Jump to content

Lengthy Primer on AC - with some thoughts on reducing the computer's impact on yours

Recommended Posts

This was originally posted in this thread: http://www.computeraudiophile.com/content/Mac-Mini-Linear-Power-Supply


I've re-posted it here in the event that a broader audience might be interested.


As I said in the earlier thread, I reached out to the person I go to for all things AC - a former poster here who shall remain anonymous per his request.


Here is what he had to say:


"After thinking for a while about a response, I realized that doing even a moderately complete job would require a substantial chapter in a substantial book. I know this because some experts have done just this and it took that much space. And that was in books targeted toward electrical engineers with some years of experience. For an audience primarily of people who are not electrical engineers, or better yet physicists, it would take much, much more than what anybody would want to read in some hobbyist forum. More than I’d want to type, too.


So, this will be as simple as I can make it, hoping that people will spend time researching ideas that they aren’t familiar with and knowing that the subject will barely be scratched.


This will also be as apolitical as I can make it, with one exception. You can easily find billions and billions of posts on various internet forums about these audio topics. Many of these are from highly educated and smart engineers who wave their arms and proclaim that none of this is really important. Here’s the problem: From the beginning, engineers (and some other professionals – you know who you are) are trained to make various simplifying assumptions about any problem. This is done not only to streamline the educational process, but also to make really complicated problems understandable for us puny humans. But, that doesn’t mean that the details that were ignored actually go away. In the end, they may not really be significant after all. In many cases they are not. Maybe just as often the details are significant. If you wanted to calculate the planetary motion of the Earth around the Sun, you might not include the effects of some piece of rock the size of a Buick that is floating around somewhere beyond the Moon. It probably wouldn’t make any meaningful difference in your calculation. But if that rock decided to land in your neighborhood one evening, you might find the effects really significant to you.


The power system in your house is really, really complicated. It may not look it, but it is. It is not only the delivery mechanism for 60 Hz power (50 Hz in many parts of the world), but also is an unwitting connection system between everything plugged into it.


For now, we’ll entirely ignore the radiative aspects of the home power system. By those I mean radiation of noise as well as capturing various undesirable signals that are already floating about, like radio signals, wireless signals, TV, and so on. These are real bona fide conditions that can be major PITA problems, but that goes way beyond Mac Mini power supplies.


Think about what happens in your house, after the breaker or fuse box. (Lots of badness takes place outside the house, but let’s skip this one, too.) You have cables of various lengths that often have non-linear loads attached along their way at various points. What’s a non-linear load? Anything that distorts the waveform of the 50/60 Hz voltage waveform. Unfortunately, almost every load in a modern home aside from incandescent lamps and electric heat is non-linear. If DC is used in the load, the AC must be converted to DC. This is usually a non-linear process. More on this in a bit.


Non-linear loads draw current in some way different from a pure sine wave. This causes harmonics of the line voltage to be placed on the line as well as various artifacts that are nominally secondary to the actual power process.


The reason these harmonics and artifacts are important is because audio equipment is imperfect with regard to “ignoring” impure AC supply voltages. The simple harmonics of the line frequency usually are not so much of a problem, since similar harmonics are generated within the audio equipment and are normally considered in the design and execution of the product. The ones that really are a problem are normally between 50 KHz and 5 MHz. Since these are outside of what is normally considered the frequency range of what people can hear and loudspeakers can reproduce, why do we care?


Audio amplifiers and processing equipment, even those “competently designed and manufactured so that they all sound the same,” often respond to higher frequencies. Yup – it’s true. Not only do they respond linearly at these frequencies, they also often respond non-linearly to higher frequencies. Even look at the data sheets for the most popular opamps. As a simple example, notice how even some simple amplifiers can detect and play back radio signals, sometimes even quite loudly. This is one of those situations where you’ll need to do your investigations as to why this may be – it’s a long explanation. For the moment, please just go along and accept that this is true.


The non-linearity of the amplifiers at these frequencies can cause problems in the audio band, even if the amplifier is essentially perfectly linear within the audio band itself. Not that any amplifier really is perfectly linear… For example, imagine that somehow two high frequency signals manage to find their way into your amplifier. Let’s assume that these two signals are at 211 and 212 KHz. Way outside the audio band, right? Well, if applied to an amplifier that isn’t especially linear at ~200 KHz, you’ll get harmonics of the 211 and 212 KHz signals at the output, which you can’t hear. You’ll also get an intermodulation product at the sum of the two (423 KHz) as well as at the difference of the two – 1 KHz. Even I can hear that – you probably can, too. That’s not even thinking about the higher order IMD products that fall into audio range as well as out. Nor is it considering the effect of these signals on the operating conditions of the devices within the audio band. Conceptually, it’s easy to imagine that if for some horrible reason there was a signal at 1 MHz that was driving the amplifier to clipping at that frequency, you couldn’t hear the 1 MHz, but that clipping certainly would affect the performance within the normal audio band.


Of course, that was really a highly simplistic example. It’s more likely that the high frequency signals will not be simple sine waves, but rather complex signals that are likely additionally modulated by other complex waveforms and noise. The products of these that fall into the audio band are equally ugly and can sound like modulation of the noise floor or other nasty garbage that your auditory system will have a hard time processing. Simple harmonic structures occur in nature and even in your ears, so harmonics of the music are the least of your worries.


So, back to the power supply.


In a “linear” supply, the rectifier diodes are commutated (look it up) and the output smoothed and filtered to make it a close approximation of DC. Sadly, most diodes are not perfect and can produce outputs at several hundred KHz (tube rectifiers, imperfect in other ways, don’t do this). These high frequency signals not only can travel forward into your amplifier stages but also backward out of the power supply through the power cord and back into the power system in your house. How nice – they can now get into every other electrical appliance plugged into your power system. The type of smoothing filter affects the level of these as well as resonances that are found within the system. That’s partially why choke input filters have a good reputation in some circles. Toroid power transformers are inferior in this regard compared to split bobbin transformers, since the windings are right atop each other in a toroid and there is a lot of capacitance between the primary and secondary windings. Those are just two of many details that affect transmitting high frequency garbage both out of and into a piece of audio equipment plugged into the power mains. Or any other applicance for that matter.


An additional concern with a linear supply is any regulation that is used. In many applications it’s desirable to maintain constant voltage at many points in a circuit. A linear regulator does this fundamentally by dissipating any unused power that is due to the wrong voltage as heat. Basically, closed loop linear regulators are servo controlled resistive voltage dividers. The servo watches the output voltage wherever the designer chooses and adjusts the voltage drop across one of the regulator elements to maintain the desired voltage at the sense point. If you were really diligent and had really fast fingers, you could do the same yourself manually by turning the knob on a potentiometer. Obviously, that doesn’t work because of the speeds and resolutions involved. But the same principles apply as if you were doing it by hand. Making these adjustments at signal speeds makes the regulator act as a filter. The faster the regulator, the better the filtering at high frequencies. Most, but not all, regulators don’t offer much rejection for the range of 50 KHz to 5 MHz. In fact, since most regulators use the same types of servo amplifiers as found in audio amplifiers and are susceptible to high frequency interference. Whoops. (Some regulators make use of some circuitry or component properties to maintain voltage in an open loop application. These are usually less susceptible to outside interference.)


Switching regulated supplies are a little different from linear supplies. Switchers are used for one simple reason: Efficiency. The way a switching regulator works is that it creates a high frequency AC signal from a DC source, which is then rectified and filtered. Yup – you might be using two rectifier systems in series if the switching regulator is powered from the AC mains. The switcher adjusts some aspect of the high frequency AC waveform to vary the DC output voltage from its filter. It uses the same kind of servo action as the linear regulator does. Since the frequency of the AC output from the switcher is high – usually somewhere between 50 KHz and 5 MHz (sound familiar?) the components needed to nominally filter the DC are smaller than what is needed for filtering rectified AC mains frequencies. Lower value caps are needed as well as smaller inductors. That is space efficient as well as money efficient. The waveform used is made such that the diode rectification process is made to be as efficient as possible electrically. That makes for less heat, better power usage, as well as a smaller package. Since a transformer is often not needed, that also reduces cost and size, as well as offering a side benefit of allowing the supply to be used in many countries where the AC mains can be 50 or 60 Hz and anywhere from 100 to 250 VAC. The bad thing is that the high frequency stuff can blow right through your power system into your audio components, where they may have an effect that you don’t appreciate.


In the case of the Mac Mini, there is an additional consideration. The external DC supply (note that new Mac Minis have built-in supplies) only produces a single output voltage of somewhere around 18 VDC. Inside the Mini there’s a bunch of other switching regulators that provide the actual operating voltage for the various internals - none of any consequence actually run from 18 VDC. So the artifacts from this process can easily migrate out of the Mini and through the external DC supply to your AC mains. That’s not to mention all the switching noise from the actual processing components doing their job…


What do you do?


One approach is to build a suitable linear DC supply that, if properly done, eliminates the first level of problems. If done really well, the switching noise from within the Mini back into the AC mains will be minimized and filtered, too.


A second approach is to filter the AC power going into and coming out of the raw DC switching supply for the computer. There’s numerous commercial units available for this. Unfortunately, most of the ones I’ve measured have limited attenuation for between 100 KHz and 2 MHz where most of the power supply garbage is located. Some of the ones that are OK for common mode signals (look it up) aren’t great for differential mode signals. In addition, many, many use filters that try to bypass common mode noise to the safety ground in your electrical wiring. Unfortunately, at least in the United States, that safety ground is connected directly to the “neutral” wire of the system back at the breaker/fuse box. For high frequency signals where the wiring in the wall is a substantial fraction of a wavelength, that may be OK. But for the 100 KHz to 2 MHz region, that effectively can short out and negate part of the filtering. (Investigate: Balanced Power) That’s why some manufacturers use what’s called doubly isolated AC wiring so that no safety ground is used. In many countries, the safety ground plain does not exist. Read on your own the reasons given for using a safety ground – it’s not for noise filtering. Most of the general purpose AC mains filters are designed to minimize conduction of frequencies well above 1 MHz, so that they don’t get onto the AC mains system, radiate, and interfere with radio and other wireless devices. In addition, making filters that really are good for between 50 KHz and 5 MHz often requires pretty large components or circuit tricks that require large components. Even most of the AC filter guys who target the audio market don’t want to go there. For one thing, there’s a whole dogma that includes the idea that anything in series with the AC mains will wreck the performance (see below). Plus, it just plain gets large, heavy, and expensive and that isn’t what most customers want.


So, again, what do you do?


Since the AC system in your house is pretty much guaranteed to be different electrically from everyone else’s, you need to experiment. The close to infinite number of variations of what is plugged in where and all that makes a pure analytical approach pretty hard. One thing that can be done is to try some of the products used to change the impedance characteristics of your AC mains system at various points. As just one of many examples already mentioned, Alan Maher makes products that attempt to minimize voltage peaks of noise where you don’t want them. Another approach is filtering the AC mains. This includes AC power cords, which really do act as lossy cables at high frequencies because of the dielectric used and the configuration. Or, you can use a cleaner DC supply for your Mini, even including batteries. Done right, batteries may be the best performance solution for everything audio. I mean everything. They just are so tough to manage in so many ways that most consumers just don’t want anything to do with them. Not many manufacturers want to build equipment they can’t sell.


Don’t forget, it isn’t just the Mini. If you have an external disc drive that has its own switching supply, that can be a problem. Same for your TV. Or dimmer controlled lamp. Or the microprocessor controlled dishwasher in the next room.


As I said, this is complicated.


One more specific point about how AC filtering can affect dynamics and all that… Even if you have a perfect AC filter in every way, the problem may be the amplifier itself. I have found many amplifiers that are “voiced” to take advantage of the power supply artifacts to give the impression of increased dynamics. It’s an illusion, but if it works for you and you find it more realistic, great! The point is that if you remove these artifacts by cleaning up the AC or the DC by using batteries, some or all of that thrill can go away. Keep in mind that in many audio products, the signal current goes right through the power supply with limited isolation. That means that any crap in the supply will somehow affect the desired audio signal.


I told you this is complicated."


Hopefully this will help some of us (begin to) understand what we're dealing with.






PS, Chris, if it's unwieldy to have this much info in two places, please feel free to delete one and direct traffic to the other.




Link to comment

Hi Clay - Thanks for posting this over here in a new thread. I think more people will find what they are looking for (if this is what they are looking for) in this thread instead of one about the Mini.


Thanks also for posting this info in general. Very cool to read and soak up the information as another data point in an ocean of data points. Not all are created equal. This data seems a bit better than most :~)





"Toroid power transformers are inferior in this regard compared to split bobbin transformers, since the windings are right a top each other in a toroid and there is a lot of capacitance between the primary and secondary windings."


This is very similar to what a very well known engineer at Spectral Audio told me. Toriods are pumped up by manufacturers who you them to be the best thing since sliced bread and made huge as a selling point. However, the more people know about them the more they don't like them.


Also, running a MacBook Pro from battery with a bus powered hard drive is looking like a better and better option every day.


Founder of Audiophile Style | My Audio Systems

Link to comment


Please thank your friend for taking the time to type this up. He is spot on in addressing some of the many problems associated with AC power, switching supplies, and even mentions the relative lack of mains filtering in toroidal transformers (one of my pet peeves, so many audiophiles think toroids are a good "feature")!

Perhaps Chris can have this published somewhere on the site as a reference, maybe in FAQs with your friends permission, Chris?



SO/ROON/HQPe: DSD 256-Sonore opticalModuleDeluxe-Signature Rendu optical--Bricasti M3 DAC--DIY Purifi Amplifier-Focus Audio FS888 speakers-JL E 112 sub-Nordost Tyr USB, DIY EventHorizon AC cables, Iconoclast XLR & speaker cables, Synergistic Purple Fuses, Spacetime system clarifiers.  ISOAcoustics Oreas footers.                                                       

                                                                                           SONORE computer audio

Link to comment





Good to hear that this info is in sync with your thoughts, as I consider your knowledge & opinions on the subject of AC to be very high quality as well - not that I'm in any position to judge anyone's knowledge on this topic necessarily. :)





Link to comment

That was a great primer on ac effects Clay, Thank You for the time and thought put into it. It's making me rethink how I want to implement my server setup. And it showed me how I still have much to learn ;)




Ryzen 7 2700 PC Server, NUC7CJYH w. 4G Apacer RAM as Renderer/LPS 1.2 - IsoRegen/LPS-1/.2 - Singxer SU-1/LPS1.2 - Holo Spring Level 3 DAC - LTA MicroZOTL MZ2 - Modwright KWA 150 Signature Amp - Tidal Audio Piano's.  


Link to comment

Nice post !

Thanks Clay, and to your friend of course.


My findings so far are that the netbook serving the music from the battery supply is the best sounding option. Far better than a desktop solution that really sucks IMHO.


But it seems not all laptops are created equal. The asus netbook sounds really refined when the dell laptop seriously damages my ears.


Considering the number of variables here, this is down to luck when picking one equipment over another one.


I'll be trying power cables with these Bybee thingies very soon. Don't really know what to expect, apart from bankruptcy ;)


I've just bought a Silent Wire power cable (90€) to connect from the wall to the ramp. OMG the system is completely transformed, and I pick my words carefully. This will be a good comparison point I guess.


I realize I am being a bit off-topic now speaking of power cables (and the danger that it represents here nowadays), so I'll stop here.


Again, thanks,



Link to comment

Thanks for the article Clay. Your efforts are appreciated.


Wavelength Silver Crimson/Denominator USB DAC, Levinson 32/33H, Synergistic Research Cables and AC cables, Shunyata Hydra V-Ray II with King Cobra CX cable, Wilson Sasha WP speakers with Wilson Watch Dog Sub. Basis Debut V Vacuum turntable/ Grahm Phantom/Koetsu Jade Platinum. MacBook Pro 17\" 2.3GHz Quad Core i7, 8GB RAM, Pure Music, Decibel, Fidelia, AudioQuest Diamond USB Cable.

Link to comment

Thanks for this post Clay. Your thoughts are definitely well "grounded" in respect to AC quality. This is something I overlooked for a long time as well as the majority of computer audio users out there. Most people think "I don't have any hum or buzz so everything is good, I've got clean electricity"; wrong! It's really incredible what improvements can be had by addressing this deal. I'm a huge fan of the A. Maher gear as well. I'll be rewiring my audio outlets here soon and installing a new ground rod with a less than 5 ohm resistance and using star grounding. It's alot of fun playing around with various approaches and finding what makes a substantial difference in your enjoyment of the sound.


david is hear[br]http://www.tuniverse.tv

Link to comment

Dirty AC is imo an important source of audible pollution. I've played around with regenerators, which opened my eyes to AC noise. I once had several Elgar conditioners (a halfway regenerator on which Accuphase conditioners are modeled) and heard improvements adding Elgars in series ... to three in total. Wow, was my response: that delicate thing called the audio signal is sensitively vulnerable to modification by very low-level noise.


Link to comment

Interesting read.

Here in England an “audiophile” power conditioning unit capable of supplying a Kilo Watt of continuous output runs into hundreds of pounds.

At the last place I lived I had a dedicated mains supply with a local ground which I put in and then had checked by an electrician at a cost of under £100 including materials.

Depending on how your home is wired there is another option and that is if there is a separate supply from the inlet fuse box for each room or even each floor you can disconnect all but one socket from the least used ring and use that for a dedicated supply. All one needs to do is wire the disconnected sockets to an alternative ring and run a 10 or 12 gauge ground wire from your dedicated socket to an outside ear thing point; a thick copper pole with the centre drilled out to a couple of inches depth at just above the ground wire diameter and then filled with braising rod works well. Each ring will feed a certain number of sockets and allow you to draw a total current value. As long as you don’t exceed this it’s perfectly safe.




I must admit I’m a fairly surprised that you guys with the expensive kit don’t have a dedicated supply installed as a matter of course.


Despite some of the stuff I’ve read which hasn’t always been very accurate I find a UPS system has a lot going for it.

I use this, the 450VA version. I got it for nothing because the batteries had died. New batteries cost me about £30.



Apart from the conditioning aspect of a good UPS it has the advantage of having at least 2 batteries and an inbuilt charger. If you really want to know if having all your equipment run on battery power makes any difference this is a cheap way of finding out.

Here in England you can pick a UPS up for next to nothing. For some unknown reason when companies relocate they tend to throw the old supplies out and buy new.


After that I have a fairly simple self built distribution conditioning block that uses round pin sockets that accept the old style solid brass round pin plugs. All my power leads have been replaced with inexpensive but good quality double screened power cable with 1.5 mm cores.

Even if your kit draws heavy current 2.5 mm cored wire with foil and braid screen isn’t expensive. If you’re paranoid then ferrite rings on all the component cables can be added.


Can I hear the difference? Ah, well I couldn’t swear to it either way, but I’ve got it just in case ;)

The point is, my “just in case” cost not very much. It seems some contributors here “just in case” might run into thousands of dollars. Just think, you could probably buy a dac upgrade “just in case” for that amount of money ;)



Dedicated Mains Cond dis block. Custom Linux Voyage MPD server. HRT Music Streamer Pro, Linear mains powered ADUM Belkin Gold USB cable. TP Buffalo 11, Custom XLR interconnects/Belkin Silver Series RCA. Exposure 21RC Pre, Super 18 Power (recap & modified). Modded World Audio HD83 HP amp.Van de Hull hybrid air lock speaker cables. Custom 3 way Monitors,Volt 250 bass&ABR, Scanspeak 13M8621Mid & D2905/9300Hi. HD595 cans.[br]2)Quantum Elec based active system self built.

Link to comment

Far better than a desktop solution that really sucks IMHO.


Pretty sweeping statement. Care to qualify it?


I've done major troubleshooting, just in the "noise" dept. and when I popped a Lynx card into a Dell, the resulting noise floor was quieter than two different low-power mini-ITX solutions with low power PSUs (not linears).


I'd like to hear some comment about high bitrate playback on low powered cpu systems vs. quad-cores. there is no doubt we are sometimes pushing these PCs IF YOU ARE STREAMING (JRMC Library Server) which I think is the future of light-weight clients.


Striking a balancing act may be more prudent than jumping on the low-power linear shp imo.


I have talked to some very experienced audiophile DIYs and heard that batteries can still have noise, can still not give the life to the music that direct power can.


This is the realm of "lifting the veil" because we aren't about a set of 103db/w speakers suddenly going QUIET because you hook up a linear. I hooked up my EP15 regenerator and I still had to deal with grounding issues and loops. I'm confused about "noise" vs. possibly affecting overall fidelity; are these power tweaks aimed at lifting payback to some perceived and abstract ultimate level, or are we talking about eliminating traces of observed, however minute, audible noise at the driver?????????????????


DIGITAL: Windows 7 x64 JRMC19 >Adnaco S3B fiber over USB (battery power)> Auralic Vega > Tortuga LDR custom LPSU > Zu Union Cubes + Deep Hemp Sub


ANALOG: PTP Audio Solid 9 > Audiomods Series V > Audio Technica Art-7 MC > Allnic H1201 > Tortuga LDR > Zu Union Cubes + Deep Hemp Sub


ACCESSORIES: PlatterSpeed, BlackCat cables, Antipodes Cables, Huffman Cables, Feickert Protracter, OMA Graphite mat, JRemote

Link to comment

I would strongly advise not to use regular UPS devices on a high end audio system. These units are made without regard to quality of their output, and are generally just made to keep critical computer system components running in the event of a power failure. Here are some of the problems inherent in typical UPS devices:

In AC mode:

Poor quality wiring and connections, which will introduce more noise on the AC line, and will raise the AC line's impedance (resulting in a drop in system dynamics). Some UPS also run in an always on mode, where the AC output is always generated by the switching output stage; if this is the case, then all of the below problems will apply all the time.

In battery mode (this is where it really gets bad):

Not enough battery capacity to achieve low output impedance, resulting in poor system dynamics.

Most UPS devices use a switching output stage, built with very little regard to noise performance-so running from a UPS, on battery power, has much more noise than plugging directly into the wall with no power conditioning at all.

Most UPS devices output a square wave, rahter than a sin wave, this means thta they have huge amounts of harmonic distotion on their output. If you must use a UPS, definitely get one with "true sin wave" output (more expensive) but note that the output is still highly distorted, with typically around >20% distortion, still way too many harmonics to expect decent performance from an audio system.


There are very expensive lab grade UPS supplies that can avoid these problems, but one will find that these UPS cost more than the most expensive audiophile products.


SO/ROON/HQPe: DSD 256-Sonore opticalModuleDeluxe-Signature Rendu optical--Bricasti M3 DAC--DIY Purifi Amplifier-Focus Audio FS888 speakers-JL E 112 sub-Nordost Tyr USB, DIY EventHorizon AC cables, Iconoclast XLR & speaker cables, Synergistic Purple Fuses, Spacetime system clarifiers.  ISOAcoustics Oreas footers.                                                       

                                                                                           SONORE computer audio

Link to comment

What to do... some advice (in order of effectiveness):


1. Run a dedicated circuit, in the US this should be 20 amp. You are better off running your entire system from one circuit if you can (this avoids ground loop issues), but if your system might peak out above 15 amps, run two identical circuits.

2. Use AC conditioning gear. This needs to be selected carefully for your system, as different systems (and AC supply) will respond differently. My favorite solution is to use high quality regeneration (stay away from switching regenerators) for source components, and high quality passive filters for amplification. I avoid transformer isolation as these can raise the impedance of the AC line, and add distortion to the line.

3. Use power cables that are well sheilded, to avoid noise pickup and broadcast through the cables.

4. Eliminate noise causing components in the home: light dimmers are one of the worst culprits unless they are switched off. Cordless phones, computers/TVs (plug these into powerful series filters, like those from PS Audio, Panamax, and Monster). If you can, try and keep high current draw appliances on the opposite phase of your AC supply (fridge, dishwasher, washer/dryer).

5. To lower the overall noise floor of the home experiment with things like PS Audio Noise Harvesters, and Alan Maher products. (these can distributed around the home at noisy parts of your AC wiring, note that the led on the Noise Harvester gives feedback as to where they may be most affective).

6. Choose components that try and address AC noise problems internally. Many designs may have EI or R-core style transformers, these are better at rejecting incoming noise than toroids. Many designs also include some kind of AC filtering circuit internally.



SO/ROON/HQPe: DSD 256-Sonore opticalModuleDeluxe-Signature Rendu optical--Bricasti M3 DAC--DIY Purifi Amplifier-Focus Audio FS888 speakers-JL E 112 sub-Nordost Tyr USB, DIY EventHorizon AC cables, Iconoclast XLR & speaker cables, Synergistic Purple Fuses, Spacetime system clarifiers.  ISOAcoustics Oreas footers.                                                       

                                                                                           SONORE computer audio

Link to comment

"Use AC conditioning gear. This needs to be selected carefully for your system, as different systems (and AC supply) will respond differently. My favorite solution is to use high quality regeneration (stay away from switching regenerators) for source components, and high quality passive filters for amplification. I avoid transformer isolation as these can raise the impedance of the AC line, and add distortion to the line."


Which regenerators do you recommend, specifically?


And what is "transformer isolation"?


Link to comment

For regenerators, I prefer the older versions of the PS Audio units. Because I recommend these for source components only, the P-300 and P-500 will have adequate output for solid state source components. Rick Cullen can do mods to both these units to make them even better. Buying a used one and sending it to Rick is a great way to go (he can make sure it is working properly, as well as do the mods as he was the builder for all of these units). I have not used them, but the Accuphase units look like they might be very good as well (but quite spendy, and they are not quite full regenerators as I understand them). The new PS units (Power Plant Premier) are also good for those who need more power, but for low power sources (where I like regenerators) the old designs, while innefficient and heavy, have slightly cleaner output, and offer balanced power. For best performance, true regenerators should be run with plenty of headroom (they are, essentially, amplifiers after all); I like loads no more than 100-130 watts on the P-300/P-500. As regenerators are active components, they respond well to mods: capacitor upgrades, opamp upgrades, wiring and outlet upgrades can all lower their noise floor even further than stock.

Transformer based power conditioning is offered by quite a few companies, many people report good results with some of these devices, but my personal point of view is skeptical. Transformers add impedance to the AC line, and they add distortion, while they can filter out some noise, very good ones are heavy and very expensive. Some people recommend isolation transformers built for industrial use (hospitals)-while these may provide AC isolation, they are not built to provide the fast, high peak, current demands of audio systems (as no one in hospitals complains about the EKG analyizer not being "dynamic" enough!). One company claims that their transformer based AC products store energy to respond to dynamics, while in theory this is true, the stored energy in the magnetic field of a transformer is minimal-perhaps a 600 pound monster could store enough energy to not hamper dynamics, but I would not want to be around the EM field produced by something like this. For power amps, I prefer parallel style filters, as they do not mess with current delivery, and some of them even offer a little bit of power factor correction. While it is true that parallel filters do not reduce the noise level as much as series filters (or perhaps some transformer based designs) they do not pose dynamic restrictions. I build my own, but I also like the Audience products for one, and I am never sure exactly what is in Jack Bybees products, but I have heard very good sounding systems using them. Even though the Bybee stuff seems kooky, it is not. I have a great deal of respect for Jack, and use some of his Bybee filters in my system to great effect. It appears that Alan Maher's designs are parallel filter circuits as well, I think they would be worth experimenting with also.

Most important though, is to experiment in your system, and only take my advice as a starting point.



SO/ROON/HQPe: DSD 256-Sonore opticalModuleDeluxe-Signature Rendu optical--Bricasti M3 DAC--DIY Purifi Amplifier-Focus Audio FS888 speakers-JL E 112 sub-Nordost Tyr USB, DIY EventHorizon AC cables, Iconoclast XLR & speaker cables, Synergistic Purple Fuses, Spacetime system clarifiers.  ISOAcoustics Oreas footers.                                                       

                                                                                           SONORE computer audio

Link to comment

Pretty sweeping statement. Care to qualify it?


Yes you are right, better provide information here.

I have already described the whole system, while testing the weiss int202 unit. You can find this here :



I'd like to hear some comment about high bitrate playback on low powered cpu systems vs. quad-cores


I have tried to lower the cpu system within this test, but I haven't compared to light speed configurations.


What I haven't tried, utimately, is to replace the psu, with one of the silent 80+ kind. I could not get it on time.





Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Create New...