My MScaler has been in stby all day and is cool to the touch, but once turned on it gets surprisingly hot. I personally wouldn’t stack anything on top. I used to have the tt2, but they we’re side by side.
Hugo M Scaler by Chord Electronics - The Official Thread
- Thread starter ChordElectronics
- Start date
My MScaler has been in stby all day and is cool to the touch, but once turned on it gets surprisingly hot. I personally wouldn’t stack anything on top. I used to have the tt2, but they we’re side by side.
My TT2 was on top of MSC ... not any more. They can create heat but I've found side by side they are ok ... and seems to be better sound separated - might be in my head but the pair are now side by side, well about 400mm apart on the same level.
GreenBow
Headphoneus Supremus
- Joined
- Feb 6, 2015
- Posts
- 4,197
- Likes
- 1,603
Hi Rob, Is there a trend to go towards linear power supplies among consumers , a fashion almost with the belief is they are superior, when properly designed smps can be very good or better, and would it be fair to say that linear types can potentially produce an over filtered sound in music leading to the idea that an m scaler can sound less bright or smoother? I’ve used both Types with different bits of kit over the years but the debate is ongoing for using lps with both m scaler and Hugo tt2. Does Dave have a smps on board?
Rob Watts
Member of the Trade: Chord Electronics
- Joined
- Apr 1, 2014
- Posts
- 3,154
- Likes
- 12,561
Yes Dave employs a SMPS, and all the DACs are absolutely full of switching regulators.
Using switching regulators started with the Hugo design, where battery operation forces one to use switchers. But I thought it would be a good idea to listen to the SQ damage that switchers generate, by comparing them to linear regulators. Before this time I used linear regulators exclusively, thinking the extra RF would cause problems. But when listening the switchers sounded a lot better - darker, warmer and smoother. I then discovered that there was two issues for the improvement by switching regulators - switchers are more efficient, so correlated noise levels at the PSU input side is much lower, and switchers force you to use more RF filtering localised to each circuit module.
Looking at RF filtering - we actually have two problems - one is random RF noise, the other is switching noise which is a fixed frequency (I always use fixed frequency regulators). When I design RF filters I check it out by running a SPICE simulation, and to do it properly you must characterise the PCB tracks, component internal resistance and inductance for capacitors and parasitic capacitance and resistance for inductors. When you do this careful simulation, you can get very close agreement to reality. The benefit of SPICE is that you can design filters to achieve unmeasurable levels of RF; so recently I wanted to design RF filters that had pV levels of voltage (a pico volt is 1 million times smaller than a micro volt so it's pretty damn small). What was curious was for switching regulators it was very easy to get pV levels when filtering - but with random noise it was an order of magnitude more difficult, requiring hugely more complex filtering.
And it's random noise that is the SQ killer, as this is responsible for noise floor modulation - a regular switching fixed frequency will not create noise floor modulation, but fixed intermodulation products outside of the audio bandwidth. The problem with the mains is that it is full of random RF noise, so you need extensive RF treatment on the PSU - something that I do post PSU as a matter of course.
So the idea that linear is best (no RF filters) against SMPS (lots of RF filters) is simply wrong. When you dig down it is very much more complex than one would imagine.
Using switching regulators started with the Hugo design, where battery operation forces one to use switchers. But I thought it would be a good idea to listen to the SQ damage that switchers generate, by comparing them to linear regulators. Before this time I used linear regulators exclusively, thinking the extra RF would cause problems. But when listening the switchers sounded a lot better - darker, warmer and smoother. I then discovered that there was two issues for the improvement by switching regulators - switchers are more efficient, so correlated noise levels at the PSU input side is much lower, and switchers force you to use more RF filtering localised to each circuit module.
Looking at RF filtering - we actually have two problems - one is random RF noise, the other is switching noise which is a fixed frequency (I always use fixed frequency regulators). When I design RF filters I check it out by running a SPICE simulation, and to do it properly you must characterise the PCB tracks, component internal resistance and inductance for capacitors and parasitic capacitance and resistance for inductors. When you do this careful simulation, you can get very close agreement to reality. The benefit of SPICE is that you can design filters to achieve unmeasurable levels of RF; so recently I wanted to design RF filters that had pV levels of voltage (a pico volt is 1 million times smaller than a micro volt so it's pretty damn small). What was curious was for switching regulators it was very easy to get pV levels when filtering - but with random noise it was an order of magnitude more difficult, requiring hugely more complex filtering.
And it's random noise that is the SQ killer, as this is responsible for noise floor modulation - a regular switching fixed frequency will not create noise floor modulation, but fixed intermodulation products outside of the audio bandwidth. The problem with the mains is that it is full of random RF noise, so you need extensive RF treatment on the PSU - something that I do post PSU as a matter of course.
So the idea that linear is best (no RF filters) against SMPS (lots of RF filters) is simply wrong. When you dig down it is very much more complex than one would imagine.
ekfc63
100+ Head-Fier
Which of the Zenith Mk3 and the Phoenix would you say has the greater impact on sound quality?
The Phoenix has a bigger and more noticeable impact. The Zenith MK 3 is good, but the differences are a little more subtle. If you can only get one, definitely go with the Phoenix
yellowblue
100+ Head-Fier
My dealer first didn´t want to sell a Phoenix to me because he thought that the Phoenix also would do some kind of upsampling and therefore shouldn´t have impact on SQ as I owned the M-Scaler already. How wrong he was! The Phoenix works fantastic in combination with the M-Scaler. In my chain I would say it has kind of the same impact as the M-Scaler. Double fun.
ekfc63
100+ Head-Fier
Sadly too late. A deal came up a little while ago for a Zenith and I made the move. WIll get a Phoenix nevertheless. I don’t really get along with the iPeng app so maybe I’ll sell the Zenith.
Last edited:
You’ll definitely like the Zenith. And then when you add the Phoenix, you’ll like everything more.
Also, use the Zenith Mk3 as a Roon core. I feel it’s the best user experience
ekfc63
100+ Head-Fier
Some of the reviews say that the Zenith works / sounds better when its not the Roon core. If I keep the Zenith I’ll probably end up with Roon.
I’ve read that too, that separating the server from the streamer makes everything sound better. I’m skeptical, as others feel there isn’t as much of a difference, but I have no firsthand knowledge.
In a moment of craziness, I once contemplated buying another Zenith Mk3, so one was the streamer and one was the server. Thankfully that moment passed...hahaha. I decided to add the EtherREGEN instead, as that would reclock my Ethernet as well (kinda like the Statement). No complaints.
Sonore optical rendu , sounds great, I tried sotm sms 200 ultra se special edition, it sounded amazing but the rendu sounded more real, less bright ( I like bright) but I couldn’t get away from the resolution of the rendu with the m scaler. Just brilliant. What’s next after all this.?
teknikk7
100+ Head-Fier
I’m a bit confused on how to output from foobar2000. I have a mix of 16/44 24/96 some higher res 32 bit files. I have USB going to MScaler, then dual bnc to TT2. I have upsampling maxed to 768. If I don’t convert files from foobar I sometimes run into issues with files not playing or playing with static. Should I convert everything to 44.1? That way I get the full 16X up sample? Kinda confused here. Any advice?
muski
100+ Head-Fier
- Joined
- Jan 25, 2007
- Posts
- 393
- Likes
- 447
Thanks for your post. I’m an ultraRendu user, and while I love it, I’m intrigued by the opticalRendu.
As earlier posts mentioned the Innuos Mk3 server, have you ever considered the SonicTransporter I9 Optical? I love the idea of a direct optical connection from Roon server to opticalRendu streamer. I’m very tempted by their server+opticalRendu combo package,
Cheers,
muski
Users who are viewing this thread
