AudioCircle
Audio/Video Gear and Systems => Power Conditioning => Topic started by: Quiet Earth on 23 Aug 2014, 09:37 pm
-
I realize that the best SMPS is no SMPS, but many of us have that one piece of gear with a SMPS inside and we don't want to modify it. Maybe it's a dedicated streamer or a computer. Perhaps it's a universal disc player or CD burner. Whatever it is, there it is and how do we treat it when we connect it to the main power strip of our linear powered audio equipment?
Many of these SMPS equipped devices already have some pre-AC noise filtering built inside. So do we just plug the SMPS device straight in to the main power strip and leave it at that?
Would additional power conditioning dedicated to the SMPS device eliminate the switching noise that goes back into the main power strip?
And last but not least, can power conditioning actually improve the performance of a SMPS powered device and make it sound better?
Thanks for sharing your thoughts on this subject.
-
They benefit in a huge way. In fact it's almost the only way they're worth anything barring some very high end ones like Paul Hynes (which can probably benefit anyway).
-
Well some SMPSs can get very unhappy with some power conditioners.
Let me look for the info.
-
Well some SMPSs can get very unhappy with some power conditioners.
Let me look for the info.
Please do.
Many people years ago found that SqueezeBoxes stock power supply was very good but only with some conditioning, otherwise replacing it was sort of manditory for a decent source.
-
Well it's in the Henry W. Ott book 'Electromagnetic Compatibility Engineering' book. It's section '13.6 Power Supply Instability' and it's only 5 paragraphs long. But putting it online might be a trick.
http://www.hottconsultants.com/EMCE_book_files/emce_book.html
Mr. Ott references another book:
"Power Line Filter Design for Switched-Mode Power Supplies"
by Mark Nave see Chapter 6
-
It says that when a SMPS has a constant output, it can compensate for voltage loss by drawing more current (due to the switching nature). If the current draw is to much then the filter may become inefficient and cause even more of the problem.
This is only an issue if you use higher resistance power conditioners or improperly sized current devices. It may even be a likely problem with some junk sold by less than audiophile companies.
This is just another reason for using CMC's twice the size of the load (or more). Well that's what I do and recommend. And yet another reasoning to stay away from cheap low current junk like those little IEC "filters" for power inlet.
The book also discredits those things for a variety of reasons that don't even cover how they totally muck up ground and sometimes have ferrite trash in them.
Audience
Pi
Running Springs
Blue Circle
And myself are a few to name that don't experience this under any normal operation. You'd have to plug in an impeccably large (15-20a depending on power conditioner) or inappropriately sized device for the socket to have this issue.
This problem could be replicated by just having too small of wire or anything that limits current. Generally audiophile companies far exceed current need capability.
Personally I'd consider it a non-issue with any quality company. It's an engineering problem for those that build SMPS's that have input filters, and those who want to buy trash expecting non-trash results.
-
SMPS are the bane of a music lovers and they are everywhere. I posted a picture of the new receptacles with built-in USB chargers. Smart grids have telemetry powered by small SMPS. Coffee pots have SMPS in them... Like ChickenMan: "they're everywhere, they're everywhere".
They suck. They suck any semblance of true musicality out of reproduction because of the grungy haze and grit that gets passed by power supplies that are built to price points. A properly designed linear power supply is very expensive to manufacture.
One of the difficulties in designing a filter for a SMPS is that the differences in their input characteristics. They get very finicky when plugged into AC. A good example is the power supply in the unibody Mac Mini computers. When used as a music server you will find that it sounds very different depending upon the orientation of the IEC C7 plug into the back of the Mini. Why? I have no clue. I dislike SMPS so much that I have no interest in finding out why. They are banned from my system.
Some SMPS work very well with my BUSSes. Some do not. I have been trying to develop a one size fits all conditioner that works as well as the BUSSes do feeding transformer input supplies. I am convinced that there is no such animal. There are ways to mitigate the deleterious effects of the SMPS on the rest of the system and that is much more important than finding the best way to treat a particular SMPS.
Keep the EMI/RFI out of the system is imperative IME.
-
I have no problem with them and conditioning. But as I said maybe Paul Haynes would work, but as is I prefer linear (of my own design). They can still sound good and there's no other choice with the world's most expensive DAC...
I'm sure the Apples are just as simple as an either complete or not circuit. In fact one original orientation may eliminate a ground loop, as an example of potential. It depends how it's grounded internally and what it's connected to probably.
I've been toying with the idea of a linear for NC400s.... NO ONE thinks it's a good idea but perhaps it's the ears and not the brain that needs coercion.
-
There are ways to mitigate the deleterious effects of the SMPS on the rest of the system and that is much more important than finding the best way to treat a particular SMPS.
When you have some free time, perhaps you can list some of the ways to reduce the deleterious effects of SMPS on the rest of the system. That could be helpful for many of us.
Thanks!
:D
-
I'm sure the Apples are just as simple as an either complete or not circuit. In fact one original orientation may eliminate a ground loop, as an example of potential. It depends how it's grounded internally and what it's connected to probably.
:scratch:
-
When you have some free time, perhaps you can list some of the ways to reduce the deleterious effects of SMPS on the rest of the system. That could be helpful for many of us.
Thanks!
:D
Here is an easy one. Plug system SMPS into a small isolation transformer with an adequate capacitive input filter. There are a lot of small ISO TX available on the used/salvage market.
Having a home run to your system it makes EMI/RFI treatment much easier, too.
-
In my amateur and unprofessional opinion, I'd just make sure any cheap external switching power supplies are as much as practically possible kept on a different circuit from your nice high end linear power supplied gear. I've heard it myself muck up the sound of the other components unless it's on an isolated receptacle into a power conditioner, and even then I don't trust em. I make sure to keep all those devices on their own power strip plugged into a separate receptacle that's on a circuit separate from my audio dedicated lines. Thats my switching power supply isolation / containment policy......
-
Here is an easy one. Plug system SMPS into a small isolation transformer with an adequate capacitive input filter. There are a lot of small ISO TX available on the used/salvage market.
Having a home run to your system it makes EMI/RFI treatment much easier, too.
Meh, they don't offer a benefit to the SMPS itself. But I guess it's one way to deal.
The Apples orientation could mean a connection to safety ground or not. (In your service panel)
-
Meh, they don't offer a benefit to the SMPS itself. But I guess it's one way to deal.
The Apples orientation could mean a connection to safety ground or not. (In your service panel)
The question to me was about how to mitigate the bad juju from a SMPS. Hence my answer.
An IEC C7 is a non polarized connector. There is no connection to safety ground except through the neutral which is normally bonded. I guess I am just missing what you are talking about. It's late and I have a head cold. Snot must have crowded out my brain.
-
My 2 Emotiva phono stages (XPS-1) have SMPS that I was always concerned with. However, I have plugged them into the outlets on the back of my preamps and they haven't been a problem, at least that I have noticed.
To be honest, I really don't know what a SMPS does or why some engineer would use it over a conventional power supply.....
-
I use two SMPS(each are isolated from other connections) after my DIY Power Conditioner (PC), for my Cable TV and AudioPC, and they works really well.
The other components works much better than if I don't use them in the PC, so the noise that get back to AC is very reduced or almost none...
If you can't get a LPS, try to buy a better SMPS, like the medical ones, with low ripple noise, and then make a good CLC filter to improve it even more, and the component that you want to feed will thank you for that... ;)
Currently, I'm removing a SMPS (http://www.fspgroupusa.com/ecommerce/fsp060-dbae1.html)(<150mV) for a Medical SMPS (http://www.deltapsu.com/FntProductsAction.do?method=openItemDetail&productId=13)(<22.8mV), with one more electrolytic cap 120uF400V before the transformer to improve the transient filter, and after the transformer a very good CLC to remove the rest between it and my AudioPC.
-
Like most things, it depends on the design of the smps. There are cheap, poorly designed units and properly designed smps'. Those from Hypex are very good and actually work best without anything between them and the outlet.
-
..........
-
Like most things, it depends on the design of the smps. There are cheap, poorly designed units and properly designed smps'. Those from Hypex are very good and actually work best without anything between them and the outlet.
And yet everyone sells their Hypex's and goes back to A/AB all the time. I'd argue they may not be much more intelligent than being a relic from the distortion wars with today's best learned lessons on quality engineering. You can quality engineer a paper cup, but you won't find them in an archeological dig hundreds of thousands of years from now like with ceramics.
Honestly I think SMPS's could serve as more efficient constant current devices, but not a lot else.
-
Yeah, you can tell everyone is selling them by the way all the audio classifieds are just jammed with them.... :lol:
Some of the most respected names in audio use them. It is, like many other things, about implementation. Of course there are still those confined by their out of date dogma...they can't be helped.
-
I am using a Chapter 400 watt Class D that will not work plugged into my Torus. It only works plugged into the
wall? It might have to do with the 16amp fuse, or it's adjustable grounding scheme. You can toggle between the chassis or the ground pin on xlr.
-
Yeah, you can tell everyone is selling them by the way all the audio classifieds are just jammed with them.... :lol:
Some of the most respected names in audio use them. It is, like many other things, about implementation. Of course there are still those confined by their out of date dogma...they can't be helped.
Like I said in my circle, Bruno is one smart guy. The SMPS used is a very, very good one and is part of the voicing for the amp. The only real issue that I have with it is the use of polyester coupling caps. That I'd pretty easy to fix, if a bit kludgey. I would say that if the builder does not follow the build instructions to the letter EMI "could" be a problem.
I have build several of them for customers and did a bit more in the noise mitigation mode and they sounded very good. I'm still a tube guy though. A dogmatic toobular lemming :lol:
-
To each his own. At least your opinion is based on experience...
-
To each his own. At least your opinion is based on experience...
Thanks. I have spent way too much money on this addiction 8). Slow learner, I guess.
-
Interesting thread here. There's a couple of tests we did ourselves to determine the best sonics of a Mac Mini into our DAC that support some of the posters here and their experiences with Switch Mode supplies:
Firstly, I'd say that its very hard to get better sound from the switch mode supply with any power conditioning. We've tested (in blind A/B tests) and tried quite a few different offerings by various respected high end manufacturers. To this day we haven't found any conditioners we would say - "use this one and your Mini will sound much better'. In fact, in our 25+ recommendations for tweaks with the Mini for our DAC customers to try, we've always told them to avoid any power conditioners entirely and plug their Mini directly into the wall. :nono:
There have been other non-digital components that I myself have owned over the years with SMPS designed within. That design choice always seemed to be the weakest part of the design (once I started tweaking, changing and improving the internals). From a high end manufacturer's view these SMPS are very cheap as all get out as compared to linear, so I guess it makes good business sense to design them in? As for making an outstanding sounding product though, I myself would never consider using SMPS. The downward sonic detriment is just too great IMHO.
As to other direct analysis of the SMPS and what happens when it is removed or replaced in a Mini. We now feel it's better to completely remove and replace the Mac Mini's switch mode supply with direct battery power or a linear power supply. This removal / replacement of the SMPS has been fairly popular with Mini modifiers beyond just my own company's recommendation mind you and seems to be universally accepted as a major improvement sonically by those audiophiles who have done it to their Mini. Hence, we are not alone with our findings :green:
Another tweak we call out for our customers is the "try both plug orientations on your Mini's power cord, turning it 180 degrees, one way will sound better". This surprised many a customer and even more technical guys when I told them to try it. I can only suspect that the internal ground noise must be finding a better path to the neutral side of the power in a customer's outlet than the other way. Perhaps its something different entirely? Regardless, it remains an eye opener for every Mini owner who tries it...at least with our DAC hooked up to the Mini mind you :thumb:
Cheers to the people with ears!
Eric Hider - dB Audio Labs www.dbaudiolabs.com
-
Using a SMPS after the Power Conditioner only helps to prevent that noise go back to AC and contaminate the other components...
The Power Conditioner can't do nothing about noise after SMPS...
-
I'm sorry but I believe you're both wrong on this, somewhat, db and rdsu.
First off it's more likely that the orientation of the Mac Mini's plug is breaking a connection from neutral to ground at the service panel, from making a connection through your system. You've eliminated multiple paths at that point, to anything you're connected to. Perhaps I'm wrong and it's 100% internal to the Mac Mini's grounding scheme causing a loop, but generally speaking I've seen computers making loops with other devices pretty indiscriminately.
Power conditioners can actually lower the noise in a SMPS depending on it's design. There's some things happening with change of reactance in frequency ranges they operate in, and the ability to provide a shunt to some of the noise. Some SMPS units have weak, interfering, input filters, others do not.
Again this may be best seen originally with the Felix applied to the Squeezebox stock SMPS (junky). It didn't have to be on the same power distribution as the rest of the system to see the benefit.
Well, of course I assume a linear is the upgrade to have for a Mac Mini, despite that I think depending on the SMPS design it could benefit. I haven't cracked one open but I've got a guess, and it wouldn't surprise me if while fixing one thing it causes another. My apologies that I don't want to discuss it too much more, as I don't feel it's to my personal benefit as an Industry Participant, to share every least thing in the world. :lol:
-
Power conditioners can actually lower the noise in a SMPS depending on it's design. There's some things happening with change of reactance in frequency ranges they operate in, and the ability to provide a shunt to some of the noise. Some SMPS units have weak, interfering, input filters, others do not.
I fully agree! ;)
-
Mac Minis are 2 prong no ground. I think they charge the battery while plugged and run off the battery?
-
Your thinking MacBook (laptop). Mac mini are desktop and have no battery (other than bios).
-
Ok thanks. :thumb: :duh:
-
Mac Minis are 2 prong no ground. I think they charge the battery while plugged and run off the battery?
I made a terrible drawing of a path that shows how a Mac Mini can have a path taken out of the equation if the plug is reversed, if it's interacting with other equipment in this manner. You can see how this loop won't form if you switch hot/neutral. SMPS's vary on how this works; there's numerous wiring configurations. As you can see it might be possible to have a path on the neutral as well, depending on your other equipment.
(http://www.audiocircle.com/image.php?id=104531)
-
I made a terrible drawing of a path that shows how a Mac Mini can have a path taken out of the equation if the plug is reversed, if it's interacting with other equipment in this manner. You can see how this loop won't form if you switch hot/neutral. SMPS's vary on how this works; there's numerous wiring configurations. As you can see it might be possible to have a path on the neutral as well, depending on your other equipment.
(http://www.audiocircle.com/image.php?id=104531)
I don't want to start a holy war here, but I don't think you are misunderstanding our experience with the Minis and the successes we have with them. A lot of our experience deals with the Mini being used in a battery powered system... Yep, pre, amps and DAC. The Mini will exhibit the same polarity characteristics on an otherwise battery powered system.
Bottom line is that it is characteristic of those particular power supplies.
It is easy to talk about subjects with words like : might ; could ; perhaps ; depends and probably. This make for good discussion, but 4+ years of experience has to count for something.
-
Fat fingers
-
Really, on battery power? Hm.. well that's interesting. Seperate or same batteries?
-
Hey guys,
Everyone in our group here (doing true actual blind A/B testing) and now hundreds of our customers throughout the world (in 22 countries) are all experiencing the same dramatic sonic improvements with either a battery power replacement of the SMPS or the replacement of the SMPS linear supplies. Not a singular audiophile power conditioner that any of my customers have tried (thus far) matches the sonic improvement of the removal and replacement of the SMPS comparably. Not even close.
It certainly would make my life as an Audiophile Digital Provider much, much easier to point to some sort of conditioner than the alternative of recommending the complete removal of the SMPS and the inclusion of an internal harness for battery or linear power comparably. Are there any "Audiophile Conditioner Providers" watching this thread who have yet to respond? if you can create such a conditioner that magically improves the Mini's SMPS? I have hundreds of customers for you! Hint, hint :eyebrows:
I myself don't really have a concrete technical hypothesis to why the SMPS seems to suck sonically in the Mini. And thus far why no conditioners connected to make it sound any better. I can only speculate to the technical reasons. But at this point it is pure speculation for me. Perhaps someone that really understands the Mini's SMPS implementation could explain why the Mini doesn't seemingly respond positively to any power conditioning? Hmmm, anyone care to guess?
Cheers,
Eric Hider - dB Audio Labs www.dbaudiolabs.com
-
Why not use a parallel conditioner on the same outlet? What is the forseable improvement that you are talking about that isn't happening?
-
I currently have this amp (400watt) on borrow. http://www.chapteraudio.co.uk/products/reference/couplet-1000s/
Can some one give me a good answer why this amp will not play using my Torus rm20 240V? It will turn on but i can't get music. It works flawlessly plugged into the wall. It does have a toggle between the ground pin xlr or chassis ground. Can't get music on either toggle select, but turns on, lights up when plugged into my Torus. This appears to be a problem with all conditioners not just a Torus.
-
Last night was trying a Torus RM20 w/ a 300WPC Hegel 4HSE amp. It worked, but totally killed the dynamics (slow slow slow). Not qualified to try to explain why a big isolation transformer like that won't work (rated power is greater than the one in the Hegel). What I will say is that convention audiophile wisdom is to never plug power amps into power conditioners, and particularly not high power amps. Just chokes the power delivery too much.
So I guess the simple answer is : plug that amp into the wall!
-
Really, on battery power? Hm.. well that's interesting. Seperate or same batteries?
The Mac Mini is on a 55AH AGM through a BatteryBUSS. The Mini is a little noise generator and other gear is much happier with it on a separate source. Pre and power amp are on a 100AH with proprietary output impedance correction for the battery to get the output impedance very, very low and with noise reduction to get the sulfation noise out of the line. DACs will vary, but are generally battery powered. Just like running dedicated circuits for gear...
-
I'm not sure a power conditioner that works on one would be cost effective, compared to just replacing with a linear? How much does that cost?
-
I'm not sure a power conditioner that works on one would be cost effective, compared to just replacing with a linear? How much does that cost?
You are missing the point. Batteries done right sound better IME. I build very high current low impedance linear supplies, too. It is all about getting the best sound, not being in one camp or the other.
Prices are all on my Circle. This is getting too far off topic from the SMPS question, methinks.
Once again: problems with SMPS are:
They feed noise back into power lines
They are radio transmitters
They may or may not provide 'clean' power to the device
To each, his own.
-
Easy! I was referring to Eric who said he would like a product that doesn't require replacement of the PSU. I think your success with the BatteryBuss (and Uber) is pretty well know on the board, by the way.
Linear or batteries seems like the route to take. Modifying the stock SMPS or using a $5k conditioner seems excessive for someone looking for a simple solution.
-
You are missing the point. Batteries done right sound better IME. I build very high current low impedance linear supplies, too. It is all about getting the best sound, not being in one camp or the other.
Prices are all on my Circle. This is getting too far off topic from the SMPS question, methinks.
Once again: problems with SMPS are:
They feed noise back into power lines
They are radio transmitters
They may or may not provide 'clean' power to the device
To each, his own.
So noise, hash amplified in the high frequency range.... thats it?
-
So noise, hash amplified in the high frequency range.... thats it?
Not entirely, but a large portion of the noise spectrum is > 100K. Part of it it is the reflected EMI on the power lines. That plays hell with all of the other noise on the powerlines and creates harmonic distortion.
Common argument is that a "properly designed" and I might add: properly executed; should take care of that noise. Should but all too often don't. There is a lot of noise on the ground, too. Remember the bond at the panel.
Odd harmonic distortion = glare.
-
It reminds me of those super tweeters you put on top of your speakers. On their own typically not audible but when added to an amplified source they blend in.
With out reinventing it the only thing you can use is a good quality, long usb cord. Along with the Mac Mini plugged in (outside and away) from the system. In a different room even. Then run it off mpod from an ipad.
I have a 7200 rpm HD that i tried with my BDP. When plugged into my BDP using a stock usb cable and plugged into my conditioner (due to usb length) the results were horrific. It literally had the same sonic characteristic that of my dot matrix printer at work. That printer wails a mechanical treble with every letter hit. My system had that same treble clammer that rode over the entire soundstage. The similarity was really weird.
This was substantially fixed by a 7 meter Wireworld Usb cable. And further repaired by plugging it in off the stereo circuit. In the end self-powered 3.0 HDs were the answer.
-
It reminds me of those super tweeters you put on top of your speakers. On their own typically not audible but when added to an amplified source they blend in.
With out reinventing it the only thing you can use is a good quality, long usb cord. Along with the Mac Mini plugged in (outside and away) from the system. In a different room even. Then run it off mpod from an ipad.
Have you tried a USB transformer, or power splitter?
Super tweeters might be affecting a lot of the other frequencies, and that's what you hear. Hell if I can remember exactly what the formula is, but every so many octaves a frequency modifies another to a degree. The questions are if the effect is recorded, and if we need it reproduced in the air rather than when recorded in order to get the benefit, when I think about it.
But you'd be right in an interesting way. Rarely do you hear artifacts and noises in playback, but the effects of the noise on the music itself are readily obvious the minute you plug your gear into something that reduces it. Maybe they're more audible, maybe they are not, what's clear is when they're gone it's crazy how much it can be like thinking coffee doesn't get much than Starbucks, and then having a pour over that's $5 at some local place that get special beans from Kenya...
To be on topic! One of the biggest issues with SMPS's is that there frequency range is extremely susceptible to problems with induction. While you might have big fat cables, they often see the path as looking like higher resistance because they recreate a field that incurs this... they're going to voltage drop somewhere, but the longer the path the more of a problem. That's why even inside the SMPS they do their best to decouple these frequencies as close as possible to the switch device. This is true in class D as well. But there's limits to how much, and how well you can. Plus from one engineers perspective it probably looks fine on paper, because they aren't feeling it on their ear drums. *This doesn't mean extremely low induction cords are an optimal choice for all applications, in fact you can create an antenna.
In order for any power conditioner to work with a Mac Mini at all, the cheapest mod possible might be shortening the AC cable to a knub. A problem however if you look at many specs on certain conditioners is that they concentrate more on the mhz range, than the khz range.
-
In order for any power conditioner to work with a Mac Mini at all, the cheapest mod possible might be shortening the AC cable to a knub. A problem however if you look at many specs on certain conditioners is that they concentrate more on the mhz range, than the khz range.
That's the problem with most "conditioners". Even 100K is way too high to be really effective.
As far as AC power for the Mac, shortening the cable doesn't work IME. Been there, done that. It needs a 1.5m cable to sound its' best.
-
That's the problem with most "conditioners". Even 100K is way too high to be really effective.
What do you mean exactly? Like power conditioners need to affect a lower range?
-
What do you mean exactly? Like power conditioners need to affect a lower range?
Was I unclear? It was a statement, not an equivocation. I'm sorry, but you seem to ask questions on things that are pretty clear cut in posts.
Power treatment is a well done, many times over subject. I guess it all boils down to how far does one care to lower the filtration bandwidth. One of the things that separates power conditioners is that low-pass corner and how you get there.
-
Sorry, but in my mind it seemed like you split on thought into two, expressed as sentances. In my mind most conditioners don't filter low enough, so to say "too high" is an expression that maybe brought something with it I didn't get.
-
Sorry, but in my mind it seemed like you split on thought into two, expressed as sentances. In my mind most conditioners don't filter low enough, so to say "too high" is an expression that maybe brought something with it I didn't get.
It was an answer to your statement: " A problem however if you look at many specs on certain conditioners is that they concentrate more on the mhz range, than the khz range. "
I was agreeing with your statement... didn't mean to confuse the issue. I meant they start filtering at much too high a frequency to be really effective.
-
Have you tried a USB transformer, or power splitter?
But you'd be right in an interesting way. Rarely do you hear artifacts and noises in playback, but the effects of the noise on the music itself are readily obvious the minute you plug your gear into something that reduces it. Maybe they're more audible, maybe they are not, what's clear is when they're gone it's crazy how much it can be like thinking coffee doesn't get much than Starbucks, and then having a pour over that's $5 at some local place that get special beans from Kenya...
Don't do USB. I don't like it and i don't believe it to be suitable for audio. I play files but use Xlr or spdif.
An interesting thing regarding noise and distortion I've noticed. As soon as the noise and distortion has been reduced. The remaining THD+N is far more easier to hear. IOW it teaches you how to hear noise and distortion. Or like tasting and appreciating Kenyan coffee for the first time.
-
Do swithmode power supplies respond to line conditioners?
Based on my experiences limited to a number of class A/B amps and several class D amps all using superior line conditioners, my answer is absolutely yes. More importantly, they respond positively under certain conditions. Some of those conditions are related to the SMPS and some are not.
You also said, "I realize that the best SMPS is no SMPS".
I can see why some might think this but like anything, it depends on one's experiences and the extent to which one is willing to address potential problems. So I'd like to offer a few thoughts on SMPS in general.
1. SMPS is digital and therefore by their very nature anything digital will induce a small audible amount of bi-directional digital noise back into the AC outlet. If other components are sharing that same circuit, they will be contaminated with that digital noise. Not the end of the world but it is audible and I presume the more components sharing that same outlet, the worse the amount of digital noise.
2. To the best of my knowledge, most/all SMPS' are mounted on cheap PCB boards and then the PCB board is mounted to the chassis. Unlike traditional power supplies that a heavier, more stout, ridid, etc. that are often times mounted directly to the chassis or mounted to a rigid plate that is mounted to the chassis. This implies that any vibrations induced by the SMPS itself, or from a neighboring source within the PCB board and/or chassis, or simply any air-borne vibrations captured elsewhere in the chassis can have a potentially greater effect on the SMPS because such vibrations become more trapped in/around the SMPS. IOW, if any vibrations in/around the SMPS are not free to travel elsewhere, they will begin their lengthy dissipation right there. IOW, cheap, thin, fragile PCB board material is problaby far worse than MDF which is a horrible conductor for allowing mechanical energy to travel.
That said, line conditioning theoretically has more to do with universal noisy AC inducing its harm (distortions) on anything plugged into the wall. Whether it be a refrig, an electric oven, an audio component, a traditional PS, or SMPS. Since most of us are not anal about the performance of our refrigerators and ovens the distortions induced by noisy AC there go unnoticed.
However, in a well-thought-out playback system it is easily noticeable. But then again, it can only be easily noticeable if one is able to hear that playback system with and without proper line conditioning. Proper being the keyword here, because there are plenty of line conditioners that do little or nothing, or worse will induce their own sonic harm.
Which leads to the point of my answer to your question. If, per chance, you're able to employ line conditioners worth owning, the level of improved musicality should be pretty tremendous, whether traditional power supplies or SMPS.
But you still have the added problem of bi-directional digital noise generated by the SMPS that still will go back out into the AC outlet and infiltrate your other components. Therefore, you either need to ensure your line conditioner of choice is capable of bi-directional noise filtering, and/or you need to ensure every component has its own proper line conditioner.
Some may say that dedicated circuits/lines will cure/prevent this bi-directional noise from infiltrating the other components, but there are enough in-the-know who also will claim that bi-directional digital noise is capalbe of going all the way back to the AC service panel and then infiltrate other circuits/lines and other components.
There are also two other aspect to keep in mind with class D stereo amps and class D int. amps:
1) If there is only one AC inlet to the stereo or int. amp, this bi-directional noise will be shared between the L&R channels at the single AC inlet and there's not a external line conditioner in the world that can cure this internal problem.
2) Similar problems occur with an int. amp and the active pre-amp section. The active preamp section must borrow AC power from one of the 2 channels to function. Since both channels are already generating bi-directional digitial noise, that will carry forward to the preamp section of the int. amp.
Having experienced these situations myself, knowing what I think I know and the tremendous audible improvement potentials, I would not own a stereo class D amp, nor a stereo class D int. amp. But I do own a pair of class D monoblock amps and I have some fabulous performance-oriented dedicated, passive, and bi-directional filtering line conditioners on every component. And just like my previous class A/B amps and int. amps, my class D amps benefit equally from proper line conditioning.
Finally, there's the issue of potentially more inferior / less proper vibration control with SMPS modules, but that's outside the scope of this thread.
With regard to your statement about the best SMPS is no SMPS, I recently swapped out a highly rated $8k class A/B int. amp for a pair of small mono block class D $2.4k amps and with careful consideration and solutions to their known issues, that highly-rated $8k int. amp now sounds like a $200 BestBuy receiver in comparison to the level of musicllity I'm now able to achieve via the class D mono blocks.
I've also owned other class D amps from 2006 - 2011 with not too dissimilar experiences and successes but recent revelations have allowed me to take class D much further this time around.
In sum, SMPS technology brings a bit more baggage that needs to be considered, analyzed, and addresssed and if they are properly addressed, there are potentially huge musical gains to be had. Yes, I said huge as in larger than large.
-
SMPS's are switching devices and not any more digital than class D amplifier that's not digital at all.
You could say because they operate on a on/off switching they're like digital 010101, except that class D creates a sign wave, and SMPS's don't send data, just a 0-1hz DC.
However they use very small transformers so they're much more likely to make common mode noise, which I discussed some in another post you've read. But also the power supplies tend to be pretty low impedance, which is good and bad. The funny thing is on the output you get filtration that typically starts at the input, but some tend to lack the same amount of filtration on the return paths; if any at all. It's relatively easy for noise to leave the device, in other words. This is very simplified, and not entirely universal. Basically AC (noise) seeks a low impedance path, and SMPS's may provide one that isn't through the filtration.
-
Regarding SMPS with a conditioner, I will give a different take. An SMPS will sound better with a power conditioner, but only if the SMPS is the only device connected to it.
When you have a full system connected, any SMPS individual gains are crossed out by the SMPS itself feeding hash and noise back into the local system that effects other components.
That is why you are always better off having the SMPS isolated and in the wall.
Better yet get upgraded LPS in place of SMPS, it makes a vast difference.
-
Powervar Corp. claims the best way to condition SMPS is by using a low impedance isolation transformer, with built in noise filtering. They claim the main factor is the isolation transformer. Makes sense to me.
You could use a power strip for convenience.
I see they make a new series called "Ground Guard". I may get one soon, and they are available in several sizes.