Devialet Chat

(28-Mar-2018, 21:16)Confused Wrote: [ -> ]On this subject, it only dawned on me recently that my entire system is SMPS powered! Oh the horror.  Power is as follows:

Computer - SMPS
Router - SMPS
Modified switch - UptoneAudio LPS-1, which itself is powered by SMPS "energiser"
SOtM kit - sPS-500 - which is a SMPS unit, albeit a very trick one.
Mutec MC3+USB - SMPS
Devialet Expert Pro - SMPS x 2

I wouldn't put it that drastically as Greg but yes he's right. And I made the experience many, many times that a texture of digital edge in the sound went away by pulling SMPSs out of the mains.
But there are some quiet good ones and there are bad and very bad ones !
Obviously the SMPS inside our Devialets is a good one and even nessasary to provide it's speed combined with power. The SOtM SMPS should be a good one too. And I wouldn't concentrate on the SMPS in your audio at the moment as you couldn't change it now. I would consider to concentrate on any other cheaply SMPS in the room. It's really no problem to pull off a lamp/charger/wireless phone/or anything else if it's not needed for audio. Don't make a hassle to compare before/after. Just do it before your long listening session next weekend and enjoy the music...

gui

I would be more concerned about the stuff a SMPS sends down the DC line in AC noise going through the capacitor coupling the AC to the DC side. If you follow John Swenson at Computer Audiophile he measured alot of SMPS and some stuff gets thrown back to the AC line but the high impedance leackage that is harmonics from AC 50/60Hz and gets in the DC line and travels along all sorts of cables and might end up where you dont wont it is a bigger problem. Some say when this leackage is removed from a system this digital edge/harshness is also removed but then again the number of different SMSP and audiosystems is so big that there are no real way to tell how things will behave

I have cleaned up my system from all SMPS and on my dedicated line to the stereo I only have DAC, amplifier and 2x Paul Hynes SR4 all connected to a stargrounded Ansuz D8 Mainz. A second dedicated line goes to power a HD Plex 200W PSU that powers ISP fiber, router, switch (Netgear GS108 with grounded -0V) and a NUC with Roon ROCK.

Also it's true to say there are good and bad linear PSUs. "SMPS bad, linear good" is an easy and obvious generalisation but like many generalisations (ha!) is not very helpful.

Some very interesting posts re SMPS's. To be clear, I am totally on board with addressing this as a potential issue. One point specific to my system, I do not have all devices on one mains spur, in fact they are on three. The PC and router are about 6m away from the hifi kit, connected by 10m of Ethernet cable. The rest of the hifi kit is spread between two mains outlets, which are probably about 5m apart, although I think these are probably on the same ring.

I like @Greg 's ferrite core clamp idea, it should be very easy to do and very low cost. Well worth a try. Of course @yabaVR 's SMPS test is even easier to try and free, but he has mentioned it again, so I will have to move it down yet another position on my 'to do' list. (actually, as things happen, I did have some plans for the Friday and Saturday of the Easter break, it looks like these plans have fallen through, so unexpectedly I find myself with far more free time that I expected this weekend, so maybe this test will be done soon.)

In terms of eliminating SMPS's completely, I have been thinking of getting a pair of Paul Hynes SR4's to power the SOtM kit. It would be good to get @octaviars view on the SR4 with his SOtM items. Elsewhere, the SR4 appears to be gaining a fine reputation and is relatively good value. This would leave me with the SOtM sPS-500 spare. This is a very clean SMPS in terms of power delivery and can deliver 5 amps, so one thought I had is that I could use this to power my PC. One issue here is that I am not 100% convinced that the SOtM kit will be staying in my system. It probably will stay, but I do not want to start throwing more cash on LPSU's to power something that could possibly be for sale on eBay before the end of the year.

As for the SMPS in the MC3+USB, this would require surgery to replace. I had mentioned before about rumours of an 'Empyreal' class MC3+USB, which would presumably include a LPSU similar to the REF10. There has been zero news of this recently though, and the rumour seams to be based on the fact that the REF10 is currently the only product in Mutec's Empyreal class. An announcement at the Munich show maybe? (I'm ever the optimist.....)

Picking up on @thumb5 's point, there was an interesting comparison of the performance of various power supplies on Computer Audiophile. The interesting thing was just how well some SMPS's perform relative to LPSU's, so yes, you cannot generalise here. As another generalisation buster, one of the nicest to listen to, least digital sounding systems, that I listened to at the Bristol show recently had class D power amps. (Mola Mola) I don't know how many times I have read and heard people say that class D is unlistenable, and lets say that Mola Mola system did have a slight digital edge, everyone would be saying, "well it would sound like that, it's class D".

So I agree there is much to investigate, and potentially much to gain here. That said, I still feel that all this might be missing the point completely. The key issue I have been posting about is a change in the tonal presentation of my system between the mR and SOtM kit. Some say this is due to the additional transparency of the SOtM kit. So why is it more transparent? The most common answer to this seams to be that the SOtM kit has better noise suppression. So the additional noise suppression has tilted the tonal balance away from something I enjoy, the solution to this is to then eliminate SMPS noise. This does not follow logically. OK, I may have a system now that would more than ever benefit form eliminating noise from SMPS's and similar, but I think something more is going on here. As I said in an earlier post, with decently recorded material the system is sounding quite superb, so whatever issue the work of the devil SMPS's are causing, it is hard to hear it when the source material is good, and does not bother me with the mR in place. Yet somehow switching the mR for the SOtM kit has made the system brighter, or at least apparently brighter. With the mR in place something can be enjoyable to listen to, with the SOtM kit less so, nothing else needs to change, the mR is surrounded by SMPS's as well, (at one time the mR was SMPS iFi powered and sounded OK) the amps, speakers, PC, nothing else changed. Imagine getting a favourite album, but one that is not the best actual recording quality. Listen to it with the treble turned up +5dB, it will probably be very difficult to listen to, this is irrespective of how many SMPS's one might have. Also note my car rattle analogy in post #74, it is not like the system has a fundamental problem.

The good news is that I do have quite a few ideas as to how to investigate the issue, further understand it, and identify what to do, plus a bit more free time than expected this weekend. There are a few things I want to try, systematically and logically. It is a case of information gathering, quite what this information will tell me remains to be seen. (or heard)

Quote:In terms of eliminating SMPS's completely, I have been thinking of getting a pair of Paul Hynes SR4's to power the SOtM kit. It would be good to get @octaviars view on the SR4 with his SOtM items. Elsewhere, the SR4 appears to be gaining a fine reputation and is relatively good value.

@Confused I have been runing my SOtM kit and two SR4 for a couple of weeks now (they are powered on 24/7) and I am very pleased with the performance. I do see many talk of the digital glare that some endpoints have but I cant say that in my system this was present with the sMS-200ultra and if there was a bit at some tracks to me the tX-USBultra have made this go away. I did try the album that @Confused wrote he had some problems with but to me I could not find it playing in my system. If this only are an effect of the total different gear (my TAD electronics/TAD speakers VS. his Devialets and KEF) or perhaps the SR4 brings some to the table that is just doing it right.

As we have so different systems it is hard to say what exactly is doing what as we only have sMS-200ultra and tX-USBultra in common. To me the SOtM/SR4 combo has truly made my TAD speakers sound great

Something comes to my mind here: it's very easy to get hung up about a particular aspect of your system's sound, and become super-sensitive about it.  Meaning that you get into a frame of mind in which it's hard to listen without hearing the very thing you're listening for.  As part of your series of experiments, @Confused, are you able to make some objective measurements of frequency response to determine whether there is any measurable difference between the SOtM and mR?  That would be an interesting thing to know for sure.  Simply finding that there is no measurable difference -- if that's the outcome -- might make it go away subjectively, too.

Sometimes it is best to stop overanalyzing everything and just listen to some good music......you never know when a new wave of audiophile nervousa might strike you

(29-Mar-2018, 12:09)thumb5 Wrote: [ -> ]Something comes to my mind here: it's very easy to get hung up about a particular aspect of your system's sound, and become super-sensitive about it.  Meaning that you get into a frame of mind in which it's hard to listen without hearing the very thing you're listening for.  As part of your series of experiments, @Confused, are you able to make some objective measurements of frequency response to determine whether there is any measurable difference between the SOtM and mR?  That would be an interesting thing to know for sure.  Simply finding that there is no measurable difference -- if that's the outcome -- might make it go away subjectively, too.

I do agree, this was exactly what I was referring to with my car rattle analogy.  That said, there is most defiantly a shift in the perceived tonal balance, it is too big to ignore.  I did have a visitor last weekend, when they arrived I was playing something, I can't remember what exactly, and they actually commented with something along the lines of 'that sounds a bit grindy' (non-audiophile speak) So not imagined I think.  (this sounds like one of those user reviews:  my wife was in the kitchen, and without knowing that that I had changed the cable she came into the room and asked 'have you bought a new hifi dear, that sounds amazing') 

As for objective measurements, I did actually attempt this last weekend.  I had a collection of test tones at various frequencies, and was armed with both an iPhone with Decibel X app and a professional sound pressure test meter that I managed to borrow from work.  The idea was to keep the amp at a set fixed volume, test meter in the listening position, run through the tones and check the SPL.  This could then be repeated with the mR in the system.  This almost worked.  The problem was that you could do the test, then repeat it immediately afterwards with no change to the kit whatsoever, and get slightly different readings.  For this to produce a valid result, it would need to be repeatable to within 1 dB, but this seamed to be impossible.  Even the professional meter seamed to behave very oddly with test tones.  It might be reading 75dB, and just walking behind the meter would cause the reading to move maybe 3 dB, walk away and it would hold the reading at the changed level, most peculiar!  Interestingly, the iPhone Decibel X app displayed similar behaviour.  So the final data lacked proper integrity, it did hint at the mR being slightly 'darker', but I knew that the accuracy was not good enough and I might be reading the results with a degree of bias.  What I would really like to do is some REW curves, but I do not believe this is possible via the mR or SOtM kit.  If it is possible, could someone tell me how!

I did think of trying the above test again, with the test meter maybe 4" from the tweeter, but I suspect the accuracy will still not be good enough, and this also becomes a less relevant test.  Another thought is to use REW just as an SPL meter with my UMIK-1 microphone.  Maybe I will try this again if I am in the mood.  Generating an REW curve would be much better though.

Oh, and a final note to @octaviars, I would refer you to my comment in post #79 "This weekend I will commence step 1.  Code name "operation listen to lots of music."

Quote:Oh, and a final note to @octaviars, I would refer you to my comment in post #79 "This weekend I will commence step 1.  Code name "operation listen to lots of music."

That is a good thing to be doing from time to time, just play music and dont bother so much with the technical stuff

I do recall when I used the sMS-200ultra with the original SMPS that I felt some tracks being perhaps a bit edgy, that have all gone with a couple of hundred hours on it and the SR4 powering it.

My old system consisting of Devialet 440Pro and Focal Sopra 2 was more prominent to show this edgy/harsh behavior with some albums (streaming with Roon/HQ player/AIR) but then there are so big diffrences between the tweeter/midrange of Focal and TAD how they sound.

(29-Mar-2018, 13:37)Confused Wrote: [ -> ][quote pid='73726' dateline='1522321756']
As for objective measurements, I did actually attempt this last weekend.  I had a collection of test tones at various frequencies, and was armed with both an iPhone with Decibel X app and a professional sound pressure test meter that I managed to borrow from work.  The idea was to keep the amp at a set fixed volume, test meter in the listening position, run through the tones and check the SPL.  This could then be repeated with the mR in the system.  This almost worked.  The problem was that you could do the test, then repeat it immediately afterwards with no change to the kit whatsoever, and get slightly different readings.  For this to produce a valid result, it would need to be repeatable to within 1 dB, but this seamed to be impossible.  Even the professional meter seamed to behave very oddly with test tones.  It might be reading 75dB, and just walking behind the meter would cause the reading to move maybe 3 dB, walk away and it would hold the reading at the changed level, most peculiar!  Interestingly, the iPhone Decibel X app displayed similar behaviour.  So the final data lacked proper integrity, it did hint at the mR being slightly 'darker', but I knew that the accuracy was not good enough and I might be reading the results with a degree of bias.  What I would really like to do is some REW curves, but I do not believe this is possible via the mR or SOtM kit.  If it is possible, could someone tell me how!

[/quote]
Confused, If you have 2 computers it's not that difficult to generate the REW curves through the whole chain (once through mR and once through SOtM kit) in order to compare them.

@SwissBear was once kind enough to provide me with the process:

1. Use the Generator function of REW to generate a measurement sweep and
save it as .wav. Repeat the procedure for each of the speakers and do not forget to include the timing signal.
2. Import these .wav into a player on your system (JRiver for instance)
3. On the machine with the measurement REW, select a fake output for REW
4. Select 'Measure' and make sure you have the same
parameters (sweep length, aso) you selected for generating the sweep. Also select 'Use acoustic
timing reference'. Click measure. The system will wait for the timing signal.
5. Play the recorded sweep. The system will measure the recorded sweep,
played through your complete system, including correction impulses in
case you have some.

Unfortunately I am not now able to upload the screenshots showing this