CHORD ELECTRONICS DAVE
Dec 10, 2021 at 7:38 AM Post #18,436 of 25,909
Farad super3 setup for the Chord Dave



After reading a lot on the forums about upgrading the LPS (linear power supply) of the Chord Dave and the results people are getting by adding a Paul Hynes or Sean Jacobs DC4 power supply, I was eager to get one on my own set. I already had a Farad Super3 on the Mscaler and because farad is also Dutch I contacted Mattijs the owner of Farad if he was interested in doing a project of making an affordable lps for the Dave. In the past he had this question often but I explained my own background as electronics engineer and willing to use my new Dave as test object and make some parts for the project. About around the same time he was asked to do a kind of similar lps for another DAC brand with a positive and negative power supply. So now two parties were asking him and having some ideas already in his head he agreed, notice!!! this was only approx 2 month ago.

After a brainstorming session he told me passionately about his ideas and what he thought should be on such a system for a high end DAC. Mattijs has some experience on audio design as being a founder of Pink Faun. The first key feature that was an absolute must was designing a true negative powersupply. So not pairing two positive supplies where the positive of one is attached to the negative of the other and so creating a minus when using a floating ground or rail as it is called. A true negative power supply is completely opposite build so eg. npn transistors, capacitors are the other way around - same with the diode bridge and the negative based regulators etc.. will not go into the details but the end result is a lps were all the common grounds can be attached to each other one does not have to be floating. So all rf noise can be killed at the source and not at the end of the chain. Also floating ground can be submitted to different loads from the upper and lower side voltages and this could cause a shifting or jittering ground level.

Secondly Mattijs and i are convinced that every DC voltage the Dave gets should have a separate casing and transformer so no cross contamination can occur and every lps is filtered on its own internally and externally eg. Powercord, casing, wiring etc.. Using farads stock components would also save a lot of development and testing time. These power supplies have proven themself in and out of the audio world. He designed his true negative supply in no time having the idea for a long time already, he only had to design a new pcb and had this made in his production facility in no time.



The third key feature we figured that it should have was a safety device for enabling the power to the Dave. Each voltage +-15V and +5V, three in total will come to the Dave in different timing. A -15 will start up in approx 7sec and the 5V in 1 sec. We don't know if the Dave can handle this and what happens if you are on holiday and one power supply fails and it will only get 2 out of three voltages.



The fourth but not the most un-important one, it should be affordable. I cannot give prices because they are not mine to give but you can do a little bit of the math yourself as we are using three stock farad power supplies if you add the one for the Mscaler 4 in total so thats approx 2400,- in VAT on 4 stock lps’s add the wiring and monitoring print and the aluminum inlet with plugs and maybe powercords you can guess were the basic version will start. On a later stage I will test upgrades such as fuses, rhodium inlets, but only after setting a baseline with the stock components. (While i am typing this i am listening to my ultimate playlist and are flabbergasted every few moments)
Using 4 Farad Super3 power supplies you will not have a big footprint, the supplies can be stacked or laid out flat next to each other. A lot of configurations are possible and as I have learned in this audio world everyone has a different approach on how they setup their gear and this way everybody can be happy. I have seen people building racks for just there farads.

So mattijs designed a monitoring print for inside of the Dave. The space was used were the original power supply was. This print will only pass through the three DC voltages simultaneously if all are present, a so called AND function in digital terms and a fail save if people would switch the lines. This way the dave is always safe, not a bad thing for a DAC that costs 11.400,- euro in the EU. He uses a microprocessor on this board that later on can be tweaked or reprogrammed if necessary. While mattijs was designing the lps i was doing some cnc work for the power inlet of the dave.

It still amazes me from two viewpoints how the stock $32 costing medical switched power supply of the dave still sounds so good and also that chord does not upgrade this by making a choral stand with an external power supply, how hard can it be? And people, the difference is not subtle it's huge - how much better the Dave gets by doing this i will almost call it Dave 2.

The advantage of above system is that customers pretty much can install this system themselves with a degree of safety that nothing can be blown up by doing something wrong or if/when components fail in the future. Everything can fail as we all know, the Chord as can be read on forums are also failing now and then and repaired by Chord, that's life when using electronics, crap happens. But this way it feels much better and to be honest the start up sequence is also pretty cool to watch. All separate lps’s LED start to blink red and when ready go over into a nice blue and after a fraction of a sec the dave comes alive (sorry thats my ocd kicking in).

Last week Mattijs called me that the monitor print was ready and tested. So he gave me a DIY kit of all separate components for me to install into the dave and test if it works. The first trail I do will be with stock farad super3 power supplies the only upgrade that is made are the dc lines - these are silver. The past has proven that these cables work the best between the super3 and Dave. For the power cords I had to go on Mattijs' experience and the feedback from his customers so i used TCS021 furutech cable this is slightly thinner version than the advised TCS031 version. I wanted to use one AC power connecter to 3 AC euro inlets this saves alot of space in the power block. The connectors used are also furutech and the gold versions around $30 a piece.

I soldered currently Daves own internally dc cord of its own power supply as the gold plated pins of the new connector are in back-order. I think in the future for my own setup will solder everything and will not use any connector but that's custom and not for now.

After testing everything without connected the dc to the Dave, it all worked in one go. Next I made everything fit and flipped the switch and after the startup sequence the Dave came nicely alive. I installed everything in my setup and the magic started.

I will repeat my first impressions I applied to Mattijs at 2 in the middle of the night “every instrument and voice has got its own space now, bass has much more texture and attack and depth and with vocals it is as if they are separate entities in front of you…” and than some words i will not repeat online, next notes I sent were “the fatigue that was always a bit there are just GONE the snares and trumpets tjeessss mattijs i am back to listening again bye”

This morning after listing for a few hours and thinking what to say about the change it makes out of the box and how to describe it best for now i think it's as if the sound and coloration of my gear/setup is vanished and just the music is left. Btw excuse me for my English i am not native English but Dutch and just typing as i am thinking. Our opinions above are also just mine and in my own setup. I owned almost all modern Chord dacs including all lps tweaks. In my setup i use an Aurender N10 as streamer a new Chord Dave and Mscaler connected with the high fidelity storm bnc cables. Amplifier is a Nelson Pass class A connected to the new Usher TD-10 speakers who are internally upgraded. I also have dedicated power on one phase from the powerbox just for my audio gear. As headset i use the Focal Utopia. In the upcoming week i will do some serious listening and will come back on my finding on this magical upgrade.
Can anyone explain how powering an M Scaler with an LPS makes it sound better?
I am not questioning anyone’s subjective experiences, just trying to understand what problem is being solved by it.
M Scaler is a massive RF generator and will generate noise even on a battery power.
Does LPS protect the mains from the noisy M Scaler?
What am I missing?
 
Dec 10, 2021 at 2:33 PM Post #18,437 of 25,909
I’ve got about 1 week of break in on the latest Sean Jacobs power supply. A significant upgrade. I remember adding a Blu2 to my system years ago and although there was an improvement it was minor. The power supply upgrade is one of the larger upgrades I’ve made to my system.
 

Attachments

  • 991EA1B6-4C25-4AD6-9CAB-A59765C76A21.jpeg
    991EA1B6-4C25-4AD6-9CAB-A59765C76A21.jpeg
    2.9 MB · Views: 0
  • A4B58E48-5DFE-45F2-AA62-C54A63DC9F6A.jpeg
    A4B58E48-5DFE-45F2-AA62-C54A63DC9F6A.jpeg
    4.9 MB · Views: 0
Dec 10, 2021 at 2:52 PM Post #18,438 of 25,909
Great, but this was not my question.
Yours is one of many testimonies I read.
I am just trying to understand what is the problem definition and how LPS addresses it.
Admittedly, I had been a victim of my own confirmation bias in past, so I am trying to lay some foundation prior to pursuing an upgrade.
Also, understanding a problem and a solution helps prioritising upgrades.
Is BNC cable from M Scaler to DAVE the largest vulnerability?
Is the USB input on DAVE the one and should be avoided at all cost?
Is it the grounding scheme a culprit and needs to be addressed by going optical?
If the answer “all of the above”, I am going to end up with holy mess on my hands and many thousand dollars poorer.

I do not need 100% of performance, I am pretty happy with 90% that I have now, but willing to entertain another 5% at a reasonable cost.

But reading some of the comments, I am getting impression that the stock power supplies and cables only achieve 50% of what is available, which I refuse to accept as a fact.
 
Dec 10, 2021 at 2:59 PM Post #18,439 of 25,909
Can anyone explain how powering an M Scaler with an LPS makes it sound better?
I am not questioning anyone’s subjective experiences, just trying to understand what problem is being solved by it.
M Scaler is a massive RF generator and will generate noise even on a battery power.
Does LPS protect the mains from the noisy M Scaler?
What am I missing?
I am sure others will put it more knowledgeably, but HMS generates RF both into the mains and through the air. Using a battery addresses the first route for RF to travel but not the second. In other words, taking mains RF solution away is what a battery achieves (and a LPS will not totally achieve). However, batteries bring their own issues. As I understand it a good LPS provides HMS with a stable voltage that a battery cannot match. To my ears at least, a good LPS is noticeably better than battery.
 
Dec 10, 2021 at 3:01 PM Post #18,440 of 25,909
I'm also a massive skeptic but let me address your questions with what I've gathered so far :
Is the USB input on DAVE the one and should be avoided at all cost?
From my personal perspective yes. No matter what macbook (unplugged) i used it also sounded worse than Optical out of a "noisy gaming pc ". This was true until i got my streamer where coax just sounds better than even optical to me and noise is not that much of an issue even over USB.
Is it the grounding scheme a culprit and needs to be addressed by going optical?
Is BNC cable from M Scaler to DAVE the largest vulnerability?
Yes but it's the BNC cables that screw you, and yes, the M-scaler is a massive RF generator. Some people use ferrites, i use a battery for example. I've tried a Farad with a non-ferrited M-scaler and it was significantly worse, and clearly worse than the stock SMPS. All the hallmarks of RF noise were there.
I do not need 100% of performance, I am pretty happy with 90% that I have now, but willing to entertain another 5% at a reasonable cost.

But reading some of the comments, I am getting impression that the stock power supplies and cables only achieve 50% of what is available, which I refuse to accept as a fact.
People are blowing things out of proportions. Even with all optimizations , I think Dave itself is 80% of the SQ. the M-scaler takes away more than it brings without a battery from my perspective. Be it directly or from an Opto-DX which also needs a battery for complete isolation.

The best upgrade (without any downsides) I've found to Dave is the mains filter
 
Dec 10, 2021 at 3:07 PM Post #18,441 of 25,909
Great, but this was not my question.
Yours is one of many testimonies I read.
I am just trying to understand what is the problem definition and how LPS addresses it.
Admittedly, I had been a victim of my own confirmation bias in past, so I am trying to lay some foundation prior to pursuing an upgrade.
Also, understanding a problem and a solution helps prioritising upgrades.
Is BNC cable from M Scaler to DAVE the largest vulnerability?
Is the USB input on DAVE the one and should be avoided at all cost?
Is it the grounding scheme a culprit and needs to be addressed by going optical?
If the answer “all of the above”, I am going to end up with holy mess on my hands and many thousand dollars poorer.

I do not need 100% of performance, I am pretty happy with 90% that I have now, but willing to entertain another 5% at a reasonable cost.

But reading some of the comments, I am getting impression that the stock power supplies and cables only achieve 50% of what is available, which I refuse to accept as a fact.
Everything matters ... and the differences are, surprisingly, large (percentages are tricky and potentially misleading, but I would put the potential upside at massively over 5%). You are wise to be cautious, though, about falling into a rabbit hole! The cheapest upgrade, which is very noticeable once RFI is addressed (but might not be as noticeable otherwise?) is to pop a DAVE on rollerballs. Also, if you have a HMS and a DAVE you should certainly use BNC as otherwise the HMS is not giving you anything like it should because you are not using anything like its full upscaling potential.
 
Dec 10, 2021 at 3:14 PM Post #18,442 of 25,909
I'm also a massive skeptic but let me address your questions with what I've gathered so far :

From my personal perspective yes. No matter what macbook (unplugged) i used it also sounded worse than Optical out of a "noisy gaming pc ". This was true until i got my streamer where coax just sounds better than even optical to me and noise is not that much of an issue even over USB.


Yes but it's the BNC cables that screw you, and yes, the M-scaler is a massive RF generator. Some people use ferrites, i use a battery for example. I've tried a Farad with a non-ferrited M-scaler and it was significantly worse, and clearly worse than the stock SMPS. All the hallmarks of RF noise were there.

People are blowing things out of proportions. Even with all optimizations , I think Dave itself is 80% of the SQ. the M-scaler takes away more than it brings without a battery from my perspective. Be it directly or from an Opto-DX which also needs a battery for complete isolation.

The best upgrade (without any downsides) I've found to Dave is the mains filter
Ok, this is closer to where I am with all of this.
My starting assumption is that when you spend this kind of money ($10-15K), you already entered the area of diminishing returns, right out of the box.
If this or that tweak made someone’s jaw drop, would it make that tweak worth, what, 10%?
Combine all the tweaks people on this forum mention and you got my point…
 
Last edited:
Dec 10, 2021 at 3:22 PM Post #18,443 of 25,909
I am sure others will put it more knowledgeably, but HMS generates RF both into the mains and through the air. Using a battery addresses the first route for RF to travel but not the second. In other words, taking mains RF solution away is what a battery achieves (and a LPS will not totally achieve). However, batteries bring their own issues. As I understand it a good LPS provides HMS with a stable voltage that a battery cannot match. To my ears at least, a good LPS is noticeably better than battery.
Ok, if the problem definition is M Scaler‘s contamination of the mains, is the LPS the most effective way to combat it?
Why nobody talking about ferrites, similar to the BNC cables?
Does that problem go away if M Scaler plugged into a power conditioner with its own filters?
Again, I spent money in past on things that I believed should help and “heard” improvements.
Then I moved on, forgot about them and could not tell the difference later on when I chose to revisit the subject.
 
Last edited:
Dec 10, 2021 at 3:37 PM Post #18,444 of 25,909
Great, but this was not my question.
Yours is one of many testimonies I read.
I am just trying to understand what is the problem definition and how LPS addresses it.
Admittedly, I had been a victim of my own confirmation bias in past, so I am trying to lay some foundation prior to pursuing an upgrade.
Also, understanding a problem and a solution helps prioritising upgrades.
Is BNC cable from M Scaler to DAVE the largest vulnerability?
Is the USB input on DAVE the one and should be avoided at all cost?
Is it the grounding scheme a culprit and needs to be addressed by going optical?
If the answer “all of the above”, I am going to end up with holy mess on my hands and many thousand dollars poorer.

I do not need 100% of performance, I am pretty happy with 90% that I have now, but willing to entertain another 5% at a reasonable cost.

But reading some of the comments, I am getting impression that the stock power supplies and cables only achieve 50% of what is available, which I refuse to accept as a fact.
I was not answering your question, if I was I would have proceeded my comments by quoting you 😇
 
Dec 10, 2021 at 3:37 PM Post #18,445 of 25,909
Ok, if the problem definition is M Scaler‘s contamination of the mains, is the LPS the most effective way to combat it?
Why nobody talking about ferrites, similar to the BNC cables?
Does that problem go away if M Scaler plugged into a power conditioner with its own filters?
Again, I spent money in past on things that I believed should help and “heard” improvements.
Then I moved on, forgot about them and could not tell the difference later on when I chose to revisit the subject.
HMS radiates RFI through both each and every cable: BNC and mains. A power conditioner will do nothing to help RFI travelling via BNC. I have not tried ferries, but Nick (Wave) offers them trial (at least in the UK - you could ask him re the US) ... so you could see if you feel they justify the expense? If you are weighing up ferrites against the OptoDX, then which works out cheaper depends how far down the rabbit hole you are prepared to go (although Opto DX and good LPS is likely to be the most expensive route).

By the way, I sympathise and share the scepticism. But, on playing around a little, I was very much persuaded as to the difference - not remotely something that I believed I heard and didn't notice when removed. We all, though, hear differently ....
 
Dec 10, 2021 at 3:43 PM Post #18,446 of 25,909
I have got three kemp filters in my power block, two sns and a qa filter that block the root back and forth over the ac power everybody who sits down with me and i plug and unplug these filters hears it imidiatly in the highs and the ease that comes in the music(people like friends, family etc non audifreaks like us) then the two biggest upgrades in my opinion are the dual bnc from high fidelity storm cables and now all 4 lps.

In percentages if its worth anything its 70% baseline dave, add 5% for the filters 10% for the bnc and 15% on the lps’s but the combi off three enhances everything so 1+1+1=4 hope it makes sense. Reading alot about it and people who compared it to vivaldi etc. I did not but taking this route makes the whole digital stack to aproxx 20k euro and beats alot of these 30-40k stuff. I cannot confirm first hand but asume a bit. But it really sound the bomb now compared to dave stock.

Hope this bit helpes you a bit
 
Dec 10, 2021 at 3:45 PM Post #18,447 of 25,909
Does that problem go away if M Scaler plugged into a power conditioner with its own filters?
As much as i hoped it would...It doesn't. My filter has 6 individually filtered sockets. The effects are just as noticeable.
Ok, if the problem definition is M Scaler‘s contamination of the mains, is the LPS the most effective way to combat it?
It does nothing to combat it without ferrites and sounds significantly worse than the stock brick to my ears.
Ok, if the problem definition is M Scaler‘s contamination of the mains, is the LPS the most effective way to combat it?
Why nobody talking about ferrites, similar to the BNC cables?
Rob explained a few years back that there is a ground loop that you can only break by using optical or a battery. When you do this the BNC cables basically filter the M-scaler noise themselves (to a degree, not sure if entirely ) and then you won't need Ferrites. But decent cables still help with overall SQ.
I think what some people are saying is that if you use ferrites to filter out the noise generated by the M-scaler (and the ground loop via the mains), by using a LPSU you generate less noise into the mains and overall. But then your devices won't benefit from the filtering designed for the smps.
At least those are the only conclusions I've been able to draw from way too much time wasted on forums.
I completely disagree that an LPS improves everything as if it's magic. Heard the Farad make my M-scaler and a friend's Qutest sound worse (more noise). I agree with what Rob said about it being a case of people enjoying distortion, or maybe they're just adding it via an LPSU and removing it via ferrites.

TLDR: IMO a battery is the most cost effective solution and the one that makes the most sense . I have no idea as to why you would need a "high current" battery for the M-scaler. Or what that even means, but it feels like BS to me.
As for Dave, I'm willing to try a triple Farad solution and see for myself before i comment. If it's well designed( for Dave), i imagine it might. I'm also sure people are blowing things out of proportions, as experience has taught me.
 
Dec 10, 2021 at 3:47 PM Post #18,448 of 25,909
I have got three kemp filters in my power block, two sns and a qa filter that block the root back and forth over the ac power everybody who sits down with me and i plug and unplug these filters hears it imidiatly in the highs and the ease that comes in the music(people like friends, family etc non audifreaks like us) then the two biggest upgrades in my opinion are the dual bnc from high fidelity storm cables and now all 4 lps.

In percentages if its worth anything its 70% baseline dave, add 5% for the filters 10% for the bnc and 15% on the lps’s but the combi off three enhances everything so 1+1+1=4 hope it makes sense. Reading alot about it and people who compared it to vivaldi etc. I did not but taking this route makes the whole digital stack to aproxx 20k euro and beats alot of these 30-40k stuff. I cannot confirm first hand but asume a bit. But it really sound the bomb now compared to dave stock.

Hope this bit helpes you a bit
I am curious if you've tried a battery with the M-scaler vs the Farad. Without the Wave cables.
 
Dec 10, 2021 at 3:49 PM Post #18,449 of 25,909
HMS radiates RFI through both each and every cable: BNC and mains. A power conditioner will do nothing to help RFI travelling via BNC. I have not tried ferries, but Nick (Wave) offers them trial (at least in the UK - you could ask him re the US) ... so you could see if you feel they justify the expense? If you are weighing up ferrites against the OptoDX, then which works out cheaper depends how far down the rabbit hole you are prepared to go (although Opto DX and good LPS is likely to be the most expensive route).

By the way, I sympathise and share the scepticism. But, on playing around a little, I was very much persuaded as to the difference - not remotely something that I believed I heard and didn't notice when removed. We all, though, hear differently ....
I can’t reply to HMS but preferred lifatec tos (optical $100) into Dave directly over Blu2. That was with both an Aurender N10 and Project CD box rs both have outstanding spdif outputs. I also found power cables for Dave with 20 or so clamp on ferrites helped ( Pangea ac14 se II $80 + $20) and isolating the mains with a ps audio P10 powerplant helped. The Pangea, ferrites and lifatec are cheap enough it’s worth the experiment.
 
Dec 10, 2021 at 4:06 PM Post #18,450 of 25,909
Has anyone measured the rfi produced by the mscaler? Rob says most is mitigated internally.
 

Users who are viewing this thread

Back
Top