[serengetiplains]

Thanks for the tip, Rodge. 

Guys, what's the need to ridicule?  If those supplies lack the value you claim they lack, who would then care to even talk about them?  Next ...

My tentative priority for Ncore power supplies is starting to shape up like this: comparator supply is most important, followed closely by the driver supply, followed by the input buffer supply.  My sense is that the output supply is less important for overall sonics than any of those three, although important in its own right. 

Regarding the output rails, electrolytics, to me, sound like mud in the audio band, so I'll experiment with bypassing electrolytics on the output rails, claims of parasitic resonance notwithstanding (a few big and a few tiny teflons on my Tacts, carefully placed, and wow the mud reduction).  Also, the lower the noise the better, clearly.  But I question if a regulated supply could be feasibly designed to supply current quickly enough (a maximum 27 amps of which) and be able to sink energy.  If one doesn't regulate, the best solution is differentially cancel noise via rails carrying mirror noise voltages.  Bruno says his smps's synchronous secondary rectification gives this for that use within the Ncore's differential circuitry.  I'd have to see a schematic to see how he puts this to effect, but it makes implicit, well-thought sense.  Perhaps a linear supply will better the Hypex switcher.  There are a few reasons to think this might not obviously be the case. 

Bridging should also noticeably reduce output-supply-related noise.

I don't know if the comparator can be independently supplied without mucking something up.  If it can be, I'll be tempted to put a Hynes shunt on it, fed by a Hynes series.  The driver supply can be regulated easily (and which I think would function as preregulation).  Same for the buffer supplies.

These amps might really, really fly. 

[coverto]

I am more tickled by the $6000 LIQUID music server that weighs 150 pounds. Maybe he will include a couple of freshwater Piranhas in that fish tank......but alas they will not survive in mineral oil. Good to clean earwax though.

 

Anand.

Yeah, pretty far-out, I gotta say... that's one amp you don't want to knock off the shelf! 

[James Romeyn]

I wish Bruno had consulted with more forum members before designing and selling me his crummy SMPS600! 

[serengetiplains]

Btw, my Tact is a Lyngdorf Millenium.  I replaced the digital main supply and all regulators (5) each with tx-->Hynes series-->Hynes shunt.  All the electrolytics are well bypassed, output filter caps likewise.  Higher frequencies are slightly elevated (probably the caps), but in all other respects, this amp sings.  Crystal-clear see-through and class D tone glory --- the better of what class D does well, better (notwithstanding lacking the deep coherence and timbre of class A push-pull SE, err). 

The Ncores will be an interesting competition.  Digital vs. analogue class D.

[serengetiplains]

And if the comparator is differential (can comparators be differential?), then one supply for the both.  Differential supply distortion.

[Emile]

serengetiplains, any views on the minimum amount of capacitance needed for the buffer section? Im contemplating skipping electrolytics alltogether and bundle cheap russian caps.

edit:

Positive DC voltage on J1:3 output - 21 - Vdc See note 1,2
Negative DC voltage on J1:7 output - -21 - Vdc See note 1,2

Looks like its a balanced supply with a ~42 volts potential. Didnt notice this before (didnt look very closely). So you'd need 2 lineair supplies and optionally 2 Hynes regs, this wont be an ultra cheap upgrade 

[OzarkTom]

For only $2700, you can add the Stillpoint footers and Stillpoint standoff mod.

Core Audio, now where have I seen that name before?

[poseidonsvoice]

And the plot thickens:

http://www.diyaudio.com/forums/vendors-bazaar/190434-hypex-ncore-461.html#post3074569

http://www.whatsbestforum.com/showthread.php?3618-Class-D&s=7faebcfdf0fa2e6c4a2082f8933b5324&p=118349&viewfull=1#post118349

I find it interesting that Gary's words were copied verbatim onto the Audiogon ad:

http://www.whatsbestforum.com/showthread.php?3618-Class-D&p=118354&viewfull=1#post118354



Anand.

[serengetiplains]

serengetiplains, any views on the minimum amount of capacitance needed for the buffer section? Im contemplating skipping electrolytics alltogether and bundle cheap russian caps.

edit:

Positive DC voltage on J1:3 output - 21 - Vdc See note 1,2
Negative DC voltage on J1:7 output - -21 - Vdc See note 1,2

Looks like its a balanced supply with a ~42 volts potential. Didnt notice this before (didnt look very closely). So you'd need 2 lineair supplies and optionally 2 Hynes regs, this wont be an ultra cheap upgrade 

It is a differential supply.  Bruno has placed what he calls the equivalent of an HxR regulator on-board the Ncore supplying the buffers.  The Ncore data sheet specifies that the minimum voltage input to those supplies is 16V.  I suspect the HxR-equivalent regulates to 12V, so with a, say, 3V dropout you need 15V, hence the 16V minimum (the extra V allowing for line-level fluctuations).  The DCV of the smps buffer supplies is nominally 21V, so a Hynes reg of, say, 17VDC rating with DC input from the smps will work just fine in this context.  That upgrade will not be too expensive.

[James Romeyn]

My trouble is the qualifier "improvement".  If I add decals to my amp, is it "improved?"  Maybe yes to certain individuals. 

Personally, I'd rather past, present and future similar threads be banned containing untested qualifiers like "improved".  I'd accept in their place the following:

Independent blind ABX listening test with at least a dozen independent audiophiles with no financial interest whatsoever with 100% consistently preferring the "modded" amp.  IMO the $6k cost of the linear PS justifies my 100% requirement.  DIY and lower cost mods could qualify with lower approval percentage.  Personally, it will be a cold day in hell before I modify my amp so that Hypex refuse to service it, which describes many if not most of the described "improvements."  That is not an improvement.  That is stupidly converting a perfectly fine amp in to an unserviceable toy, something I've owned and have no interest in.  Just my opinion.   

Or alternately:

Send the modded amp out to OEM-stock properly-built NC400 owners and post their reaction, something like maybe -5 to +5 rating scale:
-5 much worse than OEM
0   no difference
+5 much better than OEM

If the mod is indeed thus determined "improved" above, then indeed the word "improved" makes sense.  Otherwise it's pure conjecture and absolutely nothing else. 

IMO my position is easier to justify than labeling Ncore "improved" with absolutely no justification except, "This always improved prior amps!"  And that's all I've read to now, and I've read (I think) every "improved" post at AC.  That's my story and I'm sticking to it!  It's perfectly fine to disagree.  I don't require any justification to disagree.  Toss this post in the garbage if you like.  It's just my preferred position. 

I bought the Ncore sight unseen and sound unheard based on independent user reports, and because my need for better amps was urgent.  I based none of my purchase decision on financially biased opinions.  Conversely, in fact, I based my decision on the opinions of some persons who might be stuck selling their current amps, and whose opinions might negatively impact resale value!  That carries absolutely immense weight IMO!   

Lastly, the $6k linear PS sellers mention the quality of the NC1200 and AFAIK no one's heard it except OEM buyers and some select reviewers.  Pretty far fetched of a guy taking images on his carpet posting comments about the NC1200, sorry.  Oh, and he implies an OEM Hypex PS exists for the NC1200 and this is a false claim.  There are some untested methods to allegedly obtain US patent a small fraction of the normal cost of about $15k USD ($10k to file the patent, about $5k to contest legal claim...virtually every patent claim is contested by individuals who's purpose is to contest every patent claim...it is a war of attrition and they don't stop until they've tested the claimant's financial will and ability to the tune of about $5k).  Again, somehow I presume someone taking images on his home carpet has absolutely no funds to justify his comments concerning patent protection.  Patent requires complete and absolute scientific disclosure of the circuit, which many of the readers of this thread would immediately copy for DIY purposes, plus the design would be instantly common knowledge.       

[serengetiplains]

James, you're sounding very defensive, preemptive defensive, in fact.  I don't care that you don't care, and if in fact you don't care, play somewhere else.  Seems to me you care.  Which is it?  I don't need your approval to run my audio preferences.

[James Romeyn]

I reply to the so far unjustified (to me) claim of "improved."  That's it.  A better term is "modified."  I think, even by your own standards, it has to "compared" to be labeled "improved" does it not?

Amplifier A does not equal amplifier B.  In this case, A = stock OEM properly built/properly functioning NC400/SMPS600, B = A + "modifications".  A and B are never directly "compared" in any way. 

Please list and describe now the conditions in which B can be properly described as "improved" vs. A. 

IOW, how can B be "improved" vs. A when the two are never directly compared?

       

[serengetiplains]

Let me give you an English lesson.  The word "improvement" in the thread title implies "potential improvement."  Nothing more.  Actual improvements, if any at all, will be reported---at the earliest *after* I receive an actual Ncore module.

But back to my question, which I'll restate.  If you think the Ncores cannot be improved, why are you wasting your (and our) time here?

[jtwrace]

Gentlemen,

PLEASE stop the back and fourth and be kind.   

http://www.audiocircle.com/index.php?topic=37305.0

[James Romeyn]

For the record, I'd be happy to hear a modified Ncore to see if it was improved.  Sorry for any apparent bad feelings about this, which was and is not my intent.  I also have interest in reading about the mods too btw.   

[jtwrace]

For the record, I'd be happy to hear a modified Ncore to see if it was improved. 

I have interest in objective data on an "improved" NCore. 

[*Scotty*]

serengetiplains, Do you know if Paul Hynes makes any surface mount replacement regulators. I don't see any through hole devices on the Ncore modules board.
Scotty

[serengetiplains]

Scotty, I believe Paul does not.  He does make smallish series and shunt regulators measuring 1" x 1.5" thereabouts, but these are through-hole.  With a careful soldering hand, and a little luck with circuit traces, these can be placed right near the chip they're powering.  Their small form-factor is beneficial as circuit traces are short, benefiting performance.  Pricing is in the range of £30 to £36 thereabouts, or was when I last purchased some.

[*Scotty*]

If that is the case then I foresee some potential problems getting them in place due to the parts density on the module. I do know that regulation at radio frequencies which is what we are essentially dealing with here is a point of use proposition. Lead inductance or circuit board trace inductance will kill you at these frequencies.
Scotty

[serengetiplains]

I do know that regulation at radio frequencies which is what we are essentially dealing with here is a point of use proposition. Lead inductance or circuit board trace inductance will kill you at these frequencies.

I understand that, Scotty.  I do not yet know how the comparator is powered, but I speculate that placing a Hynes reg near the chip(s), if that is feasible given the circuit board layout, will put the regulated supply nearer the chip than the stock supply.