NL.

Upgrade from Majik DSM to Akurate DSM - pros and cons

Recommended Posts

On 10/07/2020 at 01:08, chesebert said:

If ADSM is not designed for balanced, then the same would be true for KDSM.

No, the analogue output stage of a KDSM is totally different to the ADSM’s one. The KDSM’s output stage contains an audio transformer, which isn’t present in the ADSM. And Linn strongly suggest not to use the KDSM’s RCA and XLR outputs at the same time, since using RCA "will degrade the XLR output audio", but there is no such suggestion for the ADSM form Linn. As far as I know, the ADSM’s negative polarity terminal (aka "cold") on its XLR output socket is generated by just inverting the "hot" signal also fed into the RCA jack by on additional operational amplifier. In other words, the analogue output stage of the KDSM is truly balanced, whereas the output stage of the ADSM isn’t.

  • Upvote 1

Share this post


Link to post
Share on other sites
On 09/07/2020 at 22:43, NL. said:

If I understand you correctly, Akurate DSM has been in fact designed to work with the unbalaned (RCA) inputs and outputs. But it has "a component", or an imbedded adapter so to say, which is responsible for transforming unbalanced to balanced signal, right? So its inner circuits havent been designed truly balanced (like some American style equipment, e.g. McIntosh), in fact they are just "compatible to balanced".

Yes. The "components" on the ADSM’s output side and the Ax200’s input side are operational amplifiers. On the ADSM’s side the XLR „cold“ signal gets generated by an additional operational amplifier. On the Ax200’s side. the balanced signal will be converted to an unbalanced signal (again, via opamps), because Ax200s are not a truly balanced design from input to output.

Since I’ve bought my two A4200s initially as external amplification for an MDSM, they were RCA models. At the time I’ve upgraded to KEBs (XLR outputs only…), I’ve asked Linn if replacing my RCA 4200s by XLR 4200s would result in better SQ. Their reply was that "I should listen myself". If Linn would be sure that Ax200s with balanced inputs are sounding significantly better than the ones with unbalanced inputs, I’m quite sure that I would have gotten a different answer…

  • Like 2

Share this post


Link to post
Share on other sites

Linn related topics become more understandable to me....slowly, but steadily ... :D

many thanks guys!!! :)

Share this post


Link to post
Share on other sites
On 14/07/2020 at 16:33, BB1 said:

No, the analogue output stage of a KDSM is totally different to the ADSM’s one. The KDSM’s output stage contains an audio transformer, which isn’t present in the ADSM. And Linn strongly suggest not to use the KDSM’s RCA and XLR outputs at the same time, since using RCA "will degrade the XLR output audio", but there is no such suggestion for the ADSM form Linn. As far as I know, the ADSM’s negative polarity terminal (aka "cold") on its XLR output socket is generated by just inverting the "hot" signal also fed into the RCA jack by on additional operational amplifier. In other words, the analogue output stage of the KDSM is truly balanced, whereas the output stage of the ADSM isn’t.

I don’t think this is correct. Both use single ended output stage with the only difference being KDSM uses transformer to covert SE to balanced whereas ADSM uses op amps. 

Share this post


Link to post
Share on other sites
Posted (edited)

I don’t think so. The output of the AK 4497 (i.e. the Katalyst DAC) is differential. In its data sheet you’ll see two examples showing how to create a single-ended and a differential output from the DAC’s output singals.

Example 1 (single-ended output):Example1.thumb.jpg.28de145c7bf47dcea7bdbbb132cbbfaa.jpg

Example 2 (differential output):

Example2.thumb.jpg.35879091c911fd540fa3437aea432234.jpg

At Akkurate level Linn’s output stage is based on example 1 and the "cold" signal on its XLR output gets generated by an opamp inverting "Analog Out". Therefore two conversions take place. From balanced to unbalanced (OPA 1611 in example 1) and then from unbalanced ("Analog Out") to balanced by adding another opamp inverting the unbalanced signal.

At Klimax level it would make no sense at all to drive the output transformer by a single-ended signal and "misusing" it for converting the single-ended signal to a balanced one. Hence I’d bet that the output stage at Klimax level is similar to example 2.

Edited by BB1

Share this post


Link to post
Share on other sites
Posted (edited)
On 17/07/2020 at 18:04, BB1 said:

I don’t think so. The output of the AK 4497 (i.e. the Katalyst DAC) is differential. In its data sheet you’ll see two examples showing how to create a single-ended and a differential output from the DAC’s output singals.

Example 1 (single-ended output):Example1.thumb.jpg.28de145c7bf47dcea7bdbbb132cbbfaa.jpg

Example 2 (differential output):

Example2.thumb.jpg.35879091c911fd540fa3437aea432234.jpg

At Akkurate level Linn’s output stage is based on example 1 and the "cold" signal on its XLR output gets generated by an opamp inverting "Analog Out". Therefore two conversions take place. From balanced to unbalanced (OPA 1611 in example 1) and then from unbalanced ("Analog Out") to balanced by adding another opamp inverting the unbalanced signal.

At Klimax level it would make no sense at all to drive the output transformer by a single-ended signal and "misusing" it for converting the single-ended signal to a balanced one. Hence I’d bet that the output stage at Klimax level is similar to example 2.

I could be wrong but I understand transformer gives three big benefits:

1) limits bandwidth in both directions.

2) gives galvanic isolation, making it easy to avoid earth loops, and

3) provides single-ended to balanced transformation.

This is why I think you are mistaken in your understanding of Klimax output. I am continually puzzled by why a high end SS DAC would use transformer coupled output. There are some huge disadvantages, namely LF distortion, limited bandwidth, HF resonances and others. 

There is no arguing Klimax sounds great all the way back to mark1, but the output implementation has always puzzled me.  

Edited by chesebert

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.