Modding the DirectStream DAC MKI

Stimpy, if your goal was to mitigate the strength of the emitted magnetic field, that copper wrapping (if that’s what it is) isn’t going to do much. The emitted magnetic field from a toroid is axial (not radial), and copper. - even if it were to be placed to shield the axial line - is more or less ineffective. Annealed mu-metal is needed to dampen the typical magnetic field leakage from a 50/60 Hz transformer.

You are right. Magnetic field leakage (stray flux), reduced with high-permeability material (like Mu-metal).

But in almost all commercial toroidal transformers, the “flux band” is for electric field (capacitive) shielding, not magnetic. Because strain fiels are already very low. A copper shield dumps electric noise to ground but it has to be grounded ! Which I can’t see in the picture. @Stimpy7777 , make sure you have it.

.

Regarding the toroidal transformer Triad Magnetics VPM12-2080, it has a big advantage that you can’t DIY: the shielding between primary and secondary ! Plus, the band around.

Although, I see that this transformer won’t fit Stimpy’s situation where he’s still using 12V from the main power supply.

It’s not copper around the transformer it’s MU metal that is wrapped in a red plastic covering. I did use a piece of copper foil; but only to ground the metal air vent I added to the metal of the lid.

Yes, I know. That’s why I said “if”. :slightly_smiling_face:

Hmm. It would be more effective if a circle were cut slightly bigger than the diameter of the toroid, placed 5-6mm above the transformer. If you have any left, give it a try.

In purist terms (IMO), garbage. If it is mu-metal, putting holes through it essentially ruins its effectiveness. If it isn’t mu-metal, then it is entirely useless - if its intent is to control magnetic field leakage.

How about removing the stock Amgis 50va transformer and placing it in an external box, and or at the same time, replacing with 300 va laminated ECore, Paul always say larger transformers

sound better.

It looks like the newest version of the transformers has the Flux band made of (probably) steel. I just tested the Triad Magnetics VPM12-2080. It has it, too. Not MuMetal, but some sort of a help.

There are so many weaknesses in this DAC we have to address. The newest version of the transformers is not one of those. Although, there’s a chance you could find something better, but it’s a good quality transformer.

In that regard, Triad Magnetics is superior to the newest version (Amgis) because it has a shielding between the primaries and the secondaries.

I wasn’t actively looking for a replacement for the power transformer. I found that we could use that specific Triad Magnetics VPM12-2080 transformer by accident. I suggested to use it not because it sounds better (I don’t know that for sure, but I trust Sparkos who said they sound good), but to eliminate the heat issue that was brought up to everyone’s attention recently again.

A cap MuMetal shield. Even if any holes make them less effective, they still work. Do we need them here?

For the toroidal transformers, we don’t need that, the Magnetic field is already too low, and in the newest stock transformers, there is a flux band already implemented. I don’t know about the newer model, though. (There are three different transformers in PerfectWave and DirectStream DACs)

In my DAC, I added a copper shield between the two power supplies and the analog board, the processor board, and the Bridge II card.

On the other hand, the R-core transformers will benefit from magnetic shielding. They usually have a copper flux band, which is good. But they have worse magnetic field leakage than the toroidal transformers. That’s why I put a MuMetal cap on my R-core transformer for my PSU for the analog board.

I received my APS LESSP DAC back from APS. They did the new type vocm mod and installed a new lid. I am thrilled with the sound. It was wonderful before, but now it is astonishing. The bass is even more coherent and extended. The imaging is Phenomenal. The sound of the music permeates the space and feels incredibly authentic. This DAC is among the top upgrades I have ever made. I am planning on taking it to a friends who has the new PGM DAC, We will see how it performs.

Has anyone else attempted this new kind of vocm alteration?

That’s great!

If previous version of APS Vocm is with Tantalum Polymer caps and new version of Vocm is different, no wonder new mod brought something new to the table. I wish we know what they did. The only thing JKRichards said was that no Zeners directly in the Vocm position.

Could you post what’s inside your unit?

Wow. Excellent. I sent mine in for the same upgrade. Mine is supposed to be delivered here tomorrow. Can’t wait.

Looking forward to the comparison.

While I am waiting for my LESSP with the new Vocm mod to arrive I decided to catch up on post here Wow lots of great information

No contradiction. I said Polymer caps might work in sandwich. And it was in 2024 before I dived deeper into the matter.

Now, we know that placing Tantalum Polymer SMD into a high-speed analog feedback loop like the Vocm reference is a critical mistake. Because it results in:

  • Correction currents become sharper
  • No damping at all
  • This is where the risk of subtle loop interaction increases
  • May have high-frequency ringing

It may not oscillate — but it’s the source of additional distortion. Which leads to: edgy, bright, mechanical sound.

I thought it was a simple voltage divider circuit using resistors to create the virtual ground.

Because inside the AD8132, the VOCM pin is not just a passive DC bias input.It is part of an active common-mode servo loop that continuously controls the output stage..

You probably missed this

Heres a piece of schematics with the output transformers in Mk1.

You can measure DC offset between either of pins: 1-3, 3-4, 4-6, 1-6. It’s the same!

Got my DAC back today :smiley: I didn’t get much else done as I have been listening to music all day. Upon first listening, the sound is even more incredibly clear and detailed. The instruments’ colors and textures are all beautifully portrayed. On certain recordings featuring challenging-to-reproduce murmurs from the bass drum, this enhanced DAC handles them effortlessly. Double basses are robust and raw, Violin sections are large and substantial. The LESSP impresses with exemplary orchestral timing. The texture of cellos are remarkably realistic, the overall display is highly energetic, and I have seldom experienced this sound at such a high quality. I’m quite pleased with what I have heard up to this point. The upgrade changes are worth every penny. I believe it will improve even more with adequate break-in time. Thank You Ted Smith & APS audio!

From what I remembered on my MK2 APS mod, the unit felt quite hot at the beginning (where the transformer located). It took weeks for it to cool down to normal. The SQ took a week to sound great, and a lot longer to sound its best. So, I am sure you will hear more improvement later from MK1 too.