[DeckHiFi]

Wow, for anyone that uses this dac with HQplayer and possibly missed how to get no noise/hiss in background using DSD...page 307 on this thread is a life saver. I wish I found that page sooner. Thanks guys for going over how to get no hiss. This is incredible. No wonder people love DSD. Would be a good thing to add to first page even though I know it's not really HQplayer focused usually. I do appreciate all the helpful stuff in this thread and the first post though. Thanks everyone!

 

[Jake2]

Found it: https://lessemf.com/product/magnetic-shielding-foil/

Looks good - though to be honest on the face of it it's more similar than different to the product I linked with 80% nickel (other components not specified) and 0.004" = 0.1mm thick, vs for my one 0.15mm thick 80% nickel supposedly MuMetal.. "conforming to ASTM A753 Alloy 4, MIL-N-14411C Composition 1, and 1J79" - whatever that means.

On closer review of their links and language for the one I got, it appears, despite their use of the term MuMetal, it's likely it's not actually official MuMetal, but some equivalent/similar composition from a different and cheaper source than the MuMetal trademark owner Magnetic Shield Corp who're quite protective of their trademarked name. The range of linked specs are all similar to MuMetal but different names e.g. MuShield

https://www.mushield.com/material-sales/hymu-80-magnetic-shielding-alloy-astm-a753-alloy-4/
ASTM- A- 753, Alloy 4
A.K.A. Hipernom, HyMu 80, Moly Permalloy, Magnifer 7904; is a soft magnetic nickel-iron alloy with about 80 % nickel, 4.2 – 5.2 % molybdenum, a saturation induction of approximately 8,000 Gauss, the highest technically obtainable permeability,( Max >350,000) and a very low coercive force.

Not sure that helps at all - other than to say there's a lot of suppliers with very similar materials out there!

EDIT: I see the spec sheet for your one refers A753, type 4! Snap!

 

[BShaw]

Looks good - though to be honest on the face of it it's more similar than different to the product I linked with 80% nickel (other components not specified) and 0.004" = 0.1mm thick, vs for my one 0.15mm thick 80% nickel supposedly MuMetal.. "conforming to ASTM A753 Alloy 4, MIL-N-14411C Composition 1, and 1J79" - whatever that means.

On closer review of their links and language for the one I got, it appears, despite their use of the term MuMetal, it's likely it's not actually official MuMetal, but some equivalent/similar composition from a different and cheaper source than the MuMetal trademark owner Magnetic Shield Corp who're quite protective of their trademarked name. The range of linked specs are all similar to MuMetal but different names e.g. MuShield

https://www.mushield.com/material-sales/hymu-80-magnetic-shielding-alloy-astm-a753-alloy-4/
ASTM- A- 753, Alloy 4
A.K.A. Hipernom, HyMu 80, Moly Permalloy, Magnifer 7904; is a soft magnetic nickel-iron alloy with about 80 % nickel, 4.2 – 5.2 % molybdenum, a saturation induction of approximately 8,000 Gauss, the highest technically obtainable permeability,( Max >350,000) and a very low coercive force.

Not sure that helps at all - other than to say there's a lot of suppliers with very similar materials out there!

EDIT: I see the spec sheet for your one refers A753, type 4! Snap!

All good Jake, all I can say is when I measured EMF with and without by placing the foil between the source and the meter, it absolutely murdered the emmisions. Not debating the stuff you've suggested, just another option. Actual mu metal is supposedly very tricky stuff and often not authentic. When I've seen the real stuff, it's used in flat pieces because if it's bent it has to be re-tempered ("perfect temper"). So yeah, most likely they're the same or very similar and possibly not actual mu metal as you say.

 

[m-i-c-k-e-y]

Would be a good thing to add to first page even though I know it's not really HQplayer focused usually.

Added.

 

[amrutmhatre90]

Wow, for anyone that uses this dac with HQplayer and possibly missed how to get no noise/hiss in background using DSD...page 307 on this thread is a life saver. I wish I found that page sooner. Thanks guys for going over how to get no hiss. This is incredible. No wonder people love DSD. Would be a good thing to add to first page even though I know it's not really HQplayer focused usually. I do appreciate all the helpful stuff in this thread and the first post though. Thanks everyone!

Welcome to the DSD club,
What are your settings?

 

[Jake2]

All good Jake, all I can say is when I measured EMF with and without by placing the foil between the source and the meter, it absolutely murdered the emmisions. Not debating the stuff you've suggested, just another option. Actual mu metal is supposedly very tricky stuff and often not authentic. When I've seen the real stuff, it's used in flat pieces because if it's bent it has to be re-tempered ("perfect temper"). So yeah, most likely they're the same or very similar and possibly not actual mu metal as you say.

I probably went a bit to town with the research and links etc, was just trying to make sense of confusing if similar sounding names. Hopefully useful to someone.

Main thing is our sheets are very effective like you say. Look forward to hearing your impressions of using it in a component.

I reckon a prudent rule of thumb to avoid reducing its effectiveness along the lines you mentioned, is to minimise bending and stressing the metal and try to make nice clean cuts without bending the surrounding metal. I’ll be more careful with my second attempt - will keep you posted.

Per this 2020 article, it sounds like modern fabrications are maybe a little less sensitive than they used to be…

Magnetic shielding alloys have sensitivity to mechanical shock – bending, forming, even severe flexing may give some reduction in the alloy’s magnetic permeability. But modern, vacuum-refined alloys have a lower sensitivity to shock, and normally withstand regular handling without significant loss of properties. Another consideration is cutting the alloy without affecting the alloy properties and degrading the performance. Cutting by shearing, EDM, waterjet, photo-chemical etching, or blanking dies typically only affects that portion of the alloy immediately adjacent to the edge, and the shield will exhibit normal shielding performance. If the alloy had already received its final magnetic anneal, re-annealing should not be necessary.

 

[camrector]

Wow, for anyone that uses this dac with HQplayer and possibly missed how to get no noise/hiss in background using DSD...page 307 on this thread is a life saver. I wish I found that page sooner. Thanks guys for going over how to get no hiss. This is incredible. No wonder people love DSD. Would be a good thing to add to first page even though I know it's not really HQplayer focused usually. I do appreciate all the helpful stuff in this thread and the first post though. Thanks everyone!

This didn’t work for me. Constant hiss while playing native dsd albums (DSD64-DSD512) from my library.
Just turning DSD direct on without any track loaded and hiss was present. Even when I got HQplayer and followed that it didn’t work. Granted this was some time ago and on a waaay earlier Gustard firmware
Glad it worked for you.

 

[Jake2]

OK, last MuMetal post for now.. measured, cut and fitted a shield for the second transformer and tidied up the first one. Measured and cut some caps/lids.

Approach & Observations:

• Quite a fiddly exercise but not technically complex and sonically rewarding
• Began by measuring dimensions of transformer with a ruler and cutting a piece that will be slightly deeper than the trafo is high with a length that is 3-4cm more than the circumference
• Then placed the piece loosely taped slightly larger than the transformer size gently in location over the transformer shimmying it halfway down past the wires then carefully identifying the exact location of all wires exiting the toroidal transformers and marking them on the cyclinder to then remove and cut cutouts at the bottom around 5mm deep
• even cutting the MuMetal carefully with scissors imposes visible stress marks which must have some (hopefully minor) effect on the MuMetal, ideal would be a long guillotine cut or better yet a specialist shear press (not tin snips though some long straight ones might be a bit better than scissors)
• This 0.15mm thickness is the max one could comfortably cut with normal scissors. Curved cuts are too hard and I can see they stress the metal more than straight cuts which is why I left the caps as hexagons for now
• Had to pay a lot of attention to taping up sharp cut edges, especially corners, with foil tape and//or electrical tape to ensure it doesn’t shear through power cables when one is shimmying it back in. Sanding ‘em back is an alternative suggested approach I didn’t do.
• Fitting the cylinders back in is a careful task as the wires are a very tight fit in places, especially the protective tape and with the transformer at the back by the IEC inlet - needed a pen or screwdriver to hold/push some taut wires out of the way enough to squeeze it past.
• Two layers on the lid (I had a third leftover to double up) sounds better than one
• Initial inpression is the sound is both smoother and more refined in vocals and punchier in say the leading edge attack of strings while bass from drums has more weight.

Phew.. that’s enough unplanned tweaking this week, fingers crossed it continues to sound good. If so I’ll definitely be buying for for my other gear with toroidals. Now need to get back to listening to a couple of USB regenerators…

 

[BShaw]

Nice work Jake I’m tempted to do this myself but I’m saving for a Spring KTE, and when I finally get it I’ll do the same with the transformers. Have you considered surgical scissors for the circles and tighter cuts?

 

[vvume]

Anyone using or used a Denafrips Gaia with the R26 using HDMI/I2S? If yes can you confirm it works and what mode you use(d) on the Gaia for the I2S configuration?

 

[Jake2]

Nice work Jake I’m tempted to do this myself but I’m saving for a Spring KTE, and when I finally get it I’ll do the same with the transformers. Have you considered surgical scissors for the circles and tighter cuts?

I’ll have a look into tin snips I think at my local hardware store. The thing is even with a fairly clean cut you can see the metal stress ripples spread up to 2-3cm/1” or more away from the cut edge. So assuming it reduces the shielding effectiveness slightly in that zone I suspect the lesser of two evils may be to leave a buffer around the circumference and do fewer easier and cleaner straight cuts rather than attempting cosmetically perfect and geometrically exactly matching but difficult-to-do-cleanly round cut. I might try the latter and compare it to my spare hexagon, see if I can hear a difference.

 

[Stellabagpuss]

Good work Jake, you inspired me to do a similar work. Although l'm going a different route, lron covers, and cover in 3M CN3190 copper nickel tape. l will probably use ultraperm 80 for the bottoms.

 

[Jake2]

Good work Jake, you inspired me to do a similar work. Although l'm going a different route, lron covers, and cover in 3M CN3190 copper nickel tape. l will probably use ultraperm 80 for the bottoms.

Cheers Damien. Incidentally I read that distance - which you’ve achieved I understand by removing em from the chassis - achieves the same thing as shielding as magnetic fields drop off fairly quickly with distance. That said shielding can’t hurt and is very satisfying to do. I just ordered 4 more sheets from the same place. Your alternate route/materials sound interesting.. be cool to hear how you get on.

 

[Stellabagpuss]

Yes moving the transformers gave a instance gain.
You inspired me to do some reading up, and by chance l found a link ,by a chap who used the method, l'm going to go down.

MuMetal is great material, but l think the trick is to use Nickel and Iron, so it's attracted to the magnetic waves. So the wave stays incased within the barrier.

Apprantly, apart from using a superconductor, you can't stop magnetic waves, but the methods where both employing, will work, and reduces the spread on to other components.

Thanks for sharing Jake, you inspired me to do some research. l just have to wait for the spares to arrive.

 

[Jake2]

Yes moving the transformers gave a instance gain.
You inspired me to do some reading up, and by chance l found a link ,by a chap who used the method, l'm going to go down.

MuMetal is great material, but l think the trick is to use Nickel and Iron, so it's attracted to the magnetic waves. So the wave stays incased within the barrier.

Apprantly, apart from using a superconductor, you can't stop magnetic waves, but the methods where both employing, will work, and reduces the spread on to other components.

Thanks for sharing Jake, you inspired me to do some research. l just have to wait for the spares to arrive.

Cheers again. I was aware of the basics - that you can’t block magnetic field lines, just redirect em by providing them a more attractive path via high permeability material like nickel based alloys or certain types of iron, and the fact that many high permeability materials saturate quickly so are better used in layers or in lower strength fields - but hadn’t read too deeply on it. After reading a bit more last night I have mild buyers remorse as I see 20% iron ‘Permalloy’ (80% nickel, 20% iron) has even higher permeability* than MM (typ. 80% nickel, 15% iron, 5% molybdenum) and is available on AliX at similar cost to the cheap off-brand MM I got.

Fascinating though I find this topic, it’s a little OT from the R26, I might PM you if that’s OK.

*EDIT - reading a few more sources it seems the relative permeability of MuMetal and the Ni80Fe20 Permalloy (they’re all in a general class of highly permeable alloys but helpfully the term ‘Permalloy’ was trademarked by Bell labs a century ago for the Fe20 formulation) are comparable - in the 80-100K

range. Buyer remorse diminished .