Ecaroh,
the BG's I am using now (in the super-e config) are older than the single ones I used before. These have had a burn in (with uncutted leads) for more than two months in a different system. And: BG's in super-e do solve some problems a single BG has. There is a better high frequency resolution, but at the same time the highs sound more sweet and articulated. The soundstage is wider, the bass sounds more natural and real.
Really, super-e with BG's to me is a quite magical thing...
The disctrete opamps - I would love to try them, but I saw the pics and I don't think they would fit in here, they're simply too large in size!
Now, the stangest thing of all:
Yesterday evening I just put in the 2604 again, just to confirm my prior statement, but you guys, really, this time it worked!!! No higher noise floor, and the sound is awesome!
I do absolutely not know what happenend here before!
I haven't done much listening until now, but I honestly think the 2604 sounds better than the 275.
More "drive", better details, more dynamic and cleaner sound, but yet easy to listen to. WOW!
More to come...
the BG's I am using now (in the super-e config) are older than the single ones I used before. These have had a burn in (with uncutted leads) for more than two months in a different system. And: BG's in super-e do solve some problems a single BG has. There is a better high frequency resolution, but at the same time the highs sound more sweet and articulated. The soundstage is wider, the bass sounds more natural and real.
Really, super-e with BG's to me is a quite magical thing...
The disctrete opamps - I would love to try them, but I saw the pics and I don't think they would fit in here, they're simply too large in size!
Now, the stangest thing of all:
Yesterday evening I just put in the 2604 again, just to confirm my prior statement, but you guys, really, this time it worked!!! No higher noise floor, and the sound is awesome!
I do absolutely not know what happenend here before!
I haven't done much listening until now, but I honestly think the 2604 sounds better than the 275.
More "drive", better details, more dynamic and cleaner sound, but yet easy to listen to. WOW!
More to come...
Last edited:
In any case, there's at least one issue with the USB input that inevitably penalizes it: most probably you're gonna use one of those crappy generic USB cables, believing that a "fancy" cable would make no difference. 🙂
What is this "super-e" ?? Simply two caps in parallel? I hope not so 😛
What discrete opamps? The Audio-GD ones are all pretty crappy, and the Burson seems to be more similar to them than different. Fortunately there's the OPA1611. 😉
Yes, you're right ;-)
But at least I use a usb cable without ferrits. I tested this before, and with ferrites the noise floor became more quiet, but the sound became more dull with less dynamics and resolution. So, good quality, short length and no ferrites, that's the way to go.
Btw., I really do NOT intend to buy a usb cable which costs as much as three times the dac itself...
Well, super-e is an idea of Rubycon. In short (and as far as I know), it goes like this:
The Black Gate N series are bipolar caps. But even though they are bipolar, they still have one shorter and one longer lead. That is because inside the can there is an inner foil and an outer foil, to which the leads are connected to. So, even though bipolar, they have some kind of direction.
Jelmax (or Rubycon, can't remember) now states, that there is always a marginal rest of inductance to a cap, and by connecting TWO of these caps in parallel, but IN THE OPPOSITE DIRECTION (short lead with long lead, and long lead with short lead) one might get rid of this inductance completely. That's what "super-e" is about.
To say it right away: Most people might think that this idea is simply to sell much more caps. And I don't blame anybody who has this oppinion.
But I tried it myself intensively. At the signal input of my power amp I first used one single BG N per channel with the long lead for signal input and let it run in for some weeks to study the main character of the sound. Then I marked them and desoldered them. Then I took another, new pair and connected them with the shorter lead for signal input. Again, some weeks of burn in, listening, studying.
I found that BG's to sound quite different in these two configurations. Very nice sound, but some (different) drawbacks.
Then I paralleled these two marked pairs in the super-e way, and that was a relief! Suddenly everything seemed to fit right into it's place, everything felt "right" in some way. It seemed like the best combination of both worlds (long lead vs. short lead for signal input), and I thought by myself: "Wow, this is almost sounding like no cap at all, but just slightly better!"
Ok, I know, I'm walking on very thin ice with these statements. I don't want to start a discussion about BG's and super-e. I do only say all this because you asked about the term "super-e" ;-)
Last edited:
Well, I trust your findings... I assumed that the Black Gate N would be true non-polar caps, so I excluded the series connection of them to form a true bi-polar cap. 🙂
Still I doubt that 2 electrolytics put in series can ever sound as good as a good MKP film cap... but they're surely more compact. BTW, are the Black Gate N rated at 85°C ?
Anyway...I really think you would want to try something better than the OP275, even though this chip has the tonality to match the CS4398. Going LME49725 would be a significant step up, keeping tonality about the same. 🙂 And you can have LME49725 samples from National directly almost for free...
As far as coaxial cables are concerned, a great choice for the Super Pro (having used a bunch of different cables) is this:
Supra TRICO-RCA - From £39.50 - HiFiCables.co.uk
(I bought it right there)
Hi Andrea --
I took a look at the information on that cable when you mentioned it before. I'm sure it's great. Interestingly, the specs for the actual cable used are almost identical (just a tiny hair less good) to those for Belden 1505A, as used in the "Max Rochlin Memorial Digital Cable" (diy), which you've probably already run into, but if not --
Enjoy The Music's Max Rochlin Memorial Digital & Video Cable.
I like that I can make it as short as I want. I use two of them, one for my main DAC and the other, with a BNC at one end, for an A-D converter. For RCAs, I used the listed Canares built for that wire.
Haven't made any deep comparisons. When I seemed to get an improvement over the slightly generic, though nice, digital cable that I used previously, I was happy.
Hi,
Specificatiopns might also be very similar, but the Supra cable is (obviously) a different cable, made by Supra and not by Belden, and unlike the Belden cables it has a silver plating over the core and inner shielding conductors. And the core is stranded, and not solid.
That said, I've used the Belden 1694A, as short as .5m, as well as a Gotham silver plated coaxial, and a Beresford TRC-222 (made with Neotech 75ohm cable and plugs).
They were all good, but they sounded different from each other, and each was colored in its own way. The Supra just happens to be the least colored, plus possibly the one with the bess bass and soundstage. The second one i like the most overall is the TRC-222. The Belden/Canare is very honest, and if I hadn't heard anything else I could live with that happily. Don't know how the 1505A compares with the 1694A.
The BG's are not connected in series, but parallel! Just in the opposite direction (long lead with short lead and short lead with long lead). Another benefit is the doubled capacitance of now 9,4uF to avoid any low freuqency cut of. Remember, the original epcos smd caps were 47uF! But from a mathematical point of view I'm sure that anything above 4,7uF will do the job, even when a passive pre is used.
Regarding the opamps, as said above, I'm now listening to a well working OP2604, and I think I'm really happy with it!
There is just one more thing I'm really curious about: Did anyone of you ever listened to a 2604 compared to a pair of 627's??
Or, what's your general oppinion about the 627?
(Andrea, I know, it's mentioned somewhere in the "best opamp"-thread, but I can't find it at the moment...)
Or, what's your general oppinion about the 627?
(Andrea, I know, it's mentioned somewhere in the "best opamp"-thread, but I can't find it at the moment...)
Ah, I see what you meant (now). 🙂
That said, 47uF was probably just because they were the same value used elsewhere in the circuit.
In an older Super Pro I went as far as trying with Wima MKP4 .15uF (150nF !) 250V, and I experienced no perceptible bass cut off (with the 47K input impedance of my amp) even in that 'extreme' case..
Last edited:
The OPA2604 is mainly made for audio, the OPA627 not. Why not try the OPA1642, which is kind of the perfected heritage of the OPA2604 ? 🙂
I find it mediocre, like the OPA2107.
A lot of people rave about them, of course, and I seem to remember these were the chips that first induced somebody to build a single-to-double adapter.
They are pretty much prohibitively expensive, as far as I'm concerned. Maybe I'll get TI to send me samples, some day.
Ecaroh.
Good luck, they're not in the sample program. You have to call and convince somebody at a TI distributor to send you some.
I can tell you that the cost is not without reason. The OPA627/637 is the best sounding single op amp I have listened to/experimented with bar none.
Precise, coherent, excellent tonality, perfect frequency response from top to bottom. There is nothing it doesn't do well, IMO.
Many well regarded commercial products have been built around the OPA627, including reference grade line stages.
Last edited:
Andrea, Martin, others? -- How difficult was it to desolder the output caps? Was there any glue underneath? Those pins look like it might not be easy to desolder one side, bend it up, and then do the other side.
So they used tantalum for the output caps.... Anybody else find this strange?
Thanks,
Ecaroh.
So they used tantalum for the output caps.... Anybody else find this strange?
Thanks,
Ecaroh.
Hi Ecaroh,
it was quite easy to desolder the output caps. There was no glue or anything. In my case (as I can recall) they were solderd the way my tiny little drawing shows. So I could easily desolder the right contact first and lift the cap a bit, then desolder the left contact (you can apply the heat from the soldering iron at the side of the cap due to the L-shaped leads).
it was quite easy to desolder the output caps. There was no glue or anything. In my case (as I can recall) they were solderd the way my tiny little drawing shows. So I could easily desolder the right contact first and lift the cap a bit, then desolder the left contact (you can apply the heat from the soldering iron at the side of the cap due to the L-shaped leads).
Attachments
Last edited:
Hi hags,
thanks for your comment! I think I will try the 627 in the near future. Perhaps everybody who is interested in opamp-sound has to give the 627 a shot at some point!
A general note about cap replacement:
If you feed an amp or a preamp with a rather high input impedance (like the one Andrea uses with 47kohm) you can of course use very small values, even below 1uF. But since the output of the super pro with it's opamp output stage delivers a quite high output level (I think something around 3,5Vrms), one can also use a simple passive preamp, like I do. And if this one has a 10k pot at it's input it might be better to use output caps with at least a value of, say, 3,3uF or above to avoid any low frequency cutoffs.
If you feed an amp or a preamp with a rather high input impedance (like the one Andrea uses with 47kohm) you can of course use very small values, even below 1uF. But since the output of the super pro with it's opamp output stage delivers a quite high output level (I think something around 3,5Vrms), one can also use a simple passive preamp, like I do. And if this one has a 10k pot at it's input it might be better to use output caps with at least a value of, say, 3,3uF or above to avoid any low frequency cutoffs.
Last edited:
- Status
- Not open for further replies.