Lately I have been reading about some well known companies who make amps and preamps based on switch mode (digital) power supply. Nagra, David Berning, Linn, Crayon Audio comes to mind. I have heard a couple of their products but I always seem to hear some kind of switching noise which comes through as "digital" sounding. The organic quality is somewhat robbed and replaced with some hash. I wonder if there are any designs using SMPS that can actually sound natural and organic ?
To me it seems mostly a matter of convenience to use SMPS but I would love to discuss.
but I always seem to hear some kind of switching noise which comes through as "digital" sounding.
hmmmm.....I wonder if you are imagining this? How did you trace/pinpoint the "digital" sound to exactly the SMPS in the preamp/amp?
I wonder if there are any designs using SMPS that can actually sound natural and organic ?
sure! At RMAF2013 I heard the Rowland 625 driving a Raidho stand-mount & it sounded very good. For the limited time that i was in the room on Sat & Sun I could not find any obvious faults. I, of course, discounted the fact that the Raidho, no matter how good, could not reproduce deep bass. There’s been rave words said about the Merrill Veritas amp which also uses a SMPS power supply. If I'm not mistaken Bel Canto Black is another example. There must be many others out there.....
SMPS used for convenience? Maybe. At one point i was talking to a well-regarded engineer at a class-D amp manuf & was informed that SMPS in amplifiers are just about coming into their own. Their fault, as you stated, was high(er) output noise for audio applications. We already used SMPS in other audio applications such as cellphones as the audio quality does not matter as much (different application & diff specs). I was informed, by this engineer, that the highlight of the Hypex NCore N1200 amplifier module was its SMPS. And, the final goal for any class-D amplifier was to use both a SMPS power supply + SMPS amplifier module.
I don't have solid answers for you, but I do have an experience to share which I believe is a partial answer. I wanted a better dac that got me into better ES9018 chip. I purchased a kit and did all the work. After speaking to an old friend who is fairly well known years gone by as a competent designer, he convinced me that done right, there was nothing wrong with a switching supply. Being a beyond frugal, okay, cheap person, I had found a TV/DVD switching power supply that had the exact voltages that I needed for my dac. I purchased this, put it in, fired it up.... IT WORKS... Fast, smooth, but had a glare that I couldn't deal with, waited for weeks to break in. Still glare. Took it to my buddies house, he took measurements, measured again and ended up building a filter network for the power supply. Wholey Moley what a difference. Glare is gone, detail improved, weight/authority. I became a happy camper quickly. Another friend purchased a Pass/First watt clone that Nelson offered up on DIY.com.. My buddy told him the same thing... Again, DONE PROPERLY a switching supply is fine, even on a traditional amplifier. In his case, he had already purchased 2 transformers and ran that direction. I'm a speaker guy, but I'm sure one of the amp guys out there will chime in and tell you why.
Jeff Rowland uses SMPS for low noise. That’s why he uses them in applications where efficiency is non-issue, like Capri preamp. SMPS got bad rap from crude computer applications but when is done right is extremely quiet. Most of linear power supplies are in reality primitive switchers that switch at 120Hz at the peak of the voltage causing transition noise that propagates thru the amplifier. In addition 120Hz ripple is difficult to filter out thus requiring big electrolytic caps. These cap, in series with the signal path (circuit closes thru them), are inductive and slow down response. Power transformer has to be huge at this low frequency. Modern SMPS switches at zero voltage/ zero current at frequency that is inaudible and easy to filter out (Rowland’s 625 SMPS switches at 1MHz). It has fast response time and is line and load regulated while linear supply in power amps is not. Huge transformer is replaced by small ferrite one that can carry the same power at very high frequency. Benchmark replaced linear supply in their newer DACs with switcher while their new 132dB S/N ABH2 amplifier also has SMPS. SMPS can be quiet not only electrically but also acoustically being free of buzz and with AC/DC operation, often in wide range.
I was always a skeptic with SMPS all they way until I heard Soulution. Never heard the Rowlands but they are supposed to be real good. Raidho used them at a lot of demo's. Then they came out with the Aavik which they now use. Now Aavik just came out with a class A amp and pre but I don't know if they are using a SMPS.
Given the positive and encouraging comments here with regard to SMPS what explains the continued popularity of linear power supplies? It leads me to suspect that they must have some significant inherent justification. Or is it just stubbornness on behalf of many builders? Charles
FWIW, my DEQX HDP-5 preamp/processor, which is their top-of-the-line model, uses a switching power supply. Their somewhat less expensive previous top-of-the-line model, the HDP-4, uses a linear supply.
I suspect that a factor in their decision to make that change may have been a need for additional internal real estate, because the newer model incorporates a touchscreen and some other functions that were not provided in the HDP-4. But in any event I have been very pleased with the unit’s transparency, absence of coloration, and total absence of noise at any volume setting.
Charles, regarding your question my guess is that contributing factors to the continued bias toward linear supplies that seems to be evident among many designers include their own familiarity with the use of those supplies, and perhaps also an expectation that many potential customers would be similarly biased.
Why to design complicated SMPS when the most of customers believe that it has to be big and heavy to sound good? Small and efficient bring suspicion that something (sound) has to give. In addition word "switching" suggests electrical noise.
Jaybe, I'll check that site and read the commentary presented. If the prevailing argument centers around measurements and specifications that won't do it for me. The bottom line criteria for me is what does it sound like? Al, I know you are a technical maven. Yet I still believe that you must be satisfied with the sound quality of an audio component before you'd place it in your system. I make that assumption based on your current system makeup. Charles,
Charles, yes, as I’ve said in a number of past threads my belief is that specs and measurements can often be useful, but primarily because they can often make it possible to **rule out** components from consideration. In some cases by raising doubt (or worse) as to how suitable a component may be for use in conjunction with other system components. In some cases by raising doubt (or worse) with respect to how suitable a component may be with respect to user requirements (such as a desired maximum volume capability, for example). And in some cases because a spec or measurement that is **too** good (such as extremely low total harmonic distortion, for example) may create suspicion that something else has been compromised to achieve that (such as heavy-handed application of feedback).
But I’m certainly with you in considering sonics to be THE selection criterion once the list of candidates has been narrowed down as much as possible on the basis of specs and measurements, as well as on the basis of cost, physical configuration, manufacturer reputation for support, and other such factors.
Also, as you may have observed over the years, with a few exceptions I usually tend to avoid participating in threads involving ecclesiastical controversies about the superiority or inferiority of one design approach vs. another. Such as might arise with respect to switching vs. linear power supplies. The reason being, as I know you would agree, that what usually matters most is how well a chosen design approach is implemented, not which approach is chosen.
Al, Without a doubt implementation and sheer designer talent are the major determinates of a product's quality and performance. You are unquestionably wise in avoiding the endless non-productive squabbles that are all too common. I've backed off myself the past couple of years. Charles,
I started this discussion not because I was imagining things, at least thats what I would like to believe. I had some experiences around SMPS based equipments within my own system. I had a couple of CDP from Linn (Ikemi) and Bluenote, I also had a phonostage which had the option of using an external PSU (where I could try both their SMPS model and Linear PSU model). I also had an opportunity to listen to a very well reviewed amplifier with SMPS. In fact all these equipments are very well reviewed without any mention of any issues around SMPS distortion/noise.
My personal observation though has been different. Since I have heard them all in my own system and that too besides some other non-SMPS based equipments I do not doubt what I felt about them. All of them had the ability to sound faster and more transparent. Air and Resolution also goes up compared to similarly priced equipments with Linear PSU. All this is great hence there is a certain wow factor there. Where things start to fall apart is in some inherent noise/hash/distortion whatever you call it which is not audible as a straight forward graininess or muddiness that we are typically accustomed with linear PSU. It is rather something that affects the tones and presentation which ultimately results in a less relaxed sound. Mind doesnt relax to the music but is always constantly being stirred with whats going on. There is a bit of frown, a bit intrigue, a bit of wow, all mixed up because that distortion or noise however little is neither ignorable nor passable as music, something very similar to distortion/noise of a CD vs LP.
Recently I had an amplifier at home which sounded terrific in many respects, one of the fastest, most well resolved most dynamic amps I have tried. Yes it had a very "advanced" SMPS and thats also its USP. While I was trying to get a hang of its sound I was constantly getting bothered about something that was not "usual". Especially the high frequency on the amp sounded like an early digital CD player, but then my source is a Garrard 301 played back through a Klyne phono + preamp. It is a great analog combo though a bit vintage to some ears. With this amp it sounded like digital media player which has not been fully done up especially in the highs and also in the lows, the bass suddenly reminded me of typical digital bass, which is quite unlike the bottomless analog bass we expect from a good analog source. As I was trying to decipher all this, I played an LP which had some clicks and pops, to my surprise the clicks and pops which I generally find ignorable in some of these LPs now grabbed my attention more than ever. The clicks and pops were sounding like there is some shiny disc being played in my system. They sounded hot, sparkly and sharp like flashes of fire dots. I tried various power cords and ICs especially shielded ones to see if some RFI/EMI is causing it but things remained pretty much the same. That quite nailed things for me and hence I started this thread.
Long time back I read an article which said, when one is listening to an amplifier, he is actually listening to its power supply. It is ubiquitous that all PSU do distort to some extent or other based on various factors. These distortions are definitely different in a Linear PSU vs an SMPS, typical analog vs digital. What it does to the sound is also going to be different. It is just that I find one of them to be much more forgiving and listenable than the other.
Designers probably take their stands based on various factors. They need to wow their customers and investors and get things going to the market quicker than their competiros. The fact that reviewers and most stake holders in the industry accept it without a"frown" probably makes it easy for them. Having said all that I havent heard the Rowlands and Soulution yet.
I'm doing well. thanks for asking. I see that you are also doing well & growing by leaps & bounds as an audiophile. :-) thanks for your feedback, Pani. Now your original comments make a lot more sense. Would have been nice if you had stated all this in your 1st post to give all of us a background where you were coming from. ;-) anyway, it looks like we have a variety of opinions on this topic from several owners & listeners.....
@pani As stated above SMPS has come a long way. But I can think of a reason a LPS may sound better in your home. One thing is a transformer will filter out or reduce incoming high frequency electrical noise where especially older SMPS will pass it through.
That said I'm the type of listener who stopped relying on specs and measurements when deciding what to buy. At least for me it left biased one way or the other.
Given the positive and encouraging comments here with regard to SMPS
what explains the continued popularity of linear power supplies? It
leads me to suspect that they must have some significant inherent
justification. Or is it just stubbornness on behalf of many builders?
I'd love to 'switch' (if you will pardon the expression) to SMPSs but to get the voltages we need for our products simply isn't in the cards unless we buy a lot of supplies, costing 10s of thousands of dollars. For a small business its not practical unless the power supply is conveniently available off-the-shelf. So traditional supplies will be with us a long time.
But I can think of a reason a LPS may sound better in your home. One thing is a transformer will filter out or reduce incoming high frequency electrical noise where especially older SMPS will pass it through.
Modern SMPS often contain Power Factor correction module, basically a huge inductor and capacitor, making supply appear as resistive load but also filtering any line noise. They are also line and load regulated with fast response that will eliminate in tandem with supply capacitors most of the line noise. In contrast, most of linear power supplies draw supply current in short spikes of high amplitude that "travel" thru the power cord and pollute everything around. The only thing "linear" about "linear power supplies" is that they are unregulated. I would suggest replacing word "linear" with "unregulated" since they (most of them) operate as primitive unregulated switchers - nothing linear about it.
@kijanki That’s for the clarification. I’m going by what I learned in school a very long time ago. ;-)
Just another reason I rarely rely on specs and measurements. I trust my ears the most. Ive heard stuff that has great specs and measurements that I just don't care for. And others that look horrible on paper but sound great (at least to me). Visa versa too.
The better LPSs that are made specific for audio use certainly are regulated. Look at the Sonore, Uptone, HDPlex and other well-designed LPSs and you'll find this to be true.
There's also many people of the opinion, with evidence to back it, that it's the SMPS supplies that pollute the local circuits.
I believe this can all be further fleshed out if you bring your opinion over to Computer Audiophile where there certainly is plenty of opinions, as there are here, but you will also be more likely to find some hard test data for review.
I agree that the ears tell the tale, but from some of the posts here some minds are already made up that LPSs suck and SMPSs do not.
Personally I don't get the sense that anyone is suggesting that LPS suck. This thread is conveying that SMPS can sound every bit as good and that technically LPS is "primative" relatively speaking. There are multitudes of components with LPS that sound superb, in fact state of the art.
Ralph, I am familiar with the very high sound quality of your products, are you suggesting that they'd sound even better if you could use SMPS rather than LPS? Charles,
Jaybe, I repeat again - most of linear power supplies in power amps are unregulated. I would bet that 100% of linear power supplies in class A amps are unregulated. Can you provide the link? Sonore, you mentioned are 14W, unless I'm looking at wrong website.
Bombaywalla, Thank you for the link, very interesting. The only problem is that what they describe:
The digital control circuitry constantly monitors the pulsating waves from the regulator and the rail voltages. It will then make a decision to turn the coupling transistors on or off at each zero point to add as many or few pulses as required to hold the voltage constant.
is switching regulator. Any time something is completely on or off it is switching. Idea of controlling number of full wave cycles is not new - it used to be called "Group Regulation" and was even used in such applications as quiet light dimmers. The problem is, that power supply is still "linear" and in order to respond fast it needs low inductance capacitors. The best you can get is "Split Foil" type, but it is expensive and still poor. Adding parallel low inductance capacitor helps, but also creates resonant circuit with inductance of large main caps.
Their regulator is a step in right direction (switching), but why not use switchers? Linear power supply switches anyway at 120Hz at max voltage. Yes, not many companies use SMPS with class A or AB amps but some do (Rowland, Benchmark, Linn etc). I don't understand term used in their paper "Linear Amplifier". I've never heard it before. I assume, they mean "not class D", but class D is still linear (and not digital) in all respects.
welcome Kijanki. the Magtech regulated power supply looks like a switching regulator but it is not a traditional SMPS (like the kind used by Rowland, Linn, Benchmark, etc) but it looks to me that it is some sort of Pulse Frequency Modulated (PFM) switching regulator - it monitors the pulsating waveforms out of the regulator & the rail voltages (it seems that the system has some knowledge of min rail voltage & max rail voltage & what the acceptable ripple should be) & when the rail voltage goes below the min, it multi-pulses until the rail voltage reaches the max & then shuts off. Cycle repeats when the rail voltage falls below min.
2ndly, it seems that the Magtech power supply is very much like the multi-voltage used by ARCAM in their class-G amplifiers. see this link:
Bombaywalla, I'm familiar with class G principle of operation. And you're right - their "invention" is pretty much class G. It does not matter if you switch signal or supplies - it is still switching operation. Rowland (IMHO) is the leader again with unpopular, while great sounding choices.
Interesting references. FYI, I tried to find Mr. Sanders’ patent or patent application for his regulated power supply design, which would have provided more information about it. But extensive searching at the Patent Office website (uspto.gov), using numerous search terms that seemed likely to be relevant, turned up nothing. So unless I somehow missed it the application has apparently not yet been submitted.
Let me first make clear that I was not implying anything negative about Mr. Sanders or his design. I was just reporting that I looked for the patent or patent application, in the hope of finding further detail about the design, and couldn’t find it.
The mention of "patent pending" appears in numerous articles on the web, as well as in Mr. Sanders white paper that you referenced earlier. Presumably all of those statements were based on inputs he provided, at the times the articles were written.
Regarding the unrelated patent application you linked to just above, which was filed in 2003, I can find no indication that a corresponding patent was ever issued.
I suppose there are all kinds of possible explanations for these non-findings, ranging from issues with the Patent Office’s website and/or its search functions to decisions he may have eventually made to not pursue the applications, for whatever reasons.
Thanks for the clarification, Almarg. The fact that the 2013 patent application (I believe it was 2013 & not 2003) has not issued seems par for the course. I had one patent application issue in 2015 when it was filed in 2011. Re. the 'patent pending' on his Magtech regulated power supply - not finding any patent application or issued patent is a mystery indeed.... Like you wrote - maybe he never pursued a patent & the other press is now old & written words cannot be retracted today?
Thanks, Bombaywalla. Yes, a delay of a few years between application and grant would certainly be understandable. My reference to 2003 for the "encapsulated composite electrostatic loudspeaker stator" application you had found was correct, however. As can be seen on the left side of the page, a short way down from the top: "Filed: October 30, 2003."
Patent does not mean much (I have couple of them). Patenting new principle of operation is practically impossible, so the most of people patent specific use. You can come up with any weird stuff and register it. Patent office does not check validity of the idea. They only check if it was used before - no matter how stupid. The word "Patent" suggests originality.
Of course. SMPS actually has a lot of advantage, SMPS is basically a power regenerator. Go listen to the JBL LSR305/8, it’s an all digital design. SMPS, class D, digital crossover, DSP. It’s the greatest bargain in audio right now. The 3 series 5x305+310s simply destroys $1000 to $10,000+ audiophile setups.
There are also bad SMPS designs too, eg: Benchmark.