Taiko Audio Extreme DC Power Distributor

[Christiaan Punter]

Introduction

With the Extreme DC Power Distributor, we designed a simple yet hugely effective power splitter with additional filtering that allows multiple devices to be powered from a single power supply while minimizing cross-contamination.

The 3 different filter outputs allow the user to tune the sound towards getting a neutral balance even with varying power supplies and mains conditions or throughout component changes.

Designed to pair perfectly with the Extreme Switch and Extreme Router, the DC Power Distributor is 100% compatible with any component from any manufacturer that requires up to 25 Volts and up to 10 Amps.

Background

The Extreme DC Power Distributor is a recent development resulting from discoveries made while addressing noise issues associated with Wi-Fi access points. For example, the discovery that the degradation caused by Wi-Fi noise is not dominated by the airborne / radio function but by its interaction with the supply that powers it, and how that, in turn, affects the mains supply that powers our entire systems, gave us a different perspective to investigate.

While testing a variety of power supply designs and various filters, the Extreme Switch and Router turned out to perform better by being powered by a single power supply instead of dual power supplies. This is because a 2nd power supply, even a very low noise linear design, produces more (interacted) noise than the Switch and Router combined. This is where the DC Power Distributor comes in.

Each of the DCD’s filters provides over 80dB of additional noise filtering. The filter’s 64 parts have been selected by ear; some are exotic and costly, like Duelund capacitors, and each has a clearly audible effect on the Sound Quality.

Power

You can use the Wall-Wart Power Supply that is included with the Extreme Switch or Extreme Router to obtain great results. However, higher-end power supplies will yield further increases in Sound Quality.

Features

• Chassis machined from a solid block of copper
• Voltage In = Voltage Out
• Connect your power supply of choice. The maximum input voltage is 25V

Connections

• Unfiltered: Direct-coupled to the input
• Unfiltered: Direct-coupled to the input
• Default: Neutral Filter Output, our recommended starting point
• Default: Neutral Filter Output, our recommended starting point
• Alt1: Relatively more “analytical”
• Alt2: Relatively more “natural”

Specifications

• Maximum input voltage: 25V (voltage in = voltage out)
• Maximum current capacity: 10A
• Earth/Grounding Screw: M6 (use length as needed)
• Power Connector Dimensions: 2.5mm inner / 5.5mm outer / 11.3mm length (14mm recommended)
• Dimensions with feet: 16 x 16 x 5 cm / 6,30 x 6,30 x 1,97 in
• Dimensions without feet: 16 x 16 x 4 cm / 6,30 x 6,30 x 1,57 in
• Weight: 5.2 kg (11.5 lbs)

Connection Schematic


DC Power Distributor PCB




More info can be found on the Taiko Audio Website

Taiko Audio Extreme Extreme DC Power Distributor Product Page (Bundle Page, separate DCD Page to follow)
Taiko Audio Downloads Section

 

[MarkusBarkus]

This afternoon, my DCD arrived from Taiko HQ. It took about 15 minutes to unpack it, get it into my system and pumping out musical goodness. The DCD is said to improve with time. Perhaps it will.

But it is really sweet right off the truck. It sounds to me like a little more holographic air (I have no idea what that means to others) has been added in the presentation. Sweet. A bit more bass presence too, I would add. Very satisfying. Dynamic. Open, flowing sound.

I suppose most folks think, and I'll run with those dogs, their system sounds great. My system sounds great. My power is good. My network is good. So, I'm not expecting my head to explode when I add a power filter...but this is very nice indeed. And that's on the default output. Out of the box.

First up, The Velvet Fog. I have had his 1959 "Mel Torme Swings Shubert Alley" on rotation lately, so it's on my mind. Wow. Stellar. "A Sleepin' Bee." Must be the definitive version, among a host of great renditions. And the music just keeps going through the queue.

A while back, I had ping-ed Emile to see if he would be willing to sell the DCD *without* the router, and he agreed. I may add the router at some point, but (sorry to repeat here) I have a fiber, audio-only ISP for the room, so that sounds pretty good to me. No other traffic. And dedicated power line just for network devices. And I have the Taiko Switch.

I know tech honchos have described other good things the Taiko Router is doing to improve SQ even when playing local, but I wanted to try adding the DCD to my Taiko Switch first.

I'm interested in the battery solution, but I would surely add the DCD to my smaller, second system if it becomes redundant. Or use it in some way in the main system. Let's see how it unfolds.

So, a big Markus Barkus Okie-Dokie for the DCD in an alternative use case model. Looking forward to trying the other filters when I get sorted out and settled in as it is. Happy Listening.

 

[PhP]

This afternoon, my DCD arrived from Taiko HQ. It took about 15 minutes to unpack it, get it into my system and pumping out musical goodness. The DCD is said to improve with time. Perhaps it will.

But it is really sweet right off the truck. It sounds to me like a little more holographic air (I have no idea what that means to others) has been added in the presentation. Sweet. A bit more bass presence too, I would add. Very satisfying. Dynamic. Open, flowing sound.

I suppose most folks think, and I'll run with those dogs, their system sounds great. My system sounds great. My power is good. My network is good. So, I'm not expecting my head to explode when I add a power filter...but this is very nice indeed. And that's on the default output. Out of the box.

First up, The Velvet Fog. I have had his 1959 "Mel Torme Swings Shubert Alley" on rotation lately, so it's on my mind. Wow. Stellar. "A Sleepin' Bee." Must be the definitive version, among a host of great renditions. And the music just keeps going through the queue.

A while back, I had ping-ed Emile to see if he would be willing to sell the DCD *without* the router, and he agreed. I may add the router at some point, but (sorry to repeat here) I have a fiber, audio-only ISP for the room, so that sounds pretty good to me. No other traffic. And dedicated power line just for network devices. And I have the Taiko Switch.

I know tech honchos have described other good things the Taiko Router is doing to improve SQ even when playing local, but I wanted to try adding the DCD to my Taiko Switch first.

I'm interested in the battery solution, but I would surely add the DCD to my smaller, second system if it becomes redundant. Or use it in some way in the main system. Let's see how it unfolds.

So, a big Markus Barkus Okie-Dokie for the DCD in an alternative use case model. Looking forward to trying the other filters when I get sorted out and settled in as it is. Happy Listening.

Hello

I use TAIKO SWITCH for some months and now TAIKO DC for 2 weeks. Addition of this module is as good as said MarkusBarkus. That said on demanding tracks especially on classical music or experimental , I was not completely satisfied by my streaming (I have not yet SGM Extreme but I will consider it next spring as I'm a TAIKO dealer).
I bought a WEISS Engineering PSU linear 230/12V power supply to take place of the small transfo delivered with the Switch and ... music improved so much that I was not prepared to ear much more dynamics, more refined tones, and a capacity to resolve details on metallic instruments (experimental music) that is absolutely stunning. This is a very good solution that will probably be bettered by a TAIKO power supply with batteries, in the meanwhile it's a fine way to have an all streaming system that could reproduce music with a sense of realism I got when I listened to Jordi Savall in Biel 2 weeks ago.

 

[RyanTX]

Has anyone changed their stock DC cables yet that came with the DCD? Just curious if people found any positive or negative differences when doing so?

 

[PhP]

Has anyone changed their stock DC cables yet that came with the DCD? Just curious if people found any positive or negative differences when doing so?

Not yet. I ordered a new cable at WEISS and I will report. Anyhow, using WEISS PSU 102 230/12v (or 110/12v) stabilized power supply made a huge difference.

 

[RyanTX]

Not yet. I ordered a new cable at WEISS and I will report. Anyhow, using WEISS PSU 102 230/12v (or 110/12v) stabilized power supply made a huge difference.

Yeah, from the wall I’m using the Nordost QSource and their premium DC cable going into the DCD. Happy with that combo so far. Just haven’t played with the next part of it yet—from the DCD to router and switch.

 

[PhP]

For sure TAIKO new Olympus and I/O server streamer an battery power supply will change the game. Meanwhile the WEISS unit is crazy good for that, and I received an orchestra director last week, he was amazed how good the demo system sound.

 

[marty]

i just received my DCD (and router) 2 days ago. There has been much discussion about filter settings (for Roon or XDMS) but I haven't seen any discussion of using the "unfiltered" outlets. My question is whether, if this option is chosen, it is basically the equivalent of using your LPS only? In other words, does the DCD just act like a pass through if you use the unfiltered output, or does it do something useful to the supposedly regulated output of your (in my case, an Uptone) LPS?

 

[Taiko Audio]

i just received my DCD (and router) 2 days ago. There has been much discussion about filter settings (for Roon or XDMS) but i haven't seen any discussion of using the "unfiltered" outlets. My question is whether, if this option is chosen, it is basically the equivalent of using your LPS only? In other words, does the DCD just act like a pass through if you use the unfiltered output, or does it do something useful to the supposedly regulated output of your (in my case, an Uptone) LPS?

Yes those are pass through.

 

[marty]

Yes those are pass through.

Got it. The question then is, why are they there? Why would anyone use them?

 

[Taiko Audio]

Got it. The question then is, why are they there? Why would anyone use them?

Maybe you like 1 pass through and 1 filter.

 

[marty]

I am fascinated by the option to "tune" one's sound depending on the filter chosen. This is actually a huge change from the typical mindset that many of us have had in the past where we assume that what we want from any piece of gear is something more akin to "a straight wire with gain" approach. Never mind that virtually every piece of media we play, whether it is LP or digital files, has had its sonic signature optimized by the folks who engineer and produce these products according to their taste. And surely nobody would disagree that we can certainly use equalizers (analog or digital) to optimize our sound. . But tuning the sound one hears by selecting a filter for a DC voltage output is fairly unique and makes me want to understand the mechanism a bit more if possible.

As I understand it, since the DC cables do not pass information in the frequency spectrum typically associated within the audible bandwidth, I presume their effect is mediated by altering the spectrum that is important to conducting noise in the cable. May I ask:

1) Is this noise spectrum in the KHz range, MHz range, or other?
2) Can you share any details (without revealing proprietary information) about how one modifies the noise spectrum to yield a sound that is "Relatively more “analytical” (Alt1) or "Relatively more “natural” (Alt 2). That's what I'm trying to wrap my head around. For example, are different bands of the spectrum being altered differentially with each filter or is it more simply, just a different bandwidth cut off frequency and/or slope that determines the final sound for each filter?.
3) In any event, an 80dB reduction in noise seems enormous. If we assume the S/N of many cables is around 100dB, are you saying that you can somehow lower that noise floor to -180dB? That's just mind blowing to me. Or do I not understand the claim correctly ?

Since there are 2 cables (Router and Switch) each with 4 possible DCD outputs (bypass, default, Alt 1, Alt 2) there are 2^4 (16) possible combinations to sort through, so this will keep me busy for a while! (I haven't had this much fun since since I changed virtually every plug and outlet metal in the system a while back. That took months! Fortunately, this should be a far easier and quicker exercise!) Fun times!
.

 

[PhP]

I am fascinated by the option to "tune" one's sound depending on the filter chosen. This is actually a huge change from the typical mindset that many of us have had in the past where we assume that what we want from any piece of gear is something more akin to "a straight wire with gain" approach. Never mind that virtually every piece of media we play, whether it is LP or digital files, has had its sonic signature optimized by the folks who engineer and produce these products according to their taste. And surely nobody would disagree that we can certainly use equalizers (analog or digital) to optimize our sound. . But tuning the sound one hears by selecting a filter for a DC voltage output is fairly unique and makes me want to understand the mechanism a bit more if possible.

As I understand it, since the DC cables do not pass information in the frequency spectrum typically associated within the audible bandwidth, I presume their effect is mediated by altering the spectrum that is important to conducting noise in the cable. May I ask:

1) Is this noise spectrum in the KHz range, MHz range, or other?
2) Can you share any details (without revealing proprietary information) about how one modifies the noise spectrum to yield a sound that is "Relatively more “analytical” (Alt1) or "Relatively more “natural” (Alt 2). That's what I'm trying to wrap my head around. For example, are different bands of the spectrum being altered differentially with each filter or is it more simply, just a different bandwidth cut off frequency and/or slope that determines the final sound for each filter?.
3) In any event, an 80dB reduction in noise seems enormous. If we assume the S/N of many cables is around 100dB, are you saying that you can somehow lower that noise floor to -180dB? That's just mind blowing to me. Or do I not understand the claim correctly ?

Since there are 2 cables (Router and Switch) each with 4 possible DCD outputs (bypass, default, Alt 1, Alt 2) there are 2^4 (16) possible combinations to sort through, so this will keep me busy for a while! (I haven't had this much fun since since I changed virtually every plug and outlet metal in the system a while back. That took months! Fortunately, this should be a far easier and quicker exercise!) Fun times!
.

Hi
I will let TAIKO answer about their technology. I tried unfiltered, default filter, ALT1 and ALT2. in my system, unfiltered was not interesting. ALT1 gives a presentation that I found a bit bright and forward. ALT2 not my taste, too "round". Defaut filter gives me all I expected, no compromise on details, transients, just a kind of purity and more air, better soundstage, even better density of tones. The sound is now absolutely equal to best CD or SACD red on my PLAYBACK DESIGNS dream. DC is a very interesting unit, if completed with a serious power transformer like WEISS PSU 102, and all of the "chain" - PSU - DC - SWITCH - is powered by a PRANAWIRE MahaSamadhi lightspeed Bocchino power cable and all TAIKO units are on HARMONIX bases.
KR

 

[nonesup]

Yes, I also prefer default on Switch and Router. To power my ISP modem router, I had to choose between Alt_1 and Alt_2 and I preferred Alt_1. Like you, I find Alt_2 “too round”

 

[PhP]

Between Default and Alt 1, the choice depends mainly on all the system. My router is not (yet) from TAIKO.

 

[Taiko Audio]

I am fascinated by the option to "tune" one's sound depending on the filter chosen. This is actually a huge change from the typical mindset that many of us have had in the past where we assume that what we want from any piece of gear is something more akin to "a straight wire with gain" approach. Never mind that virtually every piece of media we play, whether it is LP or digital files, has had its sonic signature optimized by the folks who engineer and produce these products according to their taste. And surely nobody would disagree that we can certainly use equalizers (analog or digital) to optimize our sound. . But tuning the sound one hears by selecting a filter for a DC voltage output is fairly unique and makes me want to understand the mechanism a bit more if possible.

As I understand it, since the DC cables do not pass information in the frequency spectrum typically associated within the audible bandwidth, I presume their effect is mediated by altering the spectrum that is important to conducting noise in the cable. May I ask:

1) Is this noise spectrum in the KHz range, MHz range, or other?
2) Can you share any details (without revealing proprietary information) about how one modifies the noise spectrum to yield a sound that is "Relatively more “analytical” (Alt1) or "Relatively more “natural” (Alt 2). That's what I'm trying to wrap my head around. For example, are different bands of the spectrum being altered differentially with each filter or is it more simply, just a different bandwidth cut off frequency and/or slope that determines the final sound for each filter?.
3) In any event, an 80dB reduction in noise seems enormous. If we assume the S/N of many cables is around 100dB, are you saying that you can somehow lower that noise floor to -180dB? That's just mind blowing to me. Or do I not understand the claim correctly ?

Since there are 2 cables (Router and Switch) each with 4 possible DCD outputs (bypass, default, Alt 1, Alt 2) there are 2^4 (16) possible combinations to sort through, so this will keep me busy for a while! (I haven't had this much fun since since I changed virtually every plug and outlet metal in the system a while back. That took months! Fortunately, this should be a far easier and quicker exercise!) Fun times!
.

Hi @marty ,

It’s indeed very interesting how these filters, utilised in this way, can exert such a significant on your resulting sound perception.

The filters start in the KHz range but work up to MHz ranges.

All filters are designed to roughly address the same frequency ranges yet the filter response differs caused by what I’ll try to explain next.

Filters are all the same in their basics and use 3 components:

1) resistors which have a fixed resistance
2) inductors which have a resistance (impedance) which rises with increasing frequency (doubles each octave)
3) capacitors which have a resistance (impedance) which falls with increasing frequency (halves each octave)

How you use these for DC filtering is largely the same as how you’d use them in loudspeaker crossovers.

You use an inductor in series with a woofer to decrease high frequency energy from passing through the woofer.

You use a capacitor in series with a tweeter to decrease low frequency energy which passes on to the tweeter.

This is then a 6dB / octave filter, if it’s impedance at 1 KHz equals the resistance of the voice coil of the loudspeaker driver you reduce power running through it by 50% at that frequency.

You can then create more effective filters with stronger attenuation by combining inductors and capacitors to create 12dB attenuation filters, or 18dB, or 24dB etc etc.

This works roughly the same for DC filtering however at higher frequencies we start running into something called “parasitics”. A capacitor also has a small amount of inductance, an inductor has a small amount of capacitance. This is something you can ignore with loudspeakers with a limited (low) frequency range to address, but not with high frequency DC filtering.

The inductor you use will have an increasing resistance up to a limit where it’s parasitic capacitance starts working against this and the impedance starts dropping again. So it’s actually more like a bandpass device with an impedance between 2 values, low and high frequency.

The same goes for a capacitor, where parasitic inductance will decrease it’s efficiency at conducting high frequencies.

PCB circuit traces start playing a significant role at some point aswell.

So with the DC distributor you have a variation in a combination of better or worse filter efficiencies with variations in frequency responses due to parasitics of the components and PCB layout.

ALT2 starts at a lower frequency but is less efficient at higher frequencies. Default filters similarly in KHz ranges, but is better at MHz ranges, extending further up with stronger attenuation at medium-high frequencies. ALT1 is the cleanest/strongest filter extending even into GHz ranges.

 

[MarkusBarkus]

@Taiko Audio when you recommend the default filter (I think it was) for the switch as last in the chain before the server, would you say that is because it is what the Extreme likes best electrically, or rather what you like best sonically?

Recognizing you urged folks to try for ourselves...and that you have a Borg-like symbiosis with the server, and what the Extreme likes and what you like are the same. Cheers, Emile.

 

[marty]

Hi @marty ,

It’s indeed very interesting how these filters, utilised in this way, can exert such a significant on your resulting sound perception.

The filters start in the KHz range but work up to MHz ranges.

All filters are designed to roughly address the same frequency ranges yet the filter response differs caused by what I’ll try to explain next.

Filters are all the same in their basics and use 3 components:

1) resistors which have a fixed resistance
2) inductors which have a resistance (impedance) which rises with increasing frequency (doubles each octave)
3) capacitors which have a resistance (impedance) which falls with increasing frequency (halves each octave)

How you use these for DC filtering is largely the same as how you’d use them in loudspeaker crossovers.

You use an inductor in series with a woofer to decrease high frequency energy from passing through the woofer.

You use a capacitor in series with a tweeter to decrease low frequency energy which passes on to the tweeter.

This is then a 6dB / octave filter, if it’s impedance at 1 KHz equals the resistance of the voice coil of the loudspeaker driver you reduce power running through it by 50% at that frequency.

You can then create more effective filters with stronger attenuation by combining inductors and capacitors to create 12dB attenuation filters, or 18dB, or 24dB etc etc.

This works roughly the same for DC filtering however at higher frequencies we start running into something called “parasitics”. A capacitor also has a small amount of inductance, an inductor has a small amount of capacitance. This is something you can ignore with loudspeakers with a limited (low) frequency range to address, but not with high frequency DC filtering.

The inductor you use will have an increasing resistance up to a limit where it’s parasitic capacitance starts working against this and the impedance starts dropping again. So it’s actually more like a bandpass device with an impedance between 2 values, low and high frequency.

The same goes for a capacitor, where parasitic inductance will decrease it’s efficiency at conducting high frequencies.

PCB circuit traces start playing a significant role at some point aswell.

So with the DC distributor you have a variation in a combination of better or worse filter efficiencies with variations in frequency responses due to parasitics of the components and PCB layout.

ALT2 starts at a lower frequency but is less efficient at higher frequencies. Default filters similarly in KHz ranges, but is better at MHz ranges, extending further up with stronger attenuation at medium-high frequencies. ALT1 is the cleanest/strongest filter extending even into GHz ranges.

Thanks Emile for your excellent reply. The sonic differences among the filters are rather obvious. What I'm trying to figure out is which one (and which combination) is sonically correct, or to be more appropriate, sonically preferred. The first thing that comes to mind, is that the filters are a good solution to a problem that may not be relevant if a battery was utilized in lieu of an LPS. But that's a story for another day. Meanwhile, it seems they can be a excellent adjunct for system "tuning" as we often like to do, which is of course, the reason you provided such good flexibility among the filter options.

If I remember correctly from my crossover building days, the parameters of impedance, capacitance and inductance are obviously important in altering the bandwidth and slope of the filter function, but inherent in that is also the phase response that is dependent on the filter design (i.e. Butterworth, Bessel, Linkwitz-Riley). I'm currently trying to assess the various filters with particular regard to phase error and their sonic consequences since it is in this area that I think the filter differences impact the sonics in a way that seem to matter most as to what the recording microphones reveal (perhaps most easily assessed for orchestral recordings). Phase errors are, for me, often the key as to the much cited ":you are there" or "they are here" effect in which the time domain changes are perhaps more relevant than frequency domain differences when listening to music reproduction with the various filters in the DCD. I'm still sorting out my preferences but in the interim I'm grateful for the filter options provided while trying to learn their respective sonic signatures.

 

[Taiko Audio]

@Taiko Audio when you recommend the default filter (I think it was) for the switch as last in the chain before the server, would you say that is because it is what the Extreme likes best electrically, or rather what you like best sonically?

Recognizing you urged folks to try for ourselves...and that you have a Borg-like symbiosis with the server, and what the Extreme likes and what you like are the same. Cheers, Emile.

Hi @MarkusBarkus ,

I don't recommend anything, it's just my personal preference

 

[oldmustang]

Thanks Emile for your excellent reply. The sonic differences among the filters are rather obvious. What I'm trying to figure out is which one (and which combination) is sonically correct, or to be more appropriate, sonically preferred. The first thing that comes to mind, is that the filters are a good solution to a problem that may not be relevant if a battery was utilized in lieu of an LPS. But that's a story for another day. Meanwhile, it seems they can be a excellent adjunct for system "tuning" as we often like to do, which is of course, the reason you provided such good flexibility among the filter options.

If I remember correctly from my crossover building days, the parameters of impedance, capacitance and inductance are obviously important in altering the bandwidth and slope of the filter function, but inherent in that is also the phase response that is dependent on the filter design (i.e. Butterworth, Bessel, Linkwitz-Riley). I'm currently trying to assess the various filters with particular regard to phase error and their sonic consequences since it is in this area that I think the filter differences impact the sonics in a way that seem to matter most as to what the recording microphones reveal (perhaps most easily assessed for orchestral recordings). Phase errors are, for me, often the key as to the much cited ":you are there" or "they are here" effect in which the time domain changes are perhaps more relevant than frequency domain differences when listening to music reproduction with the various filters in the DCD. I'm still sorting out my preferences but in the interim I'm grateful for the filter options provided while trying to learn their respective sonic signatures.

I'm not sure that you have to concern yourself too much about phase error when you are talking about DC. Or are you concerned about increasing the phase error of whatever noise makes it through the filter?

Steve Z