives. |
This design can afford sufficient LP-filtering so the noise is less important I would say. You could use almost anything but if you required low tempco, low drift, etc you have to look at the reference itself. I still think most of the design is normal, nothing weird and not very tweaked in the sence in being weird. |
|
|
| ALW |
| quote: | | This design can afford sufficient LP-filtering so the noise is less important I would say. You could use almost anything but if you required low tempco, low drift, etc you have to look at the reference itself. I still think most of the design is normal, nothing weird and not very tweaked in the sence in being weird. |
Well, here's a few questions to ponder: -
1. Why is the divider impedance around the op-amp still so low, when on the AD825 FET-input op-amp current noise would dominate (and hence imply a higher impedance)?
2. What are the disadvantages of trying to filter a noisy reference sufficiently, what effect does this have sonically, with the circuit as-is?
There's more, but it's easy to make the mistake that because a circuit looks simple, it is simple.
Measurement, Kirchoff or anything else will not answer those questions for you, but your ears will.
Andy (playing Fred's agent-provocatuer game) ;) |
|
|
| peranders |
1 Jung used the AD797 and to get low noise he used low resistor values. A basic rule is also to have as low values as possible so noise will be low and stray caps don't interfere too much. 1 kohms (if I have seen it right) is not very low, rather low.
2 No disadvantage at all! Possibly slow startup.
The reference Jung used is pretty much average, like the 431 but 6 dB lower noise. I think also Jung's text is straight forward, I agree to everything. |
|
|
| Fred Dieckmann |
"The real benefit comes from the bootstrapping the op-amp."
Well................
This is really only true at frequencies where the PSRR of the op amp is high. As the PSRR drops with increasing frequency some issues arise. Here lies the paradox. The whole point of a very low impedance supply is to keep the voltage at the supply pins free from modulation. In this regulator the op amp supply is intimately coupled to the op amps (via two followers) output currents modulation of its own supply.
That is the very antithesis of the principle reducing the modulation of supply by load currents that regulation is designed to do. An additional RC filter for the op amp in the regulator seems prudent to me. The supply AC currents to the op amp would seem to be small enough (the AC load current demands of the regulator divided the Hfe of the two followers) to make the supply for the op amps increased impedance less important than the very significant improvement in the PSSR for the op amp.
The PSRR issue becomes very interesting at the point where the inductive output impedance of the regulator resonates with output capacitor and extremely interesting with a high Q low ESR capacitor! The very low inductance with a high gain bandwidth op amp pushes this to a high frequency where the PSRR for the op amp is poor.
This is more than just an academic exercise and is what seems to be a real life issue with the topology. The behavior of audio circuits at RF frequencies can have real implications for the sonics. This has been an area of some discussion on the forum in the area of diode bridge switching noise on audio circuits.
There is no such thing as a free lunch.
"2. What are the disadvantages of trying to filter a noisy reference sufficiently, what effect does this have sonically, with the circuit as-is?"
Leakage current in electrolytic capacitors can contribute to noise and voltage drift of the reference voltage seen by the op amp.
http://www.analog.com/UploadedFiles...ation_V-Ref.pdf page 18
I will now sit and wait for the fur to fly in the ensuing cat fight.....
* Andy is quite sharp but is too modest to call himself an expert. |
|
|
| Fred Dieckmann |
"Its only 1.75" wide by 1.5" tall. I don't think thats very big.
I blew up the layout for readability.
Thanks for your comments,
Craig Beiferman"
ummmmm...... How high do the components stand off the board?
Pretty tight fit for even brand X
Yes Peranders I do have further information on the regulators pictured. Why give them free advertising here?
Meeeyowwww! |
|
|
| Elso Kwak |
| quote: | Originally posted by dipchip
peranders,
The real benefit comes from the bootstrapping the op-amp.
This means that the op-amp is powered directly from the regulated output voltage of the circuit.
Normally, if the op-amp were powered from the supply, any ripple on the op-amp supply would show up as noise on the output.
But now, the regulated output powers the op-amp,
and the regulated output is also the feedback to the op-amp.
So the noise is now much more common mode, and hence the
big 30 to 35 dB improvement over typical regulators. :yes:
Thanks,
Craig Beiferman |
Hi Craig,
Sonically I found no advantage in powering the opamps from the regulated supply i.e. the output of the regulators.
RC filters on the supplies of the opamps as also suggested by Walt Jung seriously diminished the bass slam of the supply. This is done in the Mark Levinson PLS-150 powersupply for the ML-1 preamplifier with detrimental effect. A LM317/337 combo sounded better!
This one is interesting I think..........
http://headwize2.powerpill.org/proj...ilmore3_prj.htm
Though I found sound is better without preregulators......
:yummy: |
|
|
| Fred Dieckmann |
"RC filters on the supplies of the opamps as also suggested by Walt Jung seriously diminished the bass slam of the supply.'
bigger cap, smaller resistor....... Actually the bootstrap gives extremely good input noise rejection at low frequencies. I think line noise rejection may be more important than output impedance. It may be the reason for the popularity of shunt regulators which also can be bootstraped. |
|
|
| fdegrove |
Hi,
| quote: | | I think line noise rejection may be more important than output impedance. It may be the reason for the popularity of shunt regulators which also can be bootstraped. |
Series regs do even better in that department...
Not surprisingly, really.
Cheers,;) |
|
|
| stokessd |
| quote: | Originally posted by janneman
Do yourself a favor and buy the 1995 backissue series of AudioXpress. Well worth the 20 $ or so it costs.
|
That must have been an outstanding year, then.
I buy issues from the news-stand a couple times each year when I am traveling. I have yet to be impressed by that magazine on an issue to issue basis. There's nuggets of greatness (like the Zen, Thor etc) but then there is things like low voltage biased tube headphone amps designed for lots of distortion, silly Macintosh amp output tube modification plates when simple socket rewiring would be better in every way, and on and on. Then there's articles like the moving coil transformer project which entails putting a transformer in a plastic case with RCA jacks (not even balanced inputs for god's sake!).
Then there's a project with some meat, and instead of pushing the marginal crap out of a particular issue, they spread the article over several issues.
$20 for all the back issues strikes me as "well worth it"
Sheldon |
|
|
| fdegrove |
Hi,
| quote: | | I buy issues from the news-stand a couple times each year when I am traveling. |
Maybe you should travel more often?
| quote: | | Then there's articles like the moving coil transformer project which entails putting a transformer in a plastic case with RCA jacks (not even balanced inputs for god's sake!). |
Then there's people's intelligence being able to sift the crap from the nuggets?
Come on S there's only one bible and that must be your own, right?
Cheers,;) |
|
|
| randytsuch |
Free advertising
http://www.audiocominternational.co...rregulators.asp
if you want to see the regulators Fred was referring to.
They also make a q power regulator that is even more expensive.
And I for one would be interested in a reasonably priced PWB or kit.
Randy |
|
|
| dipchip |
"Uh oh........ not even experts this time"
Ouch that hurt!
--------------------------------------------------------------------------
"The real benefit comes from the bootstrapping the op-amp."
Well................
This is really only true at frequencies where the PSRR of the op amp is high. As the PSRR drops with increasing frequency some issues arise.
Yes, thank goodness the PSSR of the op-amp is so high in the audio range that
we are interested in.
---------------------------------------------------------------------------------
"Here lies the paradox. The whole point of a very low impedance supply is to keep the voltage at the supply pins free from modulation. In this regulator the op amp supply is intimately coupled to the op amps (via two followers) output currents modulation of its own supply."
Its not a paradox, its a benefit. Whats better. The ripple before the regulator's pass stage
or after. The inherent PSRR of the op-amp insures that the ripple after the pass stage will be less than before. The op-amps inputs
are also filtered insuring that the op-amp is not producing any oscillations, thereby not directly contributing self induced modulation.
-------------------------------------------------------------------------------
"An additional RC filter for the op amp in the regulator seems prudent to me. "
There is a RC filter already in the feedback return path.
But if a lowpass filter was added to the op-amps supply.
The R in series with the op-amps supply lines would limit slew rate
and probably decrease performance.
-Craig Beiferman (non-expert)
P.S. Fred I'm a dog person. |
|
|
| janneman |
| quote: | Originally posted by peranders
I think the Jung design is pretty much normal engineering science, no tweaking at all. There are better references for instance.
Also the pcb layout, normal Kirchoffs laws, and normal science, no voodoo like at >GHz. |
Of course it is normal engineering, not voodoo. The trick is to get it all together in the optimum way. Surprisingly, the people that can do that seem to be just a very small proportion of those who know kirchoff and ohms law, even a very small proportion of all those with engineering degrees.
Could you give us the better references?
Jan Didden |
|
|
| Elso Kwak |
| quote: | Originally posted by janneman
Of course it is normal engineering, not voodoo. The trick is to get it all together in the optimum way. Surprisingly, the people that can do that seem to be just a very small proportion of those who know kirchoff and ohms law, even a very small proportion of all those with engineering degrees.
Could you give us the better references?
Jan Didden |
Hi Jan,
It surprises me Per Anders is suggesting better references while he is using the LM431 is his "Ultra Low Noise Power Suppy" QSXPS.
"Better" i.e. lower noise references would be f.a. AD586, LT1021-5, Ref02, MAX6250. Signigficantly lower noise are only the AD586 and the MAX6250 with 1µF noise reduction cap applied.
I have experimented a bit with these references compared to the LM329 fed by a constant current source comprised of a red LED and a transistor. The difference in SOUND is very small I concluded, except for the MAX6250 that seems to impose its own sonic footprint on the sound. Very weird.
As Fred remarked in a earlier discussion about these type of regulators the LM329 is easier to apply.
(You can not flipover these IC type references like the LM329 for the negative supply). But a way around is the approach by Kevin Gilmore using the negative regulator as a 1x inverting amplifier of the positive regulated supply. Is that a good thing to do?
:confused: |
|
|
| fmak |
[QUOTE]Originally posted by Fred Dieckmann
[B]"Its only 1.75" wide by 1.5" tall. I don't think thats very big.
I blew up the layout for readability.
-----------------------------------
Wow someone spent a bomb on those BGNs.
What is the rationale for using BGNs and not the lower tan delta FKs or SG OSCONs in the digital section? The OSCONs are much smaller and by selection one can find batches that are an order of magnitude better than the BGNs. I understand SOUND was the criteria.
Also in the resistors used in the same mod by Audiocom, two chip Vishays have been selected for sonics. I would prefer the single chip S102Ks. The former are of course, even more expensive than the single chip one.
Retailing expertise?? |
|
|
| peranders |
| quote: | Originally posted by Elso Kwak
Hi Jan,
It surprises me Per Anders is suggesting better references while he is using the LM431 is his "Ultra Low Noise Power Suppy" QSXPS.
"Better" i.e. lower noise references would be f.a. AD586, LT1021-5, Ref02, MAX6250. Signigficantly lower noise are only the AD586 and the MAX6250 with 1µF noise reduction cap applied.
I have experimented a bit with these references compared to the LM329 fed by a constant current source comprised of a red LED and a transistor. The difference in SOUND is very small I concluded, except for the MAX6250 that seems to impose its own sonic footprint on the sound. Very weird.
As Fred remarked in a earlier discussion about these type of regulators the LM329 is easier to apply.
(You can not flipover these IC type references like the LM329 for the negative supply). But a way around is the approach by Kevin Gilmore using the negative regulator as a 1x inverting amplifier of the positive regulated supply. Is that a good thing to do?
:confused: |
I think it's important to have low noise voltage references if you can't use heavy filtering (DAC's, ADC's). In audio it's cool to use a really expensive voltage reference but my design goal was to use normal parts, easy to get and I think 431 is a really good reference for the cost. In my case I have used a second order LP-filter and if you compare Jung's text you see that a LP-filter does wonders when we talk noise. |
|
|
| janneman |
Elso, Per,
First of all I don't like to use an inverted version of a pos supply for a neg supply. What you are doing is copying noise and ripple from one supply to the other. Especially ripple, which is correlated to the signal, should be kept out as much as possible. I admit that this is not based on exhaustive testing, just my gut feeling.
I have to disagree with Per. In isolation, you CAN say that the filter does wonders. But don't forget that the end result is the outcome of a lot of related factors. Surely you agree that using both a low noise reference AND the noise filter give a better result than a mediocre reference and a filter? The same goes for the opamp. I think I read you still use the LM324 Per, really, you're doing yourself a big disserve. This opamp can't come close to a modern wideband one in terms of wideband ripple and input noise rejection, wideband very low output impedance etc. And please, don't tell me "but it sounds better". This is an engineering discussion.
Jan Didden |
|
|
| peranders |
| quote: | Originally posted by janneman
I think I read you still use the LM324 Per, really, you're doing yourself a big disserve. This opamp can't come close to a modern wideband one in terms of wideband ripple and input noise rejection, wideband very low output impedance etc. And please, don't tell me "but it sounds better". This is an engineering discussion. |
My design idea was to use a slow opamp and creating a slow regulator because it was intended for rather constant loads. Noise isn't a big thing when opamps are used as ampiflying elements but if we talk MC, MM phono amp with discrete design it more important, very much more.
LM324 isn't so bad if they are used right but of course you can choose other types. I don't disagree on that.
16 uV noise isn't so bad (if you compare to a 78xx) but this noise is mostly generated by LM324 so it's easy to lowering the noise simply to change opamp. |
|
|
| peranders |
I'm pretty amazed by this discussion compared to this .
78xx is quite alright as long as the right brand is chosen and here is 431 a peice of junk. I'm really :confused: :scratch: .
If you have no demands of speed I think regulator choice is rather unimportant as long as you can filter out noise and the load is rather constant och the application doesn't need extra stable voltage. |
|
|
| ALW |
| quote: | | I think regulator choice is rather unimportant as long as you can filter out noise and the load is rather constant och the application doesn't need extra stable voltage. |
I think that power supplies make more difference to the sonic perfomance of a circuit than the basic topology does.
Controversial eh!
Andy (No engineering degree) |
|
|
| peranders |
A poor amp + an extreme PS creates good sound :no:
A poor amp is poor. :nod: |
|
|
| dipchip |
peranders,
fill in the blank
great amp + bad power supply = ___________
Try to use a high quality DAC with a crappy power supply.
Thats fun! :yes:
-Craig |
|
|
| stokessd |
One thing that's not being said, is the importance of power supply quality on the sound is variable depending on the circuit used. PSRR figures in highly to the sound quality vs. power supply quality debate.
For example look at the gainclones, people are building 50 watt power amps with 1000 uF filter caps. they can get away witht his because of the high PSRR of those chips.
I would argue that if you have a 90 dB PSRR, then any regulated power supply should sound the same. However many circuits (even op-amps and those power gaincard type chips) don't have that high PSRR at all audio frequencies.
Just my view from the peanut gallery...
Sheldon |
|
|
| ALW |
| quote: | | I would argue that if you have a 90 dB PSRR, then any regulated power supply should sound the same. However many circuits (even op-amps and those power gaincard type chips) don't have that high PSRR at all audio frequencies. |
I'd be inclined to agree, in principle.
The problem is I've never seen a circuit with that PSRR across the audio band, taking into account circuit noise gain etc.
Peranders,
You're right, but my point is any competent amp circuit (not obviously flawed designs) can be made to sound good, if you address the details.
Topology is almost irrelevant, but PSU's bring big benefits to everything I've tried.
Andy. |
|
|
| PMiczek |
Given the above comments on power supply quality and regulators, what regulator specifications and ps "quality" are sufficient? Meaning in the type of small, low voltage digital and +/- analog supplies controlled with 3-pin regulators, and used in DACs and related digital/analog audio designs, as opposed to power amps.
It is one thing to say that a component is not good enough, or that another is "better", but what noise and regulation specs are actually required for a supply which will not add noise and distortion to the output? |
|
|
| Fred Dieckmann |
"(You can not flipover these IC type references like the LM329 for the negative supply)."
Actually you could run a positive series reference for a negative reference voltage. Tie Vout to gnd and power Vin from the positive supply. You put a current on the ground terminal and V-out (the ground terminal) is the negative output reference. You are running the regulator series and shunt. Vin could be the regulated positive supply for best effective PSRR for the negative reference. Why bother? There seem to be more good series references than shunt. I seem to remember a few that can be used in either mode.
With seperate transformer secondaries you can tie two positive regulators together, for one to act as a negative regulator (V out of one tied to ground of the other) since the input voltages are isolated by the seperate transformer secondaries not be tied together the GND / Vout node is the only galvanic connection.
I have built a lot of line level circuits with ground as the most positive voltage. Scott Nixon used to joke that he had to turn the schematics upside down to figure out how they work. The circuits had ground for the supply and the load in parallel. Ground looks like the positive supply and the gound return. It is a shunt amplifier much the same priciple as a shunt regulator.
You guys in Australia swap all positive and negative descriptions
since you guys are upside down you realize........ |
|
|
| Fred Dieckmann |
"It is one thing to say that a component is not good enough, or that another is "better", but what noise and regulation specs are actually required for a supply which will not add noise and distortion to the output?"
You have heard the joke about the guy who ask how much something cost and the salesman saying how much you got?
It dependent on how clean your input voltage to the regulator is,
the PSRR and load currents of the circuit whos supply teminals are being regulated, and whether the circuit is anglog, digital, or vegtable. Noise, PSRR, and output impedance of the regulator vary dramatically with frequency.
And I'll bet you were afraid you would get a complicated answer like "It depends." |
|
|
| ALW |
...but have you asked the right question ;)
Rather than "what regulator specifications and ps "quality" are sufficient" how about "what regulator specifications and ps "quality" are relevant"
Aren't we a helpful bunch?
Bit like the guy asking the Irishman for directions "Well I wouldn't start from here, if I were you"
Andy (with a degree of fun). |
|
|
| Elso Kwak |
| quote: | Originally posted by Fred Dieckmann
"(You can not flipover these IC type references like the LM329 for the negative supply)."
Actually you could run a positive series reference for a negative reference voltage. Tie Vout to gnd and power Vin from the positive supply. You put a current on the ground terminal and V-out (the ground terminal) is the negative output reference. You are running the regulator series and shunt. Vin could be the regulated positive supply for best effective PSRR for the negative reference. Why bother? There seem to be more good series references than shunt. I seem to remember a few that can be used in either mode.
With seperate transformer secondaries you can tie two positive regulators together, for one to act as a negative regulator (V out of one tied to ground of the other) since the input voltages are isolated by the seperate transformer secondaries not be tied together the GND / Vout node is the only galvanic connection.
I have built a lot of line level circuits with ground as the most positive voltage. Scott Nixon used to joke that he had to turn the schematics upside down to figure out how they work. The circuits had ground for the supply and the load in parallel. Ground looks like the positive supply and the gound return. It is a shunt amplifier much the same priciple as a shunt regulator.
You guys in Australia swap all positive and negative descriptions
since you guys are upside down you realize........ |
Yes Fred you are right. But it were you own words that it was more complicated, you remember? But actually I thought the 8-pin references are so expensive I was better out using one for all regulators.
To pick up the old thread I use the low-pass filter circuit as in the OP176 datasheet and the negative reference i.e. -5V is obtained by a inverting opamp as in the Ref02 datasheet. 10k resistors and 5k to ground. This works well but not much better than LM329's.
Everyone:
How about a great supply + a great amplifier circuit?
:bigeyes:
The Kevin Gilmore circuit contains more "subtilities" but I leave these to discover by the sharpeyed reader themselve. In other words no more hints. This sentence courtesy of Jocko!:clown: |
|
|
| ALW |
| quote: | | and whether the circuit is anglog, digital, or vegtable. |
Hmm...
I wonder what those lemons I used to light a bulb in my school days sound like.
Could be a whole new market - 'Edible Audio, Music exactly to your taste'
No I'm not drunk ;)
Andy. |
|
|
| Kuei Yang Wang |
Koinichiwa,
Now I'll probably be appearing "nasty", but I fail generally to see the attraction of the Jung Reg. It is marginally stable, overly complex and has at best a moderate behaviour when faced with transients.
I would like to first point out that the "sound" of a PSU is very rarely in the DC behaviour of the supply OR in the noise. I find the "sound" of a PSU to be almost entierly determined by it's AC behaviour.
If you use 78XX/79XX regs and their fairly high noise and poor ripple rejection is no issue (it would be for a clock oscillator or a DAC chip with very poor PSRR, in most other circuits it is not) you can simply stick a really large value low impedance electrolytic on their output and I find most 78XX/79XX (in fact all I tried) to sound very similar with a few 1,000uF after the Reg. Most PCB's have enough space.
If you want something simple that actually dominates the AC impedance of the supply you could use for currents up to around 100mA the LM6171 as simple follower with a suitable reference on the input. This way the regulator will not only source current, but if required it will also sink it.
The EIN of the LM6171 is around 4uV over 100KHz bandwidth, the PSRR is > 80db below 10KHz. By adding a suitable Transistor as "Gyrator" in the positive or negative supply line the PSU line rejection can be increased far past the 80db possible with the OPA alone, I'd think 100db+ of ripple rejection should be possible in this case and the dropout Voltage would be minimised, this would likely approach the ripple rejection of something like the Jung Reg at much less effort.
The resulting "regulator" will be mostly limied by the Noise from reference for noise and the "speed" of the regulator will be very high.
With a Reference, J-Fet CCS to supply the reference, a suitable RC circuit to kill noise on the reference, two resistors to set the output voltage and one capacitor to keep AC gain at unity plus one transistor, one resistor and one capacitor we have a circuit which, relised mostly in SMD can rival the little regs from Audiocom for size (and at likley better performance).
Total part count:
1 DIL-8 IC
4 Resistor (SMD)
1 J-Fet CCS (SMD)
1 Reference (SMD)
1 200mA BJT (SMD) [darlington would be good]
3 Capacitors - several uF (Os-Con, BG whatever)
I'd suggest a layout that would place the DIL 8 IC such that it can be fitted against normal heatsinks, with the reference circutry above it.
For higher dissipation/current there are TO220 encased Power Op-Amp's which could be placed inside the feedback loop of another Op-Amp to work as "power" output stage with the first Op-Amp dominating the AC behaviour.
Anyway, some thoughts on the "78XX/79XX replacement" project. By cutting out the series pass transistor the whole regualtor becomes much more capable within the dissipation limit of the Op-Amp.
Sayonara |
|
|
| ALW |
| quote: | | Now I'll probably be appearing "nasty", but I fail generally to see the attraction of the Jung Reg. |
Not nasty at all, but have you built one?
You may then see the appeal.
Andy.
BTW, the non-AD797 ones were not marginally stable, just zone stable as is every voltage reg. |
|
|
| Elso Kwak |
| quote: | Originally posted by ALW
...but have you asked the right question ;)
Rather than "what regulator specifications and ps "quality" are sufficient" how about "what regulator specifications and ps "quality" are relevant"
Aren't we a helpful bunch?
Bit like the guy asking the Irishman for directions "Well I wouldn't start from here, if I were you"
Andy (with a degree of fun). |
Hi Andy,
The answer is extremely simple:
I build these regulators for AUDIO so the SOUND quality is what only matters to me.
I could build a sofisticated circuit wit 100 parallelled transistors kept at liquid helium temperature and claim some extreme noise spec. but will this approach sound better??? I guess not!
:confused: :confused: :confused:
I try all kinds of circuits and try to learn from the practical results rather than copying a Jung regulator. I realise I have come a long way but will NEVER reach my goal! That's life!:bawling: |
|
|
| fmak |
I build these regulators for SOUND quality is what only matters to me.
---------------------------------------------------------
Problem is: what is good sound quality; it's system, room and hearing dependent. There is a well known loudspeaker reviwer who has 3-4 cm of hair covering his ears. No wonder he likes Naim equipment!
In another Forum, there are people who claim that their clock is the only and that that those who don't use it are 'idiots' and years behind time. Again, the only evidence is that it sounds best to them.
How does one adjudicate? |
|
|
| Elso Kwak |
| quote: | Originally posted by fmak
I build these regulators for SOUND quality is what only matters to me.
---------------------------------------------------------
Problem is: what is good sound quality; it's system, room and hearing dependent. There is a well known loudspeaker reviwer who has 3-4 cm of hair covering his ears. No wonder he likes Naim equipment!
In another Forum, there are people who claim that their clock is the only and that that those who don't use it are 'idiots' and years behind time. Again, the only evidence is that it sounds best to them.
How does one adjudicate? |
Hi Fred Mak,
Yes that's the problem, as said in a application note by Burr-Brown: "soundquality is subjective"
(can not be measured by objective criteria)
I have to live by that too!
No problem for me!
BTW I have very short hair now but very long hair 25 years ago. Does it really matter?
:bawling:
Fmak If you like the KWAK-CLOCK or not does not bother me at all just as if you think I am a quirk or a quack.......:clown: |
|
|
| Fred Dieckmann |
"If you use 78XX/79XX regs and their fairly high noise and poor ripple rejection is no issue (it would be for a clock oscillator or a DAC chip with very poor PSRR, in most other circuits it is not) you can simply stick a really large value low impedance electrolytic on their output and I find most 78XX/79XX (in fact all I tried) to sound very similar with a few 1,000uF after the Reg. Most PCB's have enough space."
This whole post is the funniest thing I have read this week.......
Did you know audio supply design was so simple Andy? Boy have we wasted some time. I am so embarrased.
"I build these regulators for SOUND quality is what only matters to me."
No problem with that but some of kind of want to know why circuits sound different and how to design for that goal. Good engineering does not often negate sound quality. Some people absolutey refuse to believe this. They usually seem to be people that are not engineers for some mysterious reason. |
|
|
| PMiczek |
I realize that asking that question in this forum is only a little less dangerous than if I were to bring up 5-pin regulators.
Fred: Yes, I have heard the salesman joke. Have you ever heard that to a man with a good hammer every problem looks like a nail? The answer to that one is that if you are the type of person who routinely orders a bottle of wine which costs more than the meal, I probably will not like your advice.
ALW: As for the answer being 42, I would hope that this group had not yet reached the stage where the only interesting question is, "where do we go for lunch?"
My question was about linear supplies using a 3-pin regulator, or a reasonably facsimily thereof, with a pass transistor and a reference, which have to cope with ripple as well as some remnants of switching noise, and do not have to source more than 1 amp or more than 12V.
It is not so much which part is a few uV better, or how many voltage references will fit on the head of a pin, but what is the goal? What kind of output does a linear supply used in a digital audio design need to avoid degrading what should be a good analog representation of the digitized signal.
It's "only" audio, right?
;)
So happy to be the fun for the day Fred, even more honored to be part of the entertainment of the week. |
|
|
| Kuei Yang Wang |
Koinichiwa,
| quote: | Originally posted by ALW
Not nasty at all, but have you built one?
|
Yes. With optimised local decoupling I found no reliable improvements in sound over several much simpler solutions I prefer.
NOTE that I AM aware of the very good technical performance of these regulators, it simply fails to translate into material improvement in sound, in my experience, ESPECIALLY considering the not inconsiderable effort and cost involved for these.
Of course, there are circuits out there that are VERY senstive to PSU Line noise (Naim Preamp's come to mind, they have no PSRR to speak of - hence you hear anything happening on the supply, equally, Master clocks for CD-Players are picky about PSU quality) where the differences may very well be audible.
For many Op-Amp's and similar sturcture discrete circuits there is plenty of PSRR which helps relativate the advantages of such a circuit as the Jung/Sulzer Reg over the 78XX/79XX.
One thing I personally have noted however, is that in "audio" circuits regulators that can sink current as well as source it (Shunt regulators, Bipolar Output Regs such as I proposed) offer often quite material improvements in sound over pretty good quality "series" regulators such the LT1085/1033, something I have also found in CMOS in digital circuits.
Anyway, just some thoughs. And for those that have not tried, stick a 3,300uF/6.3V or larger value high quality Cap (Panasonic FC/HFQ) onto the output of a 7805 supplying a bit of digital circuitry with good local decoupling. You may yet be surprised...
Sayonara |
|
|
| fmak |
[BTW I have very short hair now but very long hair 25 years ago. Does it really matter?
:bawling: [/B][/QUOTE]-----------------------------------------------
Yes, if you were reviewing loudspeakers 25 years ago. |
|
|
| Fred Dieckmann |
| The answer is 42 was mine. If you ever see me give a simple answer to a question I am most likely making fun. |
|
|
| jcarr |
Thunderstone:
>I find most 78XX/79XX (in fact all I tried) to sound very similar with a few 1,000uF after the Reg.<
I have conducted blind listening tests on 78xx/79xx regulators from various manufacturers, and in the test circuit applications, the regulators were frequently followed by a fair amount of capacitance. My experiments suggest that the statement quoted above may not necessarily be true for all circuit applications.
regards, jonathan carr |
|
|
| Elso Kwak |
| quote: | Originally posted by fmak
[BTW I have very short hair now but very long hair 25 years ago. Does it really matter?
:bawling: |
-----------------------------------------------
Yes, if you were reviewing loudspeakers 25 years ago. [/B][/QUOTE]
Haha, speakers are still the same KHorns......
:clown: |
|
|
| Fred Dieckmann |
| I don't think he is bluffing since he designes some very eleborate discreet transistor regulators. Shown here with phono preamp, with carefully selected three terminal regulators. I have seen a closer up picture, its hard to see here. |
|
|
| jcarr |
Andy:
>Topology is almost irrelevant, but PSU's bring big benefits to everything I've tried.<
Overall, I agree that putting more effort into the power supply usually yields subjective sonic benefits. Nonetheless, my findings so far suggest that circuit designs in which the total load current drawn from the power supply is constant - super shunt regulation, folded-cascode amplification etc. - the sonic effect of the PS filter capacitors is substantially reduced, and the sonic effect of the transformer is also reduced (albeit to a noticeably lesser degree).
hth, jonathan carr |
|
|
| PMiczek |
Ooops, I was too overcome by the outpouring of information to give proper credit.
As for adding caps everywhere, did I mention I like to design boards to be small, diy or not, even if they are "only" low-tech linear supplies?
Sorry for the stupid question, but my experience is more in digital design, and not audio and DAC. And yes, my boards often use a decoupling cap on the digital supply, sometimes even more than one per board :rolleyes:
(I will now go and rearrange my drawer of old neglected 78xx and 79xx parts in new and more esthetically pleasing patterns...) |
|
|
| Bricolo |
aren't we supposed to only place low values caps after ICs regulators like 7805s,1085s,317s ?
I heard many times that <200uF is best, and that more isn't good, since those chips aren't designed for this |
|
|
| Fred Dieckmann |
| With a little digital masking....... a gentleman never tells. |
|
|
| Elso Kwak |
| quote: | Originally posted by Fred Dieckmann
With a little digital masking....... a gentleman never tells. |
And what do I conclude from that???
:confused: |
|
|
| nobody special |
| quote: | Originally posted by dipchip
[B]peranders,
fill in the blank
great amp + bad power supply = |
I don't know.... Gainclone?
:clown: |
|
|
| PMiczek |
About the size/type of supply I had in mind, yes.
One more try:
But what were the performance goals, or requirements? I have seen lots of simple linear supply designs, many with little or no protection against RF or digital noise getting into the analog supply.
Of course, noise from any of my digital designs is only "good noise", and anyway I usually deny all knowledge of its existence if any happens to be captured on a scope...
Besides, the real world is all digital now, right? |
|
|
| Fred Dieckmann |
"And what do I conclude from that???"
That I know what the IC types are and don't know if it's something Mr. Carr wants to share with the world. Ask him....
he might not mind for all I know. |
|
|
| jcarr |
Dear Fred:
The power supply of the prototype unit in the photo that you posted consists of R-core transformers with dual secondaries, dual-bridge rectification, industrial-grade filter capacitors, and low-dropout adjustable regulators, with composite electrolytic and stacked-film bypasses at every load. But I did put a reasonable amount of effort into the board layout and harness design, and I can state with a reasonable degree of confidence that much of the sonic performance can be credited to this, rather than the components or topology (neither of which I consider particularly remarkable).
BTW Fred, I appreciate the masking gesture, but I already took the liberty of placing the really fun stuff _underneath_ the board, where it would be hidden from view. ;)
OTOH, for the production version, I am either going to put in a super-shunt regulation scheme, or a linear pre-reg and shunt reg combo. Yes, there has been the occasional daft discrete regulator or two (or 48), but not this time :).
I will probably also incorporate a variant of the following flyingcap circuit (Fred and Andy are both familiar with this one), although if previous experiences are anything to go by, I may want to multiply the total amount of capacitance that goes into the power supply by a factor of two or three over what I would normally use.
http://www.diyaudio.com/forums/show...1191#post151191
best, jonathan carr |
|
|
| ALW |
| quote: | | Nonetheless, my findings so far suggest that circuit designs in which the total load current drawn from the power supply is constant - super shunt regulation, folded-cascode amplification etc. - the sonic effect of the PS filter capacitors is substantially reduced, and the sonic effect of the transformer is also reduced (albeit to a noticeably lesser degree). |
Jonathan,
I'm sure I'd agree on all counts (I will build those circuits at some point!), but it is fair to point out that Super Shunt Regulation probably qualifies as a good PSU, rather than an amp circuit topology ;)
My suggestion to those that found the Jung-reg's not beneficial (in analogue circuits) were not using ideal implementations, or had other masking effects, if they found 3-terminal alternatives better.
Elso,
| quote: | | I try all kinds of circuits and try to learn from the practical results rather than copying a Jung regulator. I realise I have come a long way but will NEVER reach my goal! That's life! |
Don't ever take my posts as a put down of other methods, like you, what I hear is all - more than ever I find good engineering a pre-requisite to good performance, but almost NEVER a guarantee. Your ears are the best tool, but it must be understood that when dealing with PSU's that are at these levels of engineering, the design as a whole (not just the circuit diagram) is critical. The circuit diagram tells little about a circuit performance in practice.
| quote: | | Problem is: what is good sound quality; it's system, room and hearing dependent. |
Actually I disagree with this.
The band edges do not define musical performance, neither does tonal accuracy or frequency response.
Think about how much you can tell about the emotional state of someone on a telephone - limited bandwidth and dynamic range.
Some hi-fi totally misses this emotional element.
When you go to a live performance, particularly non-classical, the sound quality, in hi-fi terms is rarely great, but it often makes you dance, tap your feet, sing, cry etc...
There are assessment methods that home in on the primary aspects of music, and they work and provide long-term musical satisfaction if you're prepared to let them.
There's always an element of subjectivity, but I don't believe that has anything to do with the fundamental musical message.
Cue another blind listening test thread :bawling:
Andy. |
|
|
| janneman |
| quote: | Originally posted by peranders
My design idea was to use a slow opamp and creating a slow regulator because it was intended for rather constant loads. [SNIP]. |
Sorry Per I don't buy that. What load was that that needed a slow opamp for a slow regulator, but apparently couldn't be served by say a 78xx? Your bike rear light? And what would be a slow regulator by your book?
Jan Didden |
|
|
| fdegrove |
Hi,
Andy...you're going to like this one:
| quote: | | If I find myself not tapping my foot to the rythm of the music on the appropriate material the whole amp is not worth a dime to me. |
Thought I'd throw it in before Peter Daniel beats me to it.;) |
|
|
| Fred Dieckmann |
"I already took the liberty of placing the really fun stuff _underneath_ the board"
Well maybe not all the fun....... I cropped what might have been a **d* ****a* out the the photo and bet there a could few more ****a*s hiding on the bottom of the PCB. Those sound very good in that circuit location and is one of the most important places to use **d* ****a* if you are going to use **d* ****a*s. Of course I could be completely mistaken, like I was about the 48 volt relay......
John Drake |
|
|
| Elso Kwak |
| quote: | Originally posted by ALW
Elso,
Don't ever take my posts as a put down of other methods, like you, what I hear is all - more than ever I find good engineering a pre-requisite to good performance, but almost NEVER a guarantee. Your ears are the best tool, but it must be understood that when dealing with PSU's that are at these levels of engineering, the design as a whole (not just the circuit diagram) is critical. The circuit diagram tells little about a circuit performance in practice.
Actually I disagree with this.
The band edges do not define musical performance, neither does tonal accuracy or frequency response.
Think about how much you can tell about the emotional state of someone on a telephone - limited bandwidth and dynamic range.
Some hi-fi totally misses this emotional element.
When you go to a live performance, particularly non-classical, the sound quality, in hi-fi terms is rarely great, but it often makes you dance, tap your feet, sing, cry etc...
There are assessment methods that home in on the primary aspects of music, and they work and provide long-term musical satisfaction if you're prepared to let them.
There's always an element of subjectivity, but I don't believe that has anything to do with the fundamental musical message.
Cue another blind listening test thread :bawling:
Andy. |
Hi Andy..
I realise my way of working is totally different from most engineers.
I try to gauge my ears regulary on a life concert but NOT pop!!! Sorry I get sick and almost have to vomit after a hour at these extreme high listening levels and distortion. It is the most severe punishment you can think of for me.
I never experienced that the "right" PCB layout has a profound influence on the sound. I usually make boards Jim Williams of Linear Technolgy style before making a final layout.
But I try to avoid stupid mistakes. If the supply oscillates I clearly did something wrong.
It is my opinion the powersupply is part of the active circuit and has influence on the sound just as the actual amplifier schematic.
Next Saterday I will visit clavicord and harpsichord concerts accompanied by appetizing morsels!.:yummy: |
|
|
| fdegrove |
Hi,
| quote: | | It is my opinion the powersupply is part of the active circuit and has influence on the sound just as the actual amplifier schematic. |
But of course it is, Elso.;) |
|
|
| Elso Kwak |
| quote: | Originally posted by Fred Dieckmann
"I already took the liberty of placing the really fun stuff _underneath_ the board"
Well maybe not all the fun....... I cropped what might have been a **d* ****a* out the the photo and bet there a could few more ****a*s hiding on the bottom of the PCB. Those sound very good in that circuit location and is one of the most important places to use **d* ****a* if you are going to use **d* ****a*s. Of course I could be completely mistaken, like I was about the 48 volt relay......
John Drake |
Hi Fred,
And so we all have are little secrets.......
:lock: |
|
|
| ALW |
| quote: | | I usually make boards Jim Williams of Linear Technolgy style before making a final layout. |
ROFL - Yep, I've got circuits like that too ;)
Andy. |
|
|
| fmak |
[QUOTE.
When you go to a live performance, particularly non-classical, the sound quality, in hi-fi terms is rarely great, but it often makes you dance, tap your feet, sing, cry etc...
There are assessment methods that home in on the primary aspects of music, and they work and provide long-term musical satisfaction if you're prepared to let them.
-----------------------------------------------------------------
There is no basis in acoustics for what you said. Sounds like Ivor of Linn. Linn and Naim who subscribe to this had organised some of the worst demos I had ever attended. To me they were no fi s but noisy onslaughts. |
|
|
| ALW |
| quote: | | There is no basis in acoustics for what you said. |
I don't understand that statement - could you explain?
Andy. |
|
|
| fmak |
| quote: | Originally posted by ALW
I don't understand that statement - could you explain?
Andy. |
-----------------------------------------------
First, the assumption is that foot tapping to a certain type of music makes a good system - perpetuated by some in UK. Many, like me, listen to Classical more often and show their responses in different ways. One can have identical responses to music being played thru a 1930s record player and my parents-in-law do. Doesn't mean that is hi-fi. There is no connection between the statements you made and physical electro-acoustics that I know, although some may claim otherwise.
Doesn't mean that you it is not real to you; it's just not generalisable as inferred. |
|
|
| PMiczek |
Andy,
The version of Walt Jung circuit which started this thread came by way of your web page (?)
I think you wrote there in part,
"The above circuit offers excellent performance in all of the major areas relevant to a regulators operating parameters, it has very low output impedance, very low noise both static and under load, and good line rejection. All of these parameters are maintained over a far wider bandwidth than any readily-available monolithic IC regulator."
What does "very low" noise, and "good" line rejection mean, relative to the specifications of a modern 3-pin, or 5-pin (there, I said a bad word, I know) regulator? Not to mention "far wider". Not trying to be confrontational here, but if I am going to throw out those 3-pin regulators I just dusted of in favor of an op-amp circuit with many more parts, or even buy a kit to replace them in existing power supplies...
Didn't someone suggest earlier that all the world needs now is more sweet capacitance, and now we find out what it really needs is a good error amplifier!?
:confused: |
|
|
| fdegrove |
Hi,
| quote: | | First, the assumption is that foot tapping to a certain type of music makes a good system |
Unfortunately I neither like either Linn nor Naim stuff...
What I do agree on is that no system that isn't inviting to either "play conductor" (on classical music). or "tapping the foot to the rythm of a tune" is doing the music any good...no matter how revealing of inner detail etc., etc....
After all we're there to listen to the music and enjoy it, not to analyse it to shreds.
Cheers,;) |
|
|
| Kuei Yang Wang |
Koinichiwa,
| quote: | Originally posted by ALW
My suggestion to those that found the Jung-reg's not beneficial (in analogue circuits) were not using ideal implementations, or had other masking effects, if they found 3-terminal alternatives better.
|
Note - I NEVER use 3-pin Jobs for analogue. I prefer film capacitors and chokes. But then, by choice, all my analogue circuits use good sounding Valves.
But more to the point, it seems my main point was misunderstood. Other than remarking upon the inability of the Jung/Sulzer (and mny others) regulator, I do not claim they are "bad".
On the contrary, in absolute measurable terms (wich exclude of course transient tests with current "pushed" back into the rail) the Jung/Sulzer reg is a major achievement of engineering at the margins of the achievable.
However, to conclude the measured superiority translates reliably into subjectively percieved superiority is arrogant. Further, considering the usual limitations posed by the use and implementation of "normal" 3-pin 78XX/79XX/317/337 et alstyle reg chips, replacing them with a 3-Pin Jung/Sulzer reg is unlikely to give much of an improvement.
Now IF the Jung/Sulzer reg where implemented with remote sensing and suitable ground layout there may be more of a point. But to make a 3-pin (78XX/79XX compatible PCB for the Jung/Sulzer reg is to throw away most of the technical benefits (or not) of that design.
So, I do not question that the Jung/Sulzer Circuit is a "good thing" (in the way Winnie the lovable Pooh sees a "good thing").
My question is:
"leaving out hype and concentrating on real applications, will the 3-Pin Jung/Sulzer circuit deliver a reliable improvemnt in subjectivly percieved sonics, comensureate with it's cost, especially when compared to other approaches?"
If the answer is not an absolute resounding "Yes", one may consider less expenive and complex alternatives that may be as goo or even better.
Sayonara |
|
|
| fdegrove |
Hi,
| quote: | | If the answer is not an absolute resounding "Yes", one may consider less expenive and complex alternatives that may be as goo or even better. |
It can't be a resounding "Yes"...there no such thing as the universally best regulator...
Sometimes, actually alot of times, I do get the feeling that engineers are just like politicians...neither have a feel for reality.
Cheers,:rolleyes: |
|
|
| PMiczek |
| quote: | | Sometimes, actually alot of times, I do get the feeling that engineers are just like politicians...neither have a feel for reality. |
And when did you first start noticing this about engineers? (LOL)
... I was only asking for a couple (real) numbers, or maybe hoping for a link to a simple graph.
Entertaining Fred and Andy in the process was a completely un-intentional side effect, if not unexpected.
(maybe if I sort these regulators by size and color...) |
|
|
| fdegrove |
Hi,
| quote: | | And when did you first start noticing this about engineers? (LOL) |
Twenty years ago already...
Put it like this, most will swamp you with numbers but when it comes to designing for music reproduction they're just as clueless as you and me.;) |
|
|
| pooge |
Several years ago I communicated with Walt Jung by email regarding the conversion of his Super Regulator to 24v operation, which I needed for my Borbely lineamp and Leach phono amp. He was unsure of doing so, since he never tried it, but offerred suggestions. Unfortunately, I didn't act on them at the time, because I thought complete boards with the preregulator and 825 amp layout might appear. Then, I lost my emails from Walt after a hardrive crash. Last week I decided to look into resurecting the project, and lo and behold, came upon this site and timely thread.
Andy, I noticed on your site that you mentioned a use for 24v supplies. Wondering if you did this with the Super Regulator topology.
Anyhow, my searching led me to several boards other than the Old Colony ones for the original Jung regulators from '95 (can't believe they don't have updates with preregulators yet!)
Borbely offers 24v regulators (EB 802/243), but not with the floating preregulators and other tweaks that the Super Regulators have. He also uses an FET for the pass transistor, which I fail to understand. All in all, not what I want to pay such a high price for, and then have to pay $20 in addition for the parts list! For anyone interested, here are some links to those regulators:
See figure 6 on this page:
http://www.borbelyaudio.com/index8.htm
Click through the download links to see data sheet for the Borbely regulators from this US distributor:
http://www.audiokits.com/products.a...ails&prodid=393
I also found a board with a 15v Sulzer/Borbely topology with preregulators (not floating) at this site, but today I couldn't get through to the links to the schematic and board pictures:
http://hitechnetworks.net/bwaldron/...1_regulator.htm |
|
|
| ALW |
| quote: | | However, to conclude the measured superiority translates reliably into subjectively percieved superiority is arrogant. Further, considering the usual limitations posed by the use and implementation of "normal" 3-pin 78XX/79XX/317/337 et alstyle reg chips, replacing them with a 3-Pin Jung/Sulzer reg is unlikely to give much of an improvement. |
I didn't actually claim that, my point was simply that unless the regulator is working impeccably, the benefits will not be realised.
| quote: | | Now IF the Jung/Sulzer reg where implemented with remote sensing and suitable ground layout there may be more of a point. But to make a 3-pin (78XX/79XX compatible PCB for the Jung/Sulzer reg is to throw away most of the technical benefits (or not) of that design. |
That's untrue, you are now making the assumption that the more obvious regulator parameters (impedance etc) are the major issue - they're not.
| quote: | | What does "very low" noise, and "good" line rejection mean, relative to the specifications of a modern 3-pin, or 5-pin (there, I said a bad word, I know) regulator? Not to mention "far wider". |
Very low = the noise of the error ampa s being dominant (10's nV / rt Hz) 3 terminals are 100's nv/rt Hz
Good = >130dB at LF, about 90dB at HF (3 terminals are 80dB max at LF)
Bandwidth - depends what you want to define as your limits, but impedance = 60uOhm at LF, rising to 1mOhm circa 100k.
Other parameters defined above.
Output Impedance vs Frequency
Thanks to Jan Didden for the above.
Andy.
P.S.| quote: | | First, the assumption is that foot tapping to a certain type of music makes a good syste |
You will have noted I mentioned other criteria too - the point being that emotional involvement in music, however it comes is fundamental - the other stuff is icing on the cake (imaging etc). Let's face it, we don't judge a guitarist by where he stands on the stage ;) |
|
|
| fmak |
[QUOTE]Originally posted by PMiczek
[B]
And when did you first start noticing this about engineers? (LOL)
-------------------------------------------------------
As an engineer, though not electronics, I sometimes find these threads overbearing. The original questions are often not addressed, and the arguments go on and on. In this case, we are still on about the Jung regulator of considerable vintage. Nothing about the new breed of micropower regulators like the LT1761-1764. I ordered a batch and couldn't get the 300 mA SMT 8 to work properly using an adaptor (hum) but the TO22O 1.5A one worked beautifully, with 40 uV noise using 100 uF output cap and 15 uV with 2200 uF cap!. At $3 a go, this has to be worth it. Pity that manufacturers rarely give you the whole spec such as output impedance etc.
Anyone with similar experience on the SMT verssion? I suspect it is the thin traces on the adaptor , and contact resistance. Threre is 400 uV of 50 Hz I couldn't get rid off. |
|
|
| peranders |
| quote: | Originally posted by Elso Kwak
Hi Andy..
I realise my way of working is totally different from most engineers.
I try to gauge my ears regulary on a life concert but NOT pop!!! Sorry I get sick and almost have to vomit after a hour at these extreme high listening levels and distortion. It is the most severe punishment you can think of for me. |
Why don't you do like I: I use earplugs which I trim for best frequency response. Rolling Stones and David Bowie sounded excellent! Pink Floyd sounded excellent without earplugs. The concert a couple a years ago at Ullevi Göteborg (>50 k people), was my best consert experience ever :att'n::) |
|
|
| peranders |
| quote: | Originally posted by pooge
He also uses an FET for the pass transistor, which I fail to understand..... |
Mr Borbely :cloud9: JFET's and MOSFET's. That's the simple explaination. :nod:
Advantage with a MOSFET is no power to drive it in "steady state". |
|
|
| ALW |
Pooge,
24V is easy - 680R / 1k8 for feedback divider (gives 25V nom).
Just choose suitably rated caps!
The output voltage is calculated from the following formula: -
Vout = Vref * (1+R8/R9).
Additionally R8 // R9 (the impedance of the two resistors in parallel) should equal 500 Ohms.
To make this easier the following formula has been derived: -
R9 = (500 * Vout) / (Vout – Vref)
Once the above is calculated, apply the value obtained above to the following formula: -
R8 = R9 * ((Vout – Vref) / Vref)
The two values then can be selected from the closest standard value, or made up from parallel / series combinations.
Andy.
P.S. A FET has an order of magnitude disadvantage in these reg's as the pass device. |
|
|
| ALW |
before the knives get draw ;)
The graph above has an impedance rise at LF, this is due to the inadequate value of coupling cap used to couple the error signal into the reg. and not the reg. itself.
| quote: | | As an engineer, though not electronics, I sometimes find these threads overbearing. The original questions are often not addressed, and the arguments go on and on. |
I actually agree with you on this. All I ever do is present my experiences, sometimes, as with power supplies, I cannot disclose all of the facts I'm aware of due to various political / commercial arrangements and the fact that so far the knowledge I have would not seem to be public domain.
But I do give plenty of clues for those that care to look, and really try to encourage people to both approach things in a detailed rigorous engineering fashion, but then to use their ears.
In audio the latter is critical and fundamental and also easy to get wrong, on first impressions. I also strongly encourage you all to think outside of conventional wisdom.
The Jung reg's are old, but I've only heard one alternative that outperforms them at present. In all likelihood there's lots of alternatives, supported by my view that topology is often the least critical element, just correct implementation of that topology.
Arguing mine's better than yours is probably specious, but I'm happy for people to copy my design (that includes, by necessity, the PCB) and compare to alternative topologies.
The Jung reg's, to this day, do things that many of the alternative topologies do not - I've tried loads, and barring one circuit I'm working with now, and in all likelihood some of Jonathan Carr's excellent posts, there's little to better it.
I've never used a circuit insensitive to power supplies, but would be happy to have one shown to me.
I am, despite what may seem to be the case, about the most open-minded person I know and always willing to learn - so teach me stuff :cool:
| quote: | | One thing I personally have noted however, is that in "audio" circuits regulators that can sink current as well as source it (Shunt regulators, Bipolar Output Regs such as I proposed) offer often quite material improvements in sound over pretty good quality "series" regulators such the LT1085/1033, something I have also found in CMOS in digital circuits. |
I'm sure this is true, but have you tried the same by effectively pre-loading the regulator output, in low current circuits this is a more elegant, lower cost approach, with a small penalty in dissipation?
Push-pull output stages can bring about effects that I know would be sonically detrimental unless ideally implemented. I'm sure a shunt / supershunt would work well, but am less convinced that many could build a push-pull series reg that would work as well.
Andy. |
|
|
| peranders |
| quote: | Originally posted by ALW
P.S. A FET has an order of magnitude disadvantage in these reg's as the pass device. |
I could image that lower gm is one thing and another is much high caps so "garbage" leaks over to the regulated side easier. |
|
|
| Kuei Yang Wang |
Koinichiwa,
| quote: | Originally posted by ALW
have you tried the same by effectively pre-loading the regulator output, in low current circuits this is a more elegant, lower cost approach, with a small penalty in dissipation?
|
Yes. I generally pre-load 3-pin's wherever I use them. I tend to use LM317/337 as pre-regulators and I prefer to use very low value resistors in the voltage divider setting the output voltage. This serves to "pre-load" these. For all other position I prefer to use the LM/TL431 as simple shunt, with optimised bypassing, giving a wideband, non-reactive PSU line with low but not excessively low impedance.
Unless of course I use Valve Circuits (which I mostly do) and where I do not use any regulation, just loads of chokes and Film Cap's... (or 'lytics if it is the LT)
| quote: | Originally posted by ALW
Push-pull output stages can bring about effects that I know would be sonically detrimental unless ideally implemented.
|
In some experiments I found driving the PSU Pin's of Op-Amp's from the same type Op-Amp (which is a "Push-Pull Reg") with NO capacitance in the PSU line (this does take very TIGHT, three-dimensional layouting to work) sounded probably as good as Batteries with Filmcap only bypassing (Siemens naked chip stacked film MKT's).
I don't know if that qualifies as "ideally implemented", but it sure works well.
Sayonara |
|
|
| janneman |
| quote: | Originally posted by Kuei Yang Wang
Koinichiwa,
[snip]In some experiments I found driving the PSU Pin's of Op-Amp's from the same type Op-Amp (which is a "Push-Pull Reg") with NO capacitance in the PSU line (this does take very TIGHT, three-dimensional layouting to work) sounded probably as good as Batteries with Filmcap only bypassing (Siemens naked chip stacked film MKT's).[snip
Sayonara |
Sayonara, this is exactly the info I was looking for. I even started a new tread (local regulator). Can you give me some more info pls?
Jan Didden |
|
|
|
