or Connect
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › Chameleon DAC listening and modifications
New Posts  All Forums:Forum Nav:

Chameleon DAC listening and modifications - Page 44

post #646 of 1144
Originally Posted by Bill Allen View Post
Yea I am with qusp on this one (with a Q) I would stick with the films. No electrolytic's here. Nugent actually suggests higher quality surface mount caps. I will see if i can needle a brand name out of him. Installing those blue Vishay's looks like a real pain unless you do them before the Blackgate's go in... but I sure do like how they look!
well i'm with steve, and I can make some recommendations, the evox/rifa (now kemet) PPS and PET SMD films and wima PPS SMD films are fantastic! also the sanyo POSCAP range are SMD tantalum/solid organic polymer and they even have palladium/gold 'leads' but be careful, these are polar, there is no non-polar version. AVX makes some great SP SMD type caps and COG type high quality ceramics.and yes the vishays and wima etc short lead caps are a PITA tp work with because of the tiny leads

I pretty much like everything that Chis VanHaus sells including the V-Caps and I am positive the new copper teflon's are another leap forward in signal transfer. However when passing a full range signal round is round ... is sounds round. This is actually a pretty big discovery and I encourage you to re-evaluate you position by further research, don't box yourself in. Now its always possible this effect is subtle in your rig for whatever reason.
well sure, like I said, i'm already using flat caps in some places and I love them, but i'm unsure its because they are flat, I doubt there is any problem with resolution in my rig
135DNR for source and preamp/headamp @
frequency response of 0hz->25mhz
Equivalent input noise: 0,6 µV
Signal to noise ratio: 112 dB at 1 V out (this is the spec for a single stereo amp, but i'm running 2 in dual mono bridged, so the spec will be significantly better)
Dynamic headroom: 129 dB (ditto)
Distortion: 0,00046%
Slew rate: 1200 V).

my headphones are 119db at 1mv
my heavily modified biamped active monitors are around 98db, but they could use some improvement, thus my next active monitor project

^^ just for reference, not getting into a pissing match, I have no doubt your speakers are more revealing than mine (particularly of soundstage character), thus my drive to make my own; but for detail retrieval the JH13 slaughter any speaker I have ever heard, tho some particularly stunning ATC actives (25K worth) sure give them a run for their money

but the chances of me doing to much more capacitor research are fairly slim, one dac is DC coupled and the other is transformer coupled. knowing me tho, now that you have mentioned it, I will find myself unable to resist the curiosity.

In my system it is clearly evident, this is not a small difference. With 100db open baffles this is blown up like a microscope, the soundstage takes on a completely different presentation with the flat caps, much like proper output transformers. Speaking with Chris Young there are no differences in construction material between the orange Jupiter's and the new vintage one except they are flat wound. I encourage any of you with Jupiter's to take the Pepsi challenge and try these new flat ones.
I dont doubt your result, just hasnt been mine, but I cant say I have analyzed it quite that much and yes perhaps your system is more revealing of soundstage, because my rig is primarily headphone based (exceedingly revealing headphones JH13) and HD600, perhaps i'll report back when i'm done with the actives, but yeah I dont think i'll be using output caps (probably no input caps either as I have full control of offset)

Now if we could encourage Mr. VanHaus to build some flat copper teflon's the superior dielectric would probably set yet another benchmark. I just wish every hobbyist, manufacture (and engineer) was as open to new ideas as TeraDak is.
interesting idea, but I dont think there would be much to drive him with the overwhelmingly positive response he receives. as it is the CuTF required a whole new manufacturing process

Ok, the cast Duelands sure do look round to me but the internal construction may still be flat wound or maybe something even more exotic. However we are now approaching silver transformer land pricing. At what point do we say stop and cry ... Uncle! Lunar landscape soundstages aside, it sure is exciting that the lowly capacitor is now maturing into transformer like sound in coupling applications with the advent of the "new" Duelands, Jupiters & V-Caps. The advantage being even the layman hobbyist can easily experiment and decide for himself.
indeed the internal construction is flat, but its also round in shape and encased in epoxy(CAST) as you mention tho, there are certainly some significant things you can do with the money required for CAST. certainly sounds like we are pretty much on the same page though.
post #647 of 1144
Originally Posted by Wood View Post
Ok, quys, you talked me out of trying the bg for small values (so much wanted to play with those little cute nx's) but heh put the money towards some flat caps instead.

Yes its a bit of a pain putting those decoupling caps in, had to take out 2 rows of BG's first, and put them back, lifted a trace in the process, can't get a piece paper inbetween them now. If Steve does give you a hot tip on smd caps please let us know Bill, as I don't have room on the digital board for anymore 0.1 vishay. Also it would be damn useful to know of a good smd cap!
hehe, well played; not too subtle for me ;D didnt mean to harp on it mate, but I see it all over trhe place people using BG where they simply arent any good. those poscaps are the hottest thing going at the moment as far as chp caps, there are a few other chip films too that would do nicely. unfortunately if you dont have room for vishay, you wont have room for the SMD PPS evox/rifa, which IMO are simply the best thing going for this purpose, but the poscaps are a close second to films and they come in values up to 150uf as well. make sure to get the 10v, 16v or 20v versions. there is a reason that even tho they are fairly new they are almost constantly out of stock direct from sanyo
post #648 of 1144
Originally Posted by Wood View Post

I see you did the 74hc04 MOD. Or did that come stock?

Those of us who did added a couple of resistors to avoid reflections between the USB, Spidf receiver and the XO.

There should be two 47R resistors between the XO and the S/PDIF and USB receiver.

The possible reflections will introduce jitter.
Easiest thing to do is to remove the connections between pins 3&4 and 10&11 and put the two resistors there.

Here is Rhodes54 diagram, Picasa Web Albums - therealrhodes54

I did the resistor update though I did not myself hear any problems without it. Oh and you can remove and bypass those 3 resistors on the top right as you can see from my pic.

Though if this is Teradaks new stock version then perhaps nothing needs doing.
As promised here's an updated picture of the 74HC04 Mod.

The complete Mod description can be found here
post #649 of 1144


Well spotted, qlad you’re paying attention

Please continue to play your harp Qusp, no point in being mealy mouthed, your directness is much appreciated by all and sundry.

I've learnt to be cautious over recommendations, no matter how passionate, the biggest disappointment are Sanyo Oscon SP, I have used them in different places other than this dac and get the same result, a hi fi rounded pleasing but b*llsh*t, sound (fake warm valve stereotype) though you see enthusiasts swear by them all over the forums. Wimas are also a disappointment but are far less universally praised, and more controversial.

The Sanyo Oscon SEPC on the other hand… lots of kick, extension and importantly transparency all the way down. They play well with the Vishay too, similar character for higher frequency. So the difference, between SP/Wima and SEPC/Vishay, night and day, yet I would not of been so easily aware of this difference surfing forums. I had to listen for myself.

I wouldn’t have believed a few months ago if you told me (actually I think you did) that decoupling caps on a digital board had the same effect on SQ as coupling caps. But boy they do, so now I have to keep an ear on the SQ of these caps not just an eye the datasheets, aka sales pitch (which I confess I don’t understand much anyway).

Hence the idea of using BG NX, the larger BG caps are transparent but have rich dynamic tone (yes I know we use larger electo's because there is no film equivelant), so perhaps the smaller bypass NX will maintain that. The Blackgate cultist’s mantra is they do, and are to die for, the other camps mantra is ‘any film cap is better than any electrolytic’. One thing for sure Bg Nx are better IMO, then Wima’s, as low value decoupling. So my mantra is ‘suck it and see’ but with the caveat ‘don’t put just anything in your mouth’.

The Vishay make a big impact but are they letting it all through? Plastic film, definitely, has a character that’s different from organic caps, and BG; fast clean but ultimately a tad behind in tonal richness. Less; being there!

So Mundorf silver, gold in oil, are more real than the straight Mundorf silver, gold (they say) and there’s the new Vcaps you like. That’s why I have a couple of West-Cap capacitors on order to try as input caps for my Sure TK2050 amp project.

Nice thing is the industry is growing up, we are moving away from the false dichotomy of hoody/black tee-shirt wearing solid state, clean powerful, good for rock/pop/rap, but not real, and cardigan wearing tubey oil and paper cuddly, but smeared sound for classical lovers. The Chameleon is a good example of doing it right, though when I listen to it I feel strangely compelled to wear a black tee-shirt/cardigan/hoody ensemble, just need to nurture a goatee now and take up pipe smoking.

Anyway that’s my pathetic attempt at writing a longer post then Qusp.

Of topic, but Qusp do you have any recommendations for good chokes/inductors for output filtering on type D amps.
post #650 of 1144
There is a strong following behind the BG Hi-Q NX (.1uf 50V) being used as bypass cap Wood....stick to your plan and make your own assessment. A buddy of mine loves using these in his tube phono stage with his results having been very very positive. Is there better stuff out there ? Of course there always is but I think the NX is a unique approach that doesn't cost all that much to try out but is well worth perusing.

QUSP & Bill A....great stuff gents

post #651 of 1144
Actually I'm doing a little experiment, Ive got a nx decoupling the pulse on the digital board, and in a month, (they take 1000hrs), I'm going to pop in a vishay to compare.
post #652 of 1144
My teflon bypass caps have about 340 hrs on them. Somewhere around the 200 hour mark though, they seem to have "fallen apart" they just don't sound right. The soundstage is off, the bass is flabby, there's to much mid-range.. I had to take them out, they just don't jive with the Dynamicaps. I took them out and am listening now, it's much better. Im sure they are good in some setups I think it's just a synergy issue with the Dynamicaps. Im still quite pleased with these dynamicaps actually, they are very well extended in all directions, they could be more transparent as they do have a warm flavor that can't be denied. Im going to try bypassing the WIMA in the FRM mod with one of these 0.022 teflon's and see what happens. Those flat jupiters are tempting, but a bit too rich for my blood. My next dac project will be a Sabre32 design, but that is a ways off. I don't know if I want to invest to much more money in the Chameleon.

I never did heatsink the DC-30 and it has ran for 2 weeks straight, I don't notice any issues, it does get hot, but levels off and maintains it's temp.
post #653 of 1144
wood, yeah I see no reason to sugar coat things on a forum, this isnt a popularity contest. totally with you on SP vs SEPC (hehe and they come in SMD) datasheets can only get you part way there, absolutely listening tests must be the final word. your comment of BG being less 'being there' is likely closely related to the fact that soundstage information and spacial cues is mostly in the HF, barely audible and even inaudible and BG simply arent fast enough to eek out the last bit of this info (this is a fact, not an opinion, look at their own datasheet, electrolytic caps impedance increases with frequency, so by the time we are dealing with the switching speeds of diodes, FETs and even comparatively lower frequencies of DACs, they struggle badly). I actually quite like BG, especially N and FK bypassed with a small film and even smaller BGs can do OK as bypasses in the analogue signal path, but standard can be beaten by any number of caps IMO and thats just about all many have left in stock. However I absolutely do NOT feel they should be used specifically for parsing or shunting high frequency information, most notably places like you mention above wood, I cannot think of a place I would use them after that TBH. great mids and upper bass though, even for an electro. they just arent built for speed, electros just arent. films can be crap too tho. the latest mundorf Mlytic with titanium oxide will replace BG FK as main supply caps in my pre/headamp too, fantastic stuff and totally available. sanyo poscap are like SEPC with faster transient response, being that they use a mix of tantalum and pretty much the same dry solid polymer as SEPC and yes they have shades of electrolyte, but construction and materials mean much better performance; they should still be bypassed with a COG ceramic, film or other very fast cap in some positions to get the best of both worlds.

which wimas are you talking about BTW?, there are many different sorts in basically the same package FKP4, and PPS are probably the best for decoupling and power supply bypassing, but PET is more for strict power supply duties and available in higher ratings for voltage.

gonna cut it 'short' there, I could blab on, but i've made my point and who am I to argue with subjective results, they are what they are.

Mike, teflons have THE LONGEST burn in and are the moodiest of any caps around, and yes sometimes they sound downright BAD!!, you have to be patient with them, but they do pay off in my experience. some take longer than others. BG taking 1000hrs is not my experience at all (sus IMO) some say forever. I find they reach most of their potential in 50hrs, then take a backward step and are pretty much there after 200, with a very slight and slow improvement over the next 2-300. but folklore reigns supreme with BG like no other. the fact that BG provoke such strong opinion does speak of their character though, great caps yes, defy the laws of physics...nnnnoooo not for me they dont
post #654 of 1144
QUSP is absolutely correct about Teflons...they will come around after the "bad period". I have found out the T1/T3 series take around 700 hours to mature (+ or - 50 hours) according to my experience with them in the last 2+ years.

post #655 of 1144
Indeed as said in my earlier post http://www.head-fi.org/forums/f7/cha...ml#post6463741

But Qusp sure is luckier then me with Blackgates, I found teflons a stroll in the park by comparsion.
post #656 of 1144

Praise indeed

Originally Posted by PatOMalley View Post
Just to get a point in that my Chameleon DAC has exceeded all expectations at this point. It closely rivals my analog 'rig'. Animated, flowing, human, giving up the "ooh yeah."

No apologies, I love the way the thing sounds now. There may be higher steps on this ladder but I am feeling levels of satisfaction..now.
Totaly with you on that Pat, same here. To say it rivals your analog rig is a big deal not something I would expect you to say lightly. That why its so intriguing, this recent talk about Duelands, and the rest, how far can this go?
post #657 of 1144
I think that's Pat's way of weasling out of pursuing the output transformer solution.

I spoke with Steve Nugent about the SMD decoupling caps on the DAC chips and he named some of the same film caps Qusp mentions. Once again he suggests a .1uf value in place of the .01uf. He also said you have to be extremely careful when soldering SMD film caps.

My experience with Black Gates is 30 days of continuous operation before I am confident they are done doing their thing. It might happen sooner but I am never confident. Teflon's caps get the same treatment.
post #658 of 1144
Thread Starter 

you weasel, you

Not so, Bill.
I am pursuing it. I would rather remove those caps. I have a sidebar going with someone on this thread about it ... more soon.

[i know when i am being egged on]

Originally Posted by Bill Allen View Post
I think that's Pat's way of weasling out of pursuing the output transformer solution.

I spoke with Steve Nugent about the SMD decoupling caps on the DAC chips and he named some of the same film caps Qusp mentions. Once again he suggests a .1uf value in place of the .01uf. He also said you have to be extremely careful when soldering SMD film caps.

My experience with Black Gates is 30 days of continuous operation before I am confident they are done doing their thing. It might happen sooner but I am never confident. Teflon's caps get the same treatment.
post #659 of 1144
The original smd are indeed .1uf, I used .01uf cos no more space. Though .01uf seem fine, it might be possible to wiggle .1uf in there, but at the risk of longer legs. So if you want .1uf then smd is the way to go (of course bg nx .1 would fit, okay I know I know).
post #660 of 1144
Yeah I know how long the teflon's are supposed to take to burn in, but they have had 340 hrs, if they are sounding this bad now, I don't see them improving enough in the next 300 hrs. Maybe im wrong? Can I go from really disliking them at 340 hrs to loving them at 700 hrs?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › Chameleon DAC listening and modifications