|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
61.92.98.16
In Reply to: RE: Dc Regulator Comaprisons posted by fmak on June 25, 2015 at 11:54:29
Hi,
> You have said little of consequence.
Really? Or maybe only people who properly understand electronics find it consequential?
> Your remarks do not mean that transient tests
> are inconsequential either.
My remark means that transient tests are inconsequential EXCEPT in cases where the the circuit being supplied is bound to generate such transients AND where if such transients or changes on the power-supply are present the circuit performance is degraded.
Such situations exist (e.g. an SE Triode Tube Amp output stage is a great example), but they are comparably rare.
And so my point remains, that using a single test to provide "proof of superiority" is rarely meaningful.
Using the step response in regulators is especially meaningless, as commonly there is enough capacitance on the circuit-board following the regulator to totally swamp the actual regulators behaviour (as you can see in some datasheets where they show this response with different output capacitor values).
> Look at the data sheet for regulators, why does LT and others
> bother to include such data?
Because it is part of the standard data set. Why do Manufacturers datasheet show squarewave response etc. for Audio Op-Amp's which will never be used with a squarewave?
> If manufacturers present the data,
They do so, because everyone does.
> it is meant for comparison in applications where fast demands
> are made, as in a computer.
Funny thing is, last time I saw a linear reg on a Computer Mainboard PCB, CPU's were still dissipation (and consuming) constant power no matter if running full load or idle and yes, you guessed it, it means no change in current. Hard drives were a different story, but they started to put switchers on the drive board ages ago.
Modern Computers use aggressive clock scaling, even turning cores off to reduce power consumption, it is this what primarily modulates the power supply current demand. The regulators used are invariably switchers, since before Y2K or so.
So, cute theories, but please give them a few shaves with old Bill of Occam's razor, to see if they even can hold any water.
But as you are sold onto this particular companies sales pitch, please give them your money and keep believing in the efficacy of their remedy.
I am sure that the results will prove as spectacular an upgrade as fitting 75 Ohm BNC connectors to gear where neither the electrical circuit driving the BNC socket or the wiring from that to the BNC comes close to being either 75 Ohm nor to the actual characteristic impedance of the BNC socket.
I am looking forward to your writing on the topic, as anything that affords some levity is welcome during all that doom and gloom and greek tragedy on the news.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
Follow Ups:
If you wish to make a substantive point, go to some other forum to discuss power supplies for the above, but not here in the context of power supplies in Computer Audio.
I am amazed that you have not sought to discuss the issue with John Swenson, Paul Hynes, audioengr and others, who in concert with users, have posted on the need for stiff, fast PSs.
Any power supply that collapses under load conditions are clearly not suitable.
How does your smps fare?
Hi,
For anyone to say "A powersupply needs to be XXXX", see above post. In SOME SPECIFIC applications this is true and the most critical element (regardless what XXXX actually is) but in most cases it is not.
I am playing on a different level to simple (un)truisms that "powersupply must be XXXX" or similar overly simplistic marketing phrases.
I prefer to accurately specify what the circuit I supply actually requires and then to design the powersupply solution that does the best job into the circuit, rather than trying to bodge the same square peg ito all the different shape, except square, holes...
> Any power supply that collapses under load conditions are clearly
> not suitable.
A powersupply that collapses under load is unregulated and it has many applications, though perhaps inside computers it is better followed by regulator of some description.
A powersupply (IC) that is exposed to load transients that cause it to show problems (such as the contentious test showed) caused is a powersupply (IC) that is insufficiently decoupled and hence has not been correctly designed into the system. The solution is of course to correct the design flaw, rather than to try replacing it with "magic regulators", just like correcting impedance mismatches requires correct circuit design, not magic connectors, in the real world anyway.
There is no substitute for solid engineering first and fairy dust plus other tweaks second.
> How does your smps fare?
Try it.
I know how much capacitive decoupling with what ESR etc. I put on the output (and this basically determines transient response in my design, not the Regulator) and how much short term over current capacity I have designed in, compared to rated output (this determines how much short term over-current it can handle).
I would say that just as with other supplies from our corner (as some testers found out to their embarrassment) what you will be measuring is almost entirely the DCR and inductance of the wires used. The contribution of the PSU itself is negligible next to that.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
Doesn't sound that great. Others use LPS too instead of it; see posts in CA and elsewhere.Mine that came with the iUSB power broke down after an hour and had to be replaced by the dealer.
If you want to argue aimlessly about Belleson's specs compared to yours, why don't you take it up with them instead of starting an argument about information that I provided a link to?
I see no comprehensive spec from your products about power supplies for a particular piece of hardware. What I did see were over the top claims about audio performance.
Edits: 06/26/15 06/26/15
Fred,
And here I was expecting you to make a substantive point, show how the SMPS fares when tested for transient behaviour in the precise kind of setup which was used in the contentious test...
And following that you would show reliable evidence that links this transient misbehaviour to poor sound quality. I even had a crow caught, executed and prepared ready for me to eat - in public.
Instead you waffle on about the broken power adapter which was promptly replaced.
Sound quality is very subjective and I see little point to debate it.
Where I see a point of debate if you make claims that Item A has a certain specific objectively tested behaviour that causes it to sound this way or that.
Past that, no you see no detailed spec for each and every power supply circuit powering each and every part different part in the unit, because what matters is actually the subjective sound quality.
As for over the top, apart from sour grapes from one or two individuals who pretty much never heard any of the products I had a hand in, I have not seen any real criticism of sound quality, especially in respect to the price. Rather the opposite.
You disagree, so don't buy any of those products. Buy anything else. Please.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
you are the one complaining about info from Belleson.
Take it up with them.
Fred,
Posted by fmak (A) on June 24, 2015 at 00:18:17
http://www.belleson.com/comparemono.php
Note the relatively poor transient performance of the LT1086, LM1117 used often in computer hardware, and the Burston.
Note the poor spelling of names and questionable grammar which is obvious to someone with english as third language.
And, no it is not a direct claim, just an attempt to spread FUD.
Posted by fmak (A) on June 25, 2015 at 21:54:54
John Swenson, Paul Hynes, audioengr and others, (who) in concert with users, have posted on the need for stiff, fast PSs.
Now here we have a claim, if arguably attributed to others, but by repetition and posting endorsed and supported by the posts author.
Posted by fmak (A) on June 25, 2015 at 11:54:29
If manufacturers present the data, it is meant for comparison in applications where fast demands are made, as in a computer.
Now here we have a claim, not attributed to others, but straight and direct.
Now, shall we dig out the whole subject of the magic BNC connectors who magically fix everything, including bad (non 75 Ohm) cabling, incorrect signal levels and incorrect drive impedance from the driving circuit and the related claims?
I am game.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
Clearly you have an agenda which you need to pursue with Belleson instead of wasting space here.AS a manufacturer, you have broken forum rules umpteen times in this exchange.
Edits: 06/26/15 06/26/15
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: