Conversion of a Chord DAVE to DC power using two Uptone Audio JS-2s

[skatbelt]

10 hours ago, asdf1000 said:

Btw: I don't write all this just to argue.

 

I just think people in the past have taken some of what each has said and tried to make a generalisation. 

 

To write it a different way: has Rob Watts measured an LPS-1.2 or JS-2 ? Has John S measured Chord DAVE's power supply? I haven't read anything anywhere that says yes to either question.

 

I'm happy to be proven wrong though. Maybe John S will chime in.

 

 

I agree with you. With the development of DAVE, Rob Watts mission was to design a state of the art DAC within the technological possibilities (at the moment). I therefore see the choice of a SMPS as part of the overall the design decisions. If a LPS would have fitted better within this philosophy and contributed positively to the objectives, then he would undoubtedly have chosen that route. 

 

There are very good SMPS's and very bad SMPS's just as there are good and bad LPS's. The problem in high end audio is that SMPS almost seems to be a curse word because of the image that is created by the large group of very cheap and noisy ones.

 

I also have reservations about the objectivity of the OP's assessment. Unless he directly A/B-ed his modified DAVE with a 'standard DAVE' ceteris paribus I cannot take 'shockingly better' very serious.

 

On the other hand, I have experienced that DAVE is sensitive to the quality of the power provided. For example, a dedicated power line, an isolation transformer with balanced power and a better power cable all brought audible improvements in my system.

[skatbelt]

1 hour ago, DAVE JS2 said:

Thanks for your comments @skatbelt

 

One would hope Rob / Chord took the approach you describe, I'm simply reporting that I heard very meaningful improvements after the switch (as have others).

 

You raise a fair point about objectivity, and of course as I don't have two DAVE's I can't easily do repeated A/Bs. The conversion took maybe 20 minutes in all, so there is that gap too. However, set against those potential issues are the facts that I was using tracks that are VERY well known to me, same volume levels, the conversion was done after about an hour or so of acclimatisation of listening with my friend (so his and my wife's opinions too, not just mine), and I was actually not expecting that much to change (if anything). That's really the reason I posted about it; it was such a dramatic change when there are many reasons to expect it to be subtle, at best. I remain delighted with it.

 

On another topic, I'm really interested in your experiences with your isolation transformer, having had a dedicated mains spur put in over the summer.  I'll try and read up more about it myself, but presumably the reason you do this is to further reduce mains noise? Do you have any before & after measurements? Also a question about the transformer size, am I direct in thinking that do you need to match up the transformer spec with the needs of the electronics 'behind' it, particularly the power amp?

 

Thanks in advance for any information.

 

I won't start a psychological discussion... :)

 

There is a whole thread in this sponsored forum about isolation transformers, infused by - who else? - John Swenson:

 

 

 

When implemented properly it will reduce high frequency common mode noise on the AC line. All details can be found in the thread. It is quite a read so in short you'll need an isolation transformer that has enough headroom to power the needed wattage behind it and one that has extreme low so called interwinding capacitance. You need to power all components behind the transformer from one power strip and make this part of the system a closed group in electrical terms. So, no copper connections with powered components that aren't part of the group to prevent from leakage currents.

 

I have mine wired for balanced power. It is a moderate 1KVA model but this plenty enough to power a dCS Network Bridge, DAVE and a Lyngdorf TDAI-3400. The uptream network components are isolated via an etherREGEN and a separately powered subwoofer is isolated via a Jensen transformer.

 

I perceived implementing this as a full component upgrade with added tranquility, ease and realness and at the same time more details in sound reproduction. The difference between balanced/symmetrical vs. asymmetrical was quite shocking to stay in your terms... :)

 

 

 

[skatbelt]

1 hour ago, asdf1000 said:

 

Let's not forget the time Rob said the Audioquest Jitterbug made a difference to DAVE's USB input !! 😃

 

For any technical disagreements please direct them to Rob, not me !

 

 

 

These are exactly the kind of topics that led me to ban USB a long time ago. In my experience, even USB cables alone make huge differences. This connection type is so brittle. I settled with AES/EBU in on DAVE and have yet to hear a better combo than with the dCS Network Bridge. May be something like a Grimm MU1 will top this but I like to keep things simple and minimalistic and like to keep as much as computer hardware far away from the DAC as possible in noise injecting terms.

 

The discussion about RFI and the apparent need for expensive cables with certain ferrite-configurations has also prevented me from trying out M Scaler technology yet. But may be this is a stupid choice and I don't know what I'm missing...

[skatbelt]

1 hour ago, DAVE JS2 said:

Oh my goodness @ecwl @skatbelt @asdf1000

 

This epic Head-Fi review / article / journey is incredibly comprehensive and touches on pretty much all the topics we have all talked about / around in this thread.

 

Recommended reading, when you've got a fair amount of spare time, and its worth getting through to the end. Also worth reading his linked piece about his DIY 'Ferrite Cable of Shame' as background.

 

This was mid 2019 and Ray progressed to the following. Note that he returned to USB but with the use of a low cost optical isolation system and he eliminated the M Scaler:

 

 

 

 

[skatbelt]

3 hours ago, kennyb123 said:

I have a ton of respect for the folks at Chord.  Rob is a genius when it comes to converting bits into live music.  He really gets sampling theory.  He also understands human hearing and particularly our sensitivity to timing.  His understanding places him amongst a very small group of experts.

 

Having said all that, I’d rather have someone other than Rob design the power supply for his digital products.  Guys like John Swenson, Paul Hynes, Sean Jacobs and Mattijs Vries may be to power supplies what Rob is to D-to-A conversion.  Both my Chord devices (TT2 and HMS) were massively improved with Farad power supplies.  Thus I’m not at all surprised the DAVE gained hugely from an improved power supply.

 

The other essential component when using an HMS is the OPTO-DX.  I can’t speak highly enough about this product.  I know Chord doesn’t think anything like this is needed, but I’ve benefited from tuning them out on this the same way I have tuned them out on power supplies and music servers.  I love Chord to death, but I think the breadth and depth of understanding needed to get everything to the point of being exceptional is a bit too much to expect from one or two guys.  We have to look to multiple experts and try our best to pull it all together in a way that works best for us.  

 

With all due respect, but the TT2 (and HMS) are designed for other market segments and scaled down versions of their big brothers. The choice for cheaper (Chinese) external SMPS power supplies fits this thought. I can even imagine that Chord already assumed that many of their customers would experiment with better power supplies and therefore equipped them with a DC input.

 

I am not saying that DAVE cannot take advantage of a very good external power supply but I suspect that the delta is determined by how well this supply is able to suppress mains noise and instabilities relative to the built-in supply. In this line of thought, this delta can possibly also be achieved by offering cleaner/better power to a 'stock' DAVE.

 

I briefly mentioned the broadly acclaimed Grimm MU1 in a previous post. This unit also has a purposely designed SMPS. The development of the MU1 took over 2 years and every component was careful optimised for SQ. Would they have opted for an SMPS if a LPS had been the better choice? 

[skatbelt]

I have an idea! Why not send the SMPS that you @DAVE JS2 have taken out of DAVE and is now unemployed to @JohnSwenson with the intention to test it on the important properties of a power supply he mentioned above. I think a very large group (DAVE owners but also others) is interested in the outcome and maybe it also says something in general about the state of purposely built SMPSs in high end audio products. I am sure John will take on this challenge!

[skatbelt]

1 hour ago, DAVE JS2 said:

 

Hang on...what about the cruelty of depriving ME of listening to my upgraded DAVE?!

 

It is just one that Chord has bought in (see photo), and I'd imagine @JohnSwenson has more pressing matters to attend to. But I'd be happy to send it if it would be of use to him.

 

Thanks! Hmmm, there goes my assumption of purposely built or at least custom built to spec for Chord. This is an off the shelf supply. For medical, among other applications. Costs about 50 euros a piece here in the Netherlands and a lot less when bought in large batches.

 

Technical specs: https://www.xppower.com/portals/0/pdfs/SF_ECM40-100.pdf