RME DPS-2 Linear Power Supply

[EERecordist]

If you look at the RME product line and distribution system, it is highly unlikely in my opinion, they would enter the hifi market. Their ADI-2 pro series have microphone inputs. Some of their other products with more processing power already have room-eq software and free DAW software.

I would guess they might want to expand into the stacked bit depth approach planned by commettitor and ASR member Millennia/imersiv, or partner to provide AES digital microphone technology in the German pro audio ecosystem. A digital microphone doesn't need an external preamp+ADC, RME's current business.

They are a very small company that executes well on the engineering side.

 

[VintageFlanker]

Out of interest, and to educate myself, may I ask you to further elaborate why my statement is wrong?

Because while it should be the case, it clearly isn't in all scenarios. First, I'm not talking about what you would hear, but what could be measurable.
I measured quite a few products showing 50Hz hum through XLR regardless. It was sometimes extremely minimal, but still there all the same :

It may sometimes be much, much worse :

Recent instances form @amirm :

So, "there's no hum using XLR" is a wrong statement.

Their ADI-2 pro series - which have microphone inputs - already have room-eq software

Oh ? I'm not aware of such things.

 

[markstein]

Thank you for these examples VintageFlanker, appreciated! Regarding the excellence of RME, though, I would expect little impact of such a linear power supply on the measurements of their ADI-2 lineup.

In other words: The ADI-2 certainly will not audibly hum when using balanced connections. Did not find measurements to verify this statement.

Btw: the ADI-2 has no microphone inputs, does not feature TotalMix and RoomEQ. Currently RoomEQ is limited to specific Firface UFX, UCX and - hopefully soon - 802 fs units.
The larger Fireface units do have internal power supplies, so this linear PSU is not intended to be used with them...

 

[Sokel]

Straight from RME here and here.

To make it easier I quote the two posts:

@MC_RME will just point to the manual where we clearly made our point about power supplies.

Ground issues, a ground loop, or no ground aka leakage current issues, or leakage with wrong grounding - all this could be cured by a linear power supply. But this is noise that shouldn't be there and is clearly audible as such. It's not some mysterious bass, mid, treble, soundstage, veiled issue. A different ps will not change the basic sound of an ADI-2 unit at all.

The leakage current of the ADI's external power supply runs to ground, whereever it finds one. With nothing connected you (your body) is the ground. The phones are in the way and therefore emit audible hum. Touching the chassis short-circuits the path of the leakage current, so it won't flow over the headphones anymore. Solution is the same as always: either connect a grounded device to the ADI, or directly connect it to ground.
Because of cases like yours we changed the power supplies to grounded ones. But as we have a lot of stock currently only the 2/4 Pro comes with it. And even that one will fail in case your power outlets don't have PE/ground (like many in Japan, Thailand...).
The last resort (if someone is unwilling or unable to connect it to a solid ground) is the use of a linear power supply. Their leakage current is basically free of harmonics and therefore typically inaudible.

(I bolded the interesting parts)

So we now have couple of explanations.

 

[EERecordist]

Btw: the ADI-2 has no microphone inputs, does not feature TotalMix and RoomEQ. Currently RoomEQ is limited to specific Firface UFX, UCX and - hopefully soon - 802 fs units.
The larger Fireface units do have internal power supplies, so this linear PSU is not intended to be used with them...

Thanks, corrected.

 

[Brian Hall]

One important reason why this new supply exists is that for precision measurement purposes the stock SMPS supply had too much mains leakage current which could affect measuring unbalanced gear. The high mains leakage current also caused some irritation in other scenarios.
And while they were at it, they also improved the supply voltage signal itself because for the sake of engineering.
Finally, it's also an excellent supply for other devices.

Why not just buy a better and cheaper DAC that doesn't have that issue?

 

[KSTR]

Why not just buy a better and cheaper DAC that doesn't have that issue?

For precision development work and precision measurements (of all sorts -- not only audio) there simply is no better DAC/ADC/Amp combo than the ADI-2 Pro and ADI-2/4, given the features, full remote control and all.

To avoid any chance of even the smallest "ground loop" type of problem being introduced by the measurement rig, it must be fully galvanically isolated.. For this, USB must be isolated (and/or TOSLINK being used) and the supply must be isolated as well. I'm using an USB isolator box and I modified the stock supply to fulfil this requirement.

The stock supply, while being two-prong and thus "isolated" in principle, is not that well isolated from the mains, just enough to pass the regulatory requirements (like any other supply of this kind). The new supply offers an isolated mode. Of course, I haven't used or seen the new supply so I can only guess how it actually behaves but I have zero doubt RME engineering is state-of-the-art.

Reminder, this is not a problem of the DAC proper, it's a potential system setup issue from the supply in the way it is coupled to the mains grid. It is, however, not a problem of the supply voltage itself the supply provides to the unit (people need to understand the difference).

 

[antcollinet]

Because while it should be the case, it clearly isn't in all scenarios. First, I'm not talking about what you would hear, but what could be measurable.
I measured quite a few products showing 50Hz hum through XLR regardless. It was sometimes extremely minimal, but still there all the same :

View attachment 365607

View attachment 365608

It may sometimes be much, much worse :
View attachment 365609
View attachment 365610

Recent instances form @amirm :
View attachment 365645View attachment 365646View attachment 365647

So, "there's no hum using XLR" is a wrong statement.


Oh ? I'm not aware of such things.

To build on this:

First XLR will only help with noise picked up on the interconnect between sender and receiver (typically ground noise, but could also be radiated interference). Any noise generated within the electronics on either side of the interconnect is not impacted. Having said that - audible noise generated in the electronics is rare (and is very poor design). Audible noise from ground loops is common.

Secondly - it is never perfect - it dramatically attenuates the noise rather than killing it alltogether. (Typically - from what I've read - around 40dB attenuation with well designed kit)

 

[wwenze]

To be honest I expected the manual to show differences in measured performance in XLR/RCA/HP out and ADC in. If there are improvements, it would be a selling point, and let the owner choose between price x marginal performance improvement.

There is no successful attempt at creating a measuring standard for measuring common-mode noise performance.

What should they show, show something that says their product is flawed while the others don't show that?

 

[Scytales]

To build on this:

First XLR will only help with noise picked up on the interconnect between sender and receiver (typically ground noise, but could also be radiated interference). Any noise generated within the electronics on either side of the interconnect is not impacted. Having said that - audible noise generated in the electronics is rare (and is very poor design). Audible noise from ground loops is common.

Secondly - it is never perfect - it dramatically attenuates the noise rather than killing it alltogether. (Typically - from what I've read - around 40dB attenuation with well designed kit)

To build on your building, one can read this very... balanced articles published by Rod Elliot on his website, as well as the reference articles in their respective bibliographies, which all teach about the challenges in designing proper balanced output and input stages and tackle some issues rarely addressed in the more commun technical litterature (even in Doug Self's books) :

Balanced Line Driver with Floating Output
Design of High-Performance Balanced Audio Interfaces
Balanced Input and Output - The Things No-one Tells You

 

[antcollinet]

There is no successful attempt at creating a measuring standard for measuring common-mode noise performance.

What should they show, show something that says their product is flawed while the others don't show that?

To clarify that a little.

There are no differences unless there is a source of external noise injection into the system.

The difference between balanced and unbalanced is in their ability to reject external noise, and nothing to do with their inherent (internal) performance. (ignoring typical voltage levels).

 

[Chrise36]

Because while it should be the case, it clearly isn't in all scenarios. First, I'm not talking about what you would hear, but what could be measurable.
I measured quite a few products showing 50Hz hum through XLR regardless. It was sometimes extremely minimal, but still there all the same :

View attachment 365607

View attachment 365608

It may sometimes be much, much worse :
View attachment 365609
View attachment 365610

Recent instances form @amirm :
View attachment 365645View attachment 365646View attachment 365647

So, "there's no hum using XLR" is a wrong statement.


Oh ? I'm not aware of such things.

It would be interesting to see measurements while powering the DUT with a battery setup eg router ups or a double conversion ups.

 

[antcollinet]

It would be interesting to see measurements while powering the DUT with a battery setup eg router ups or a double conversion ups.

Even so - you won't measure the ground noise at the output of the DAC, but at the other end of the interconnect (ie the input to the amp).

And if you are only testing the DUT (without the rest of the system) then any external noise injection is a fault of your test setup.

 

[wwenze]

It would be interesting to see measurements while powering the DUT with a battery setup eg router ups or a double conversion ups.

It gets more complicated than that. Portable amp headphone setups can hear the GSM signal coming... despite the whole thing floating

Galvanic isolation from mains doesn't solve it either, since the transformer secondary is already isolated from mains. The low voltage side is always floating, usually at around half of mains AC voltage, while the power supply voltage is being generated from magnetic field.

 

[antcollinet]

It gets more complicated than that. Portable amp headphone setups can hear the GSM signal coming... despite the whole thing floating

Galvanic isolation from mains doesn't solve it either, since the transformer secondary is already isolated from mains. The low voltage side is always floating, usually at around half of mains AC voltage, while the power supply voltage is being generated from magnetic field.

Depends on the device. Those with class 1 insulation will have the housing connected to mains ground. More often than not, they'll also have the electronics grounded to that also.

 

[Chrise36]

It gets more complicated than that. Portable amp headphone setups can hear the GSM signal coming... despite the whole thing floating

Galvanic isolation from mains doesn't solve it either, since the transformer secondary is already isolated from mains. The low voltage side is always floating, usually at around half of mains AC voltage, while the power supply voltage is being generated from magnetic field.

I have noticed this with gsm and wifi router near my hp amp. Would a better shielded rca cable make a difference? Power and integrated amps are susceptible to smps noise from the mains though. In the case of using a battery powered DAC the only noise should come from the battery's electronics or not?

 

[Ken Tajalli]

Let us agree on the following: I do not notice any hum with my equipment, using balanced connections.

Out of interest, and to educate myself, may I ask you to further elaborate why my statement is wrong?

I have a 20+ year old, single ended Meridian 556 poweramp, and I can hear NO hum or noise with my ears to the speakers, at night, in a dead silent room.
Meridian had quoted a noise figure of -90dB for the amp.
All this talk regarding audible noise and hum is a bit over rated.

 

[Salt]

Just a question of time this would have happened, and as RME is always aware of all circumstances of real environment they promoted the WAF by including
https://www.audiosciencereview.com/forum/index.php?attachments/sr-pink-jpg.365355/,
or not?

 

[wwenze]

I have noticed this with gsm and wifi router near my hp amp. Would a better shielded rca cable make a difference? Power and integrated amps are susceptible to smps noise from the mains though. In the case of using a battery powered DAC the only noise should come from the battery's electronics or not?

Yes and no depending on situation and in majority of situations it's probably no

Yea HP amps are particularly sensitive since unlike speakers with a sensitivity of 86dB @ 2.83V, headphones... I just randomly picked Grado SR 325i , is 117.90 dB/V

Shielding makes sense when you have a very solid ground voltage that doesn't move. So you can protect the signal, usually by wrapping it with the ground.

But when you setup has common-mode noise problem, it already suggests your ground voltage between source and load is unequal, and the signal - which is being referenced against that ground voltage - will be interpreted incorrectly. In majority of such cases, the cause of noise isn't due to noise getting into the signal, so cable construction has limited benefits.

 

[johny_2000]

For a thousand-plus dollars, any OCD audiophile can buy 12V car batteries and a charger for the next decades. This will solve all these problems together.