| Search | Login/Register
   Home » Melquiades Amplifier » Full range Melquiades implementation (82 posts, 5 pages)
  Print Thread | 1st Post |  
Page 2 of 4 (82 items) Select Pages:  « 1 2 3 4 »
   Target    Threads for related reading   Most recent post in related threads   Forum  Replies   Views   Started 
  »  New  Melquiades For Dummies™ - step by step...  Amp still open...  Melquiades Amplifier  Forum     28  213470  08-29-2007
  »  New  Building Melquiades: Chronicle of full-range..  VR2 issue leading to jump in current on 6C33 tube...  Melquiades Amplifier  Forum     44  433020  06-09-2006
  »  New  Building Melquiades: questions and answers...  PeaceMAT XS™...  Melquiades Amplifier  Forum     96  564821  10-09-2007
05-19-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 26
Post ID: 18190
Reply to: 18189
Early days
fiogf49gjkf0d
Yes, it is built according to the schematic, but by someone else, from whom I have only just received it. The opt is custom-made nanocrystalline amorphous core, gapped at 300mA and running 200V/240mA.

It is quiet, precise and very fast (which last the unusual coherence of the SH50 helps). But what is most interesting, to my ear anyway, is the way in which the character and dynamics of tone appear to be signal dependent. Whereas an amplifier usually has a static tonal signature, imposing it on whatever is played in a uniform manner, this one intelligently adjusts to the input. A subtle accentuation of contrast that is helpful in a quiet passage is not there in a loud one, where it would not be. Though even harmonic distortion is there, as one would expect, it does not colour instruments, such as violins, where a richer pattern of distortions is essential to their tone. In general, the real microdissonances that give music vigour are reproduced, but without the cost of introducing fake ones elsewhere.  It is eerie.

What parameters would you suggest tuning?

bw, d
05-19-2012 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,658
Joined on 10-12-2006
Post #: 27
Post ID: 18191
Reply to: 18190
"Drive"?
fiogf49gjkf0d
Good for you, Decoud!  That picture ought to warm your cockles 'til summer!

I don't know if you have an audiohile-nervosa morbid streak (like I do), but are you curious to find the limits of this little amp with the big speaker?  Are you sneaking up on the limits, or have you already gone for it?  Are you running it FR, or high passing?  If x/o, then where and how?  If FR, then how is LF?  What about fortissimos?

As for tuning, how about clarity vs. tonal saturation, or impact vs ambience?

Best regards,
Paul S
05-19-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 28
Post ID: 18192
Reply to: 18191
Limits
fiogf49gjkf0d
No, I have not pushed it hard at all: I am just trying to understand it in the mean before I look at the extremes. Though 100db sensitive, the SH50 is designed to handle 1000W or some such absurd power, so it is surprising that it works so well. It is currently running full range, and the LF shows the interesting effects I mentioned, but I'd like to hear it when it is high pass filtered, as Danley recommends. Yes, I am sure you are right about those as outcome measures of tuning, but the question was rather what parameter of the amp itself to play with (beyond output stage current).
05-19-2012 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,658
Joined on 10-12-2006
Post #: 29
Post ID: 18194
Reply to: 18192
Parameters
fiogf49gjkf0d
Well, the usual pivots are the tubes' basic current and voltage operating points.  Like Romy has described many times, there is on the one hand an "optimal" convergence point at clipping.  Then, there are the audible "polar" "qualities" that one gets to "choose between" via the current/voltage/wattage choices, and plate loading, broadly speaking, not to mention the bridge, and the means of regulation, grounding, etc, etc.  Since (I presume...) you know what this speaker can do, it will be interesting to hear from you your impressions of how the amp fares and your sense of overall results with a large, "efficient" direct radiator of of the pro ilk.

Best regards,
Paul S
05-20-2012 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,160
Joined on 05-28-2004
Post #: 30
Post ID: 18196
Reply to: 18190
Do not forget to detonate to homeless Cats shelter.
fiogf49gjkf0d
 decoud wrote:
Yes, it is built according to the schematic, but by someone else, from whom I have only just received it. The opt is custom-made nanocrystalline amorphous core, gapped at 300mA and running 200V/240mA.

Very good. The 200V/240mA with 300mA gap is a sensible mode. With 300mA gup what the lowest cut off you get at full 18W-20W power?
 decoud wrote:
It is quiet, precise and very fast (which last the unusual coherence of the SH50 helps). But what is most interesting, to my ear anyway, is the way in which the character and dynamics of tone appear to be signal dependent. Whereas an amplifier usually has a static tonal signature, imposing it on whatever is played in a uniform manner, this one intelligently adjusts to the input. A subtle accentuation of contrast that is helpful in a quiet passage is not there in a loud one, where it would not be. Though even harmonic distortion is there, as one would expect, it does not colour instruments, such as violins, where a richer pattern of distortions is essential to their tone. In general, the real microdissonances that give music vigour are reproduced, but without the cost of introducing fake ones elsewhere.  It is eerie.

Yes, I would say that the effect mode applicable to dynamics (it was what I called viscose dynamics or smart dynamics) not to “tonal signature” as I did not feel the amp modify colors during play. It does happen with dynamics and key is the driver stage.
 decoud wrote:

What parameters would you suggest tuning?

I suggest do not modify anything. If you have any specific sonic problem or specific requirements to navigate sound in one or another direction that I might to pith to you some way to tune the amp slightly…

Rgs,
Romy the Cat

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
05-20-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 31
Post ID: 18197
Reply to: 18196
Education
fiogf49gjkf0d
Many thanks for the suggestions, both of you, and many thanks Romy for making the schematic available to others. This really is a remarkable piece of work and I would strongly encourage others to try it. 
05-21-2012 Post does not mapped to Knowledge Tree
RonyWeissman
Lyon, France
Posts 138
Joined on 05-29-2004
Post #: 32
Post ID: 18198
Reply to: 18190
Builder
fiogf49gjkf0d
Hi decoud,Could you share with me the contact details of your builder please ?
Thanks,
R Weissmanronyweissman@hotmail.com
06-14-2012 Post does not mapped to Knowledge Tree
RonyWeissman
Lyon, France
Posts 138
Joined on 05-29-2004
Post #: 33
Post ID: 18288
Reply to: 18197
Melquiades update please
fiogf49gjkf0d
Hi decoud,
Are you still enjoying your amplifier ? I may go for it, or I may try a leben 660 or almarro 318 . Any feedback would be helpful.R weissman
06-16-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 34
Post ID: 18296
Reply to: 18288
Risk
fiogf49gjkf0d
My view remains unchanged: it is an extraordinary amplifier, at least with my speakers.  It is also relatively simple, so if something goes wrong you will have no problem correcting it. I would change a few small things with the implementation -- just structurally -- but more appropriate to email me about it.
07-23-2012 Post does not mapped to Knowledge Tree
RonyWeissman
Lyon, France
Posts 138
Joined on 05-29-2004
Post #: 35
Post ID: 18439
Reply to: 18296
Melq building
fiogf49gjkf0d
Thanks for response I will e-mail after the holidays. i have begun exchanging mails with the builder and I think I can swing this project, financially speaking.
R Weissman
08-08-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 36
Post ID: 18498
Reply to: 18197
Coupling capacitor
fiogf49gjkf0d
Hi Romy, I imagine the coupling cap in the full range melq acts as a first order high pass filter: is there any reason why one could not lower its value to raise the cutoff to a level one would otherwise redundantly need a speaker-level filter to achieve?  rds, d
08-08-2012 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,160
Joined on 05-28-2004
Post #: 37
Post ID: 18502
Reply to: 18498
Sure, you can do it.
fiogf49gjkf0d
 decoud wrote:
Hi Romy, I imagine the coupling cap in the full range melq acts as a first order high pass filter: is there any reason why one could not lower its value to raise the cutoff to a level one would otherwise redundantly need a speaker-level filter to achieve?  rds, d

Yes, sure the coupling cap in the full range Milq acts as a first order high pass filter and this is what I use in DSET configuration. If you have a wide range amp and need to high-pass loudspeakers at let say 30Hz then it would be a good idea to implement the roll off by coupling cap. The OPT still need to be at least 1.5 octaves lover however. I do not stress this meted in full range amp as I do not like the loudspeakers with closed bottom, or the loudspeakers that need a high pass.

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
09-14-2012 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 617
Joined on 01-07-2006
Post #: 38
Post ID: 18595
Reply to: 18498
Nichicon caps source?
fiogf49gjkf0d
Decoud (and others), where do you source big Nichicon caps, pref. in EU?

I'm slowly preparing to attack a Staxquiades - a mastodont Milq inspired
amp to power Stax headphones (PP 6E5 DC to PP 2A3/300B/EL34).
I need 10.000uF/500V and 10.000/350V caps.
Mouser, srouser and other trouser sell them at prohibitive
prices and I believe one can have them way cheaper.
(I got a very reasonable-looking "quote" from Hong-Kong, but I susect the seller has little
idea which voltages she quotes)

Thanks for any help,
N-set


Cheers,
Jarek
STACORE
09-15-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 39
Post ID: 18596
Reply to: 18595
Eu sources
fiogf49gjkf0d
Looking for a reliable one myself: a capacitor just failed. Similarly for Electrocube 950s which are obscenely priced in low quantities direct, and not on the ground this part of the world.  
09-15-2012 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 617
Joined on 01-07-2006
Post #: 40
Post ID: 18601
Reply to: 18596
Nichicon failed?
fiogf49gjkf0d
 decoud wrote:
Looking for a reliable one myself: a capacitor just failed. Similarly for Electrocube 950s which are obscenely priced in low quantities direct, and not on the ground this part of the world.  


Really? Did you form it before putting on HV?


I got scared just looking at it:

http://www.ebay.de/itm/221040095771?_trksid=p5197.c0.m619

I consider getting a pair...would prefer a bit smaller but...heck

Electrocube-I passed on the idea and used Arcotronics KP172 in EAR834

http://uk.rs-online.com/web/c/?searchTerm=KP-172&sra=oss




Cheers,
Jarek
STACORE
09-16-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 41
Post ID: 18604
Reply to: 18601
Nichicon small
fiogf49gjkf0d
Rather than using a single can my builder made a bank of 470uf units, one of which failed. I thought I'd replace the entire bank, for simplicity. 
09-16-2012 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 617
Joined on 01-07-2006
Post #: 42
Post ID: 18605
Reply to: 18604
Trouser?
fiogf49gjkf0d
At those small values mouser, didgikey etc should be ok, no?
Any reasoning behind such a approach aprt fro aviability?
Also ROmy has sunstancial capacitance at 15000u etc, do you
use 30 caps in a bank?


Cheers,
Jarek
STACORE
09-16-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 43
Post ID: 18606
Reply to: 18605
Bank vs single
fiogf49gjkf0d
The decision to use banks was my builder's: whether it was availability or other reasons I have not asked. I want to change to single capacitors because I like limiting the failure degrees of freedom, so to speak.
09-16-2012 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,658
Joined on 10-12-2006
Post #: 44
Post ID: 18608
Reply to: 18606
Rise Times and Jobbers
fiogf49gjkf0d
One of the reasons cited for using several smaller caps is rise/cycle time, versus 1 large cap.  I understand it is also easier to wind the smaller caps to the desired high-zoot specs.  Muse caps are said to have very low rates of failure, compared to others, and they have famously low ESR, along with excellent rise times, and good service lives.

I have no personal reason to explore this further, but I have been given to understand that Marantz used some sort of modern, industry-only "super caps" with large values in the MA-9S2.  As it has been several years since this amp hit the market, these caps might be available to jobbers now.

Best regards,
Paul S
09-16-2012 Post does not mapped to Knowledge Tree
N-set
Gdansk, Poland
Posts 617
Joined on 01-07-2006
Post #: 45
Post ID: 18610
Reply to: 18608
Single vs. multiple
fiogf49gjkf0d
Yes, I've heard this theory too that paralelling caps lowers the esr/esl => better rise time.
No idea if it's true or only an excuse for not using big, hard to get and expensive single caps.

PSA quote from arandom physicsforums thread:

"In the general, the larger the capacitance, the smaller the ESR becomes. So you would have to hunt for a giant cap which would probably cost a fortune"

 http://www.physicsforums.com/showthread.php?t=422231


If that's true, this can be the "secret" behind Romy PSU philosophy.  Huge cap becomes more invisible?


Cheers,
Jarek
STACORE
12-17-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 46
Post ID: 18824
Reply to: 18610
6c33c socket cooling
fiogf49gjkf0d
it is worth reiterating in this thread -- and perhaps Romy should update his advice in the relevant section -- that exotic japanese teflon sockets such as the one made by yamamoto soundcraft are no good for the 6c33c, whether because of insufficient cooling or the nature of the contact i dont know but the heater contact is unstable.
Has anyone tried cooling the socket with heatpipes, i wonder?
12-17-2012 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,160
Joined on 05-28-2004
Post #: 47
Post ID: 18825
Reply to: 18824
Do not cool socket but cool wire
fiogf49gjkf0d
 decoud wrote:
it is worth reiterating in this thread -- and perhaps Romy should update his advice in the relevant section -- that exotic japanese teflon sockets such as the one made by yamamoto soundcraft are no good for the 6c33c, whether because of insufficient cooling or the nature of the contact i dont know but the heater contact is unstable. Has anyone tried cooling the socket with heatpipes, i wonder?
I do not use the Japanese Teflon sockets but somebody have reported before at this site that they are not good. Since I did not use them I do not why they are not good but I presume that they spent time and money for relevant thing – namely the material of sockets and they did use expensive and more or less exotic Teflon. The truth is that the material of the sockets is more or less irrelevant and the key is the contraction of the pins. Usually the pines are made self-springy by the shape of the pin. The Teflon sockets use density of Teflon to provide the spring action and that might be fine for low current pins but if pin meant to care high then it might be a bit problematic from cooking perspective as Teflon is very good thermo-isolator. I do think that this time of socket still might be used if you suck outs heat from socket. Use a low gauge wire as you approach socket, take the isolation off and wide some kind of heat exchanger from that wire. It will suck out hit from the pin of the tube but I have no idea if it not going to impact sound. In other words: do not cool socket but cool wire – filament and perhaps plate.

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
12-17-2012 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 48
Post ID: 18828
Reply to: 18825
Heatpipe wires
fiogf49gjkf0d
Yes, this is what I meant: link the pins using heatpipes to a heat exchanger, in fact I wonder if one could use the heatpipes as the filament conductors direct.
The yamamotos are beautifully made but badly designed. The receptacle of each pin is a split hollow cylinder. When the pin enters it the sides are splayed such that there is contact at only one point, not along the length of the pin as is the case with conventional septar sockets. So at high current the pins cook and you intermittently lose the connection. 
The choice of teflon for the substrate is strange. One really ought to have a material that dissociates electrical and thermal conductivity, so that the heat is transmitted from the pins to the chassis, e.g. aluminium nitride. 
02-15-2013 Post does not mapped to Knowledge Tree
decoud
United Kingdom
Posts 247
Joined on 03-01-2008
Post #: 49
Post ID: 19005
Reply to: 18828
Delayed impressions
fiogf49gjkf0d
Strangely, no one but its architect has written here about the sound of the full range Melquiades, and so I thought I would say something with the benefit of several months of listening to a version of it. I say a version because though the circuit is executed essentially as in the original it is hard to know what impact incidental features of the implementation have. But then it is hard to know what impact other components of the system might have, etc, etc, we are familiar with the usual caveats.
I should also say I take it as axiomatic that playback is not a transcription of a recording but a translation; it is never the same as the original performance and it could be good for reasons that have nothing to do with the accuracy of the translation. The features of playback are correctly interpreted along the lines of the features of a musical instrument replaying the original. This means summary measures such as bandwidth, distortion, etc are neither here not there, just as they would be neither here not there if one used them to compare a Guarneri to a Stradivarius. The standard summary measures are in any event only helpful in relation to linear stationary signals, and of course what defines music is that it is aggressively non-stationary and non-linear. Nobody talks about summary measures for non-stationary signals because no such measures exist. So the objective/subjective debate in audio is mostly bullshit: we do not have adequate objective measures and never will, so the only thing that will ever matter is the quality of the subject giving you the subjective and only view anyone could possibly give. If none of this rings true to you, ignore the rest of what I say.
Now the most striking thing about the Melquiades is that it seems to handle music *intelligently*: responsively to its content. I have no idea why it should do this, but that is how it sounds. The aura of resonances that gives a musical instrument its distinctive essence is conveyed as if the pattern of distortion *adjusts* to match the pattern of resonance of the instrument itself. Someone might of course say that it is simply replicating the pattern in the input -- mere accuracy of transcription -- and it may be so, but given that amplifiers with (I would guess) substantially less distortion do not do this I suspect it is not the answer. In any event, the result is a very special capacity for tonal differentiation. It is most impressive with complex music, inevitably, for that is where tonal homogeneity is most palpably felt as a defect. And I suppose it is because that kind of music is rarely the target for an amplifier designer that we see it so rarely in the designs of others. 
02-18-2013 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,160
Joined on 05-28-2004
Post #: 50
Post ID: 19019
Reply to: 19005
Melquiades attitude.
fiogf49gjkf0d
Thanks for comment, decoud. If you live with Melquiades for a while then you might discover that it will moderate your listening interests with time. I do not know what you are listening now but be if you spend substantial amount of time with Milq then you might find yourself to gravitate toward Wagner, Bruckner, Bach….

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
Page 2 of 4 (82 items) Select Pages:  « 1 2 3 4 »
   Target    Threads for related reading   Most recent post in related threads   Forum  Replies   Views   Started 
  »  New  Melquiades For Dummies™ - step by step...  Amp still open...  Melquiades Amplifier  Forum     28  213470  08-29-2007
  »  New  Building Melquiades: Chronicle of full-range..  VR2 issue leading to jump in current on 6C33 tube...  Melquiades Amplifier  Forum     44  433020  06-09-2006
  »  New  Building Melquiades: questions and answers...  PeaceMAT XS™...  Melquiades Amplifier  Forum     96  564821  10-09-2007
Home Page  |  Last 24Hours  | Search  |  SiteMap  | Questions or Problems | Copyright Note
The content of all messages within the Forums Copyright © by authors of the posts