Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Audio-gd discrete op-amps reviewed: OPA-Earth, OPA-Moon, OPA-Sun v.2
New Posts  All Forums:Forum Nav:

Audio-gd discrete op-amps reviewed: OPA-Earth, OPA-Moon, OPA-Sun v.2

post #1 of 392
Thread Starter 
Thanks to my fellow from forum.MP3store.pl (Powered by Invision Power Board) whose personality is known to me and Kingwa, the discrete op-amps designer, I received three types of discrete op-amps from Audio-gd for review. The schematics and descriptions can be seen here: н¨ÍøÒ³ 1
The modules I have tested are:
dual OPA-Earth
dual OPA-Moon
dual OPA-Sun v.2
All came with capacitor upgrade and I didn't modify them in any way for the time of the tests. Each of them spent at least 24 hours of burn-in, observing if any changes occur. When I realized nothing changed any more, I started critical listening sessions. The equipment I used were:

Grado GS1000 headphones

Here is not much to comment. I regard these headphones as the top class dynamic model, with extraordinary spatial presentation and timbral abilities. Their bad side is long burn-in time, making some annoyed guys sell these headphones in premature condition to people like me.

Grado RS2 headphones

These are pretty well known to the head-fi community. I'll just point that mine were without buttons, I bought them new in February 2008 and in the last weeks, after trying some burn-in tricks working with the GS1000's these settled down on very nice sonic signature. Their role were minor in the whole test but this will be clarified later.

Creative Aurvana Live! headphones
For the purpose of the fellow who sent these discrete op-amps to me and chose his favorite using these headphones and the Zero DAC, IIRC. Mine are re-cabled, BTW.

Technics SL-PS840 CD player
This one has been modified heavily but I used it as the digital transport only. The mods I applied were among others:
- ultra fast MUR120 diodes in place of 1N4003 rectifying diodes (8 pieces)
- Elna and Sanyo capacitors replaced in many sections of digital and analog power supply
- long and thin PCB traces doubled with solid tin-plated copper wire
- precise 5ppm gold pin OMIG quartz put in the SOAD70A mechanism servo circuit
There were other strictly analog part related mods like capacitors and op-amps replacements which improved the Technics' sound significantly but it still was not the class of the modified...

CS8416/CS4397 DIY DAC kit as the DAC
I bought the kit from gigawork on ebay, choosing the SMD version which some people consider as the better choice because of short traces and separated power supply built on another printed circuit board. This DAC has been “discovered” and popularized mostly by Lukasz Fikus. What hi did to this DAC, you can read here: CD DAC Lampucera lampizator
I'll just admit that the stock form of this kit will not catapult you in the hi-end sound realm, it's rather significant upgrade to budgetary CD players and comparable to devices like 2x OPA627 upgraded Zero DAC, based on my forum mate's opinion. However, after some mods, this DAC kit starts to compete or maybe win (last upgrades haven't been verified yet) with $1000-$3000 DACs. At least it could fight with: AudioNemesis DC-1 upgrade+Tara Labs internal wiring, the NorthStar M192 MkI or theTheta DSP Gen Va. So, here is the short list of upgrades making this DAC a reliable, I hope, test platform:
- 20VA 2x 14V AC toroidal transformer in place of the original one (hum with OPA-Moon showed it's too weak)
- Elna Tonerex capacitors on each voltage rail on the DAC PCB
- shortened ground traces for the op-amp slot
- all supply voltages of the analog section of the DAC bypassed with the Elna Silmic II capacitors
- digital voltage supplies of the CS8416 SPDIF receiver and the CS4397 DAC chip bypassed with the Sanyo WG capacitors (better specs and sound than OSCON)
- Toslink receiver supply supply bypassed with the 220uF tantalum capacitor
- shorted output capacitors
All those mods were thoroughly tested. I tried many different mods in the last months, some were unnecessary, some required testing many different parts, etc.

Beresford TC-3618 optical digital cable
This is the best optical cable I have ever tried. Not distinguishable from the more expensive Profigold PGD5000, much better than some hyped “glass” optical cables having both mediocre build and sound quality. Use this one or choose a trusted manufacturer, that's my advise for today.

Moonlight v.8 headphone amplifier
The name tells you nothing and I know that because it's my own design. The details I'm willing to reveal now are:
- class A output
- no regular op-amps in the signal path (to hear the op-amps influence in the DAC)
- DC signal path with passive offset cancellation
- virtual ground architecture with some unique solutions
- external regulated DC power supply
This amp was roughly compared to the Yamamoto Soundcraft HA-02 and the RudiStor NX-33, from memory to the ASL Twin Head MkIII with upgrades (caps, DACTs, tube sockets, different tubes). For an opinion on this amp it's best to ask Piotr Ryka or fallow81 at this moment. Just in case you're not sure I could rely on the sound.

Analog interconnect
Having disassembled DAC and amp, I could get rid of this element, using a couple of centimeter long wires to connect one PCB to another. This sounds best to me. If not, I take the Conducfil SPEAKFIL 8896 cable, the Neutrik NYS373 plugs, and make my own IC connecting wires exactly how Oyaide did in the PA-02 TR interconnect. Then I could sell the Oyaide product as being less transparent and precise than the professional Conducfil cable.

Ayo “Gravity at Last”
Diana Krall “Live in Paris”
J. S. Bach “Brandenburgian Concertos I – VI”, Czech Broadcast Recordings
W. A. Mozart “Requiem”, Polish Cathedral Choir (forgot exact name)
Blue Note Trip 7 – Maestro compilation
Seal "Soul"
Basement Jaxx "The Singles"
Silverchair "Frogstomp"

Other albums, headphones and amplifiers will be subject to update this thread whenever something interesting appears, so be patient if you won't find all you're looking for at the moment.

As you all know, the discrete op-amps were designed to replace generic or brand name, available for sale in electronic shops integrated op-amps. Discrete circuits are considered to be superior to integrated circuits and it's explained in several ways:
1)The circuits are created for a specific purpose and voltage range, unlike the universal integrated op-amps. Here we have audio.
2)Discrete passive elements are isolated from the die and manufactured in the optimal technology for a specific solution. Implementing resistances and capacitances within the op-amp silicon die is limited by the integrated circuit technology.
3)There is no crosstalk and capacitive coupling between the circuit sections.
However, integrated circuits were created because of their advantages and these shall be listed below.
1)Unmatched thermal coupling between all active elements, keeping parameters like transistor current gain or DC offset at the output well compensated in the whole temperature range.
2)Short signal paths allowing weaker bandwidth compensation and using faster active elements for negative feedback circuits allowing higher slew rate, shorter settling time and lower phase shifts at high frequencies. Lower EMI vulnerability in general.
3)Minimized silicon-metal signal transitions. In the integrated circuits almost all current flow happens in the same piece of silicon. In the discrete circuit it comes from the pin to the silicon and goes to the pin-PCB trace-pin path before reaching another transistor's structure. Count the transistors in the the whole I/V or voltage gain section and see how many times the signal goes through the barriers like silicon-metal joints and solder joints. Sure, it can be done with high quality and possibly no loss to the signal, but in a continuous signal path within the same medium there is no such loss for sure.

The discrete op-amps were designed mostly as the replacement for the regular op-amps but also as circuits around which you can build your own audio equipment. On the other hand, when going discrete from the scratch, it's disputable to use global feedback loop dedicated circuits, so let's stick to the upgrade role of the Audio-gd discrete op-amps for now. In my DAC circuitry they worked as the low-pass filter and buffer circuit converting symmetrical CS4397 voltage output into unbalanced signal. The supply voltages were +-12V. I allowed each discrete op-amp to heat up before I started listening. For each discrete op-amp the additional grounding necessity was evaluated and appropriate comments will be provided.

Unfortunately I don't have Burson discrete op-amps in stock so cannot compare to these from Audio-gd but there are several things regarding popular integrated op-amps that should be pointed out. I tested lots of them, many times due to possibility of receiving free samples, many other times making trades or just buying them. I could say I know almost all of those you could use for audio, so let's categorize them somehow.
Category 1 – waste of time and money
Without further comments, I'll just tell you I put all the popular audio op-amps here, too much for listing. Instead of this I'll mention the op-amps I think are fine but couldn't use in the DAC: AD8021, AD8022, ADA4841-1/2, OPA637(?). Those I regard somehow more than Category 1 will appear later in this section with a remark.
Category 2 – nice but incomplete sound
AD823 – quite popular in voltage stages of amplifiers and CD players after upgrades, doing fine but keeping it's own timbre, not allowing for full color reproduction.
AD8620 – the A series is a bit harsh and aggressive, the B series is smooth but still too present in the sound. Going to the bright side and keeping sounds a bit confined. I'd appreciate more independence.
AD8397 – this is a smart one, really. At the first moment you think it does everything perfect, the sound is complete with no obvious flaws and predefined timbre. What it does bad is shrinking and flattening of sounds, some offensive forwardness due to the thinned sound contour.
OPA2111KP – holly crap, this one is pretty neutral and sounds better than you could expect from the specs. Very similar to the OPA2107 in this regard but sonics just better. Truer bass with more impact, more coherent and convincing rest of the spectrum. On the bad side is a piece of mess and congestion. Sounds get mixed and interfering each other starting from higher midrange sometimes.
2x AD825 – man, I've liked this one for a very long time. I still could find use for these, like entry level headphone amps or budgetary DAC output stages. This one is lively, rich, fast, with long treble finish, nice midrange timbre. Sounds like a sweet spot between the AD8620 and the AD744 which gets to warm, thick and uncontrolled sometimes. The AD825's unique midrange can “save” your rig when you cannot achieve enough transparency. The overall order and separation is less than perfect but not obviously flawed. I blame the highs for being a bit accentuated and with artificially sustained decay.
OPA2604 – a classic audio op-amp. This one can sound really bad when supplied badly. Just see its PSRR specs. Mere 84dB. I could confirm it. Unregulated power supply makes it sound congested, harsh, unstable, messed-up and offensive. Well regulated and filtered supply voltages keep it quite neutral, with dimmed and unfocused trebles, good pace (LOL at the 2x OPA627A here) and overall good tone.
LM4562 – just a lesser version of the LME49720. Its midrange is on the warm-bright side instead of neutral, overall good and rich presentation, unfortunately too obviously defined hence predictive and limited sometimes.
Category 3 – sound almost complete but imperfect
Here I guess additional comment is required... Some people say op-amps are bad. They just sound bad and sound like op-amp. Op-amps do weird things to the sound and using transparent amplification or acoustic transducers will let you hear that. People who want go higher with their equipment just abandon the op-amp realm and go full discrete – using transistors and/or tubes only.
The effects of op-amps are something like these:
there is an invisible barrier in the soundstage, in front of you, or surrounding the stage
the three-dimensional image of the sound is flat, corrupted or fixed to the soundstage like a picture to the wall
not enough freedom, clarity, space and timbre differentiation, and other nuances making you just hear sonic spots instead of seeing and feeling the instruments and performers' breaths.
So let's enter the area where you can experience touch of true performance, with exceptions but we're yet on the other side.
2x LT1028CN8 – wow, with the Blue Note Trip I can say I'm between the performers. The trumpet is properly drawn, I feel the air and the presentation is convincing. There is a slight tint of nervousness in the midrange, which gets sometimes slightly dry but sure you can hide it using passive components. Regarding the dryness – these op-amps just blow away the popular LT1364 in this field. If you like the latter, use 2x LT1028 instead.
AD8599 – this one can do things most other op-amps cannot do, period. It's like a sport car with big supercharged engine and huge carbon brakes. It can accelerate and brake rapidly thus keeping perfect rhythm, pace and timing, proper release and decay and exact texture. It's warmer than the LT, with more delicate, thinner trebles, no nervousness but this strange direction of the soundstage. It feels like it's not completely open towards the listener.
OPA2228P – say, the OPA2604 killer when you can use bipolar and uncompensated op-amp. Higher midrange complexity, more coherence and focus. Sometimes laid-back and the trebles could use more hues.
Category 4 – like a completely arranged room, not quite tidy yet
I'd say, the fight with the Audio-gd op-amps starts making sense here.
Did you ever know that the op-amp grade matters for audio? Yes, it does, at least for OPA2134, OPA2132, LT1028, AD797 and OPA627 which I can confirm. So, let's continue:
2x OPA627BP – welcome to Category 4! Unlike the AP or AU series, this one is not lacking proper attack. Sure, it's not lightning fast but at least this one can keep the rhythm, it's actually pretty similar to the AD797ANZ with the advantage of JFET inputs. Lack of the last refinement and oomph makes it less than perfect. I'll get back to this later.
2x AD797ANZ – this one I regarded for a long time as a piece of wire equivalent when used as a buffer. It lacked some level of the lowest detail but added pleasant tube-like timbre, in a delicate form of course. This one is a bit better than the AD8599, with less accentuated sonic signature.
2x LT1028ACN8 – the better LT1028 grade. Sure it's audible. Instead of the slight dryness you get a bit longer sustain in the mids, with a slightly higher midrange coloration, hardly perceivable when your ears get used to it. The soundstage is correct, the imaging is correct, tha PraT is correct, and what I would like from it to be perfect is a bit more independence of the events in the furthest background plans. It's sounds like it's a bit hazy there, however you can like it.
2x OP27GP – no kidding! Just see the prices of the OP27E from Texas. There is a trick – this one, I mean the one from Burr-Brown I own, requires class A biasing. Otherwise it's not worth considering. I remember I liked it even more than the 2x LT1028ACN8 combo. As a side note I can write that the OPA627 was designed to have this op-amp offering JFET inputs. Same with the OP37 vs. OPA637.
2x OPA211 – the king of the space, I guess it's even exaggerated. In terms of overall order, clarity, timbre transparency and rhythm, this one is a direct upgrade from the OPA2228. Slight but worth it. The increased shape detail and huge soundstage is probably the result of something I'd call sound disassembling. It's like losing spatial coherence of each sound, having the tone in one place, and the harmonics just aside, not surrounding the right tone. Strange but true.
Category 5 – complete sound traveling in the clean air
2x AD797BRZ – this is the op-amp truly deserving the name of the Analog Device. This one is a smoother and fuller sounding version of the AD797. Switching from the ANZ to the BRZ series is like going vinyl instead of digital. Everything is smooth to the moment your music craves for the harsh texture like aggressive violin passages, distorted guitars, or stronger double bass phrases. The readability of the furthest plans is great, without going into image sharpening effect which makes you see the details more but lose the feeling of perspective at the same time. Roughly, it's like the AD79ANZ with all advantages of the OPA627BP added. The soundstage is nowhere limited and the sound image of whatever possible shape. This is a reference integrated op-amp for me so let me know as soon as you find better, and I will verify it.

Let the show begin... first impressions
I started with the OPA-Earth as being the oldest and most recognized design from Audio-gd. I put it direct after AD797BRZ pair which I normally use and wow! Bigger soundstage, more saturated timbre. However, after some time I realized the sounds are too punctual, the soundstage is a bit predefined like in some op-amps and I started thinking what to improve? I used the ground wire. There are two polar opposites regarding my DAC where I can connect this wire – as close to the DAC chip as possible, or straight to the output connector. The difference was almost none, if really any, so I kept the DAC ground connected. This improved the vocals in the first place. The singers were no longer strangled, the shape of the voice got opened and not so focused, say more natural. Overall sound imaging improved, getting a bit less spectacular due to lower timbre emphasis but the exaggerated punctuality disappeared as well and the presentation became simply truer. What I still want? Give me back the smoothness and intimacy I had with my AD combo. We'll talk later as you're straight out of the box.

The second one to try out in this impatient rush caused by hunger for knowledge was the OPA-Moon. Let's see what guys call “sentiment” and tube-like presentation. Hmm, sound is more mellow and intimate than with the OPA-Earth but it's going to create it's own world more obviously as well. There is more transparency within the soundstage but on the other hand you hear it's borders stronger. It's like inflated from the inside but limited at the borders. The sound is smoother and a bit rounded but the soundstage is slightly uncontrolled. I connected the ground wire again and it helped for the second time. The change is slight but for the better. The imaging got less spectacular again but with lighter intrusion into natural imaging as well. On the other hand I had impression that the OPA-Earth is still closer to the transparency.

So, finally the Sun v.2 enters the socket and from the very beginning I know I like it! What a positive sounding device! The first thing making it superior to the other two is that the headphones disappeared from my head. The presentation is so coherent and convincing that I feel no strain and necessity of additional resolving of what I hear. This one is brighter than the previous two, with denser presentation and overall thicker sounds. The timbre is rich and nice, pretty straightforward. Connecting the ground wire I found as a bad idea. The sonic presentation got flatter, less of being everywhere and the instruments got a bit confined to their localization. In case of the OPA-Sun I decided I'd choose more of the sonic freedom and joy instead of stability making the presentation more flat at the same time. Another good thing about this device is that it's sound is unique and you can forget all integrated circuits when looking for sound like this.

Ayo “Gravity at last” album impressions
Well, well, well. This CD is very well recorded and mastered. Very transparent stage, rich tonality, good presence of voices and mechanically produced sounds. This one is better mastered IMHO than the previous “Joyful” album.

I forgot previously to define the competitors. I decided it to be 3 against 3, so the representatives of the better class of OPA627, LT1028 and AD797 will fit the bill, I guess. Going from the best to the worst, it looks like this:
4.OPA-Sun v.2
The winner got it right. Maximal midrange complexity,transparent ambiance, great performers' presence near you. The OPA-Moon had maybe more interesting midrange tonality at the first moment but it tried to put all the sound in it's own frames, and all in all the midrange wasn't quite complete, however rich and enjoyable. The LT was actually similar to the OPA-Sun v.2 but won due to smoother and softer presentation. I found it more true than that of the OPA-Sun because the latter felt a bit like hard wood in the texture. Both are the brighter part of the bunch. The OPA-Sun v.2 easily won with the OPA627BP pair. For this great album the demarcation line of the “op-amp smell” rose to the level of the OPA627BP. It was a bit grainy and flat in the foreground. Good tonality but sounds embraced with a barrier. The OPA-Earth got it even slightly worse. Similar problems and some congestion added. Just like the OPA627BP without the ambiance transparency.

2.OPA-Sun v.2
First two places provide pure enjoyment. The AD797BRZ won due to clean atmosphere (again), natural, not offensive tone and great instrument definition. The piano showed its supremacy and the woodwind second. With the piano you hear great string decay echoed in the wooden box. Yes, you hear it's all like that. The OPA-Sun moves everything towards the “wooden” side a bit. The trombones played tad too low, without upper finish but overall presentation was logic, coherent, powerful, enjoyable, keeping great rhythm and pace. There was no congestion or disorder. OPA-Moon sounded mellow, with lesser energy, rather nostalgic which somebody might enjoy, especially with some tracks. The sound was kind of caramel-sweet but it was a slight intrusion, not forced in every note. The soundstage wasn't completely open but on the other hand, presence of the instruments was well accentuated. The OPA-Earth presented a bit nervous and congested sound. It's like you feel there is something in the air and you can't relax. However it was still on the good side with overall correct texture, a bit flatter presentation than the top three, and surely better than the pair of the LT1028's. These showed too rounded tone, to the level it was greasy and boring. Not enough separation in this music sauce for me. The OPA627BP's were disappointing. Nice tone but corrupted piano with no piano presence. It was simply synthetic, with uncontrolled note decay and flat, “electric” timbre. Like plastic toy. No fun, no score.

Seal “Soul” album impressions
6.OPA-Sun v.2
Instead of live instruments and minimal mastering, we have here Warner Bros logo at the bottom of the CD box and obvious sonic idea about this album. It sounds sentimental, smooth, with no aggression, giving impression of a big hall, with somehow distant vocal. Slight echo adds the feeling of empty audience, just the performer and the orchestra more in the background.
Looks like heavily engineered mastering is a domain of chip op-amps. The first two places were quite close to each other. I rate neutrality higher, so the AD is the winner. Another reason is that the AD797 reached further on the vertical axis. The LT sounded fuller and more plushy in the higher midrange, making vocals more kind and friendly. The AD was more groovy in the voices department. The OPA627BP's showed simple but coherent presentation with good transparency. Leaning towards plastic but acceptable. The OPA-Moon was more physiological but with lesser clarity. I'd say they were equally good and somebody could change the order between them in the ranking just due to preferences. The OPA-Earth was similar to the LT but with worse clarity and uneven texture. The LT1028's presented more culture and coherence. The OPA-Sun v.2 with it's specific texture preferences did not match Seal's voice and emotionality of this album at all. The sounds were kind of crude and thick, not getting smooth enough when expected.

6.OPA-Sun v.2
Here what I care for is tonal complexity, layering, soundstage dimensions, textural abilities, merging in music - simply said. The AD provides the magic touch that makes you perceive the sounds not only with your ears. They make you imagine what other senses could project in coherence with hearing. The OPA-Earth provided some positively edgy fun, with complete tonality and precise texture, however not that transcendent you'd expect from good electronic music composers. This one is correct and enjoyable without entering the fourth dimension. The OPA-Moon is more emerging in the whole music ocean, and if you care more for this instead of every possible timbre delivered, this one, with its caramel presentation will fit your expectations. The soundstage is perceivably huge, especially wide, and you're in the center of the music event. The LT1028ACN8's were smoother but sometimes distant, making sounds escape a bit from you. The tonality was complete but the texture too rounded, and the haze in the background. You can like it but it's added value so I rate it as a disadvantage. The OPA627BP's made it simple. If you don't know your recordings well, you think it's the clear picture of them, with plain notes, correct, nice tones, not so big soundstage and kind of disappointing simplicity. The OPA-Sun v.2 tried to paint a rainbow using roughly five crayons when seven is required, and the hand was shaking when the picture was born. The footprint of this shaking is this repetitive texture, not differentiated enough, and limited tonality due to missing paints. Using nice colors only is moderately fun. The forward presentation doesn't feet electronica, either.

Silverchair “Frogstomp” album impressions
I'd say it is the rock/metal capabilites test. This album was their first and best, with it's young spirit and garage studio feeling. I like it all, some harshness, some anger, true acoustics, nicely distorted guitars. So, the score goes like this:
4.OPA-Sun v.2
The winner is in another league, magic of reality and presence, with no further excuse. Places 2 to 4 is roughly the same, a bit lower league where the OPA-Earth added some nice harshness to the vocals, emphasized the distortion of guitar strings and kept great rhythm and timing. The LT1028ACN8's scored a bit higher due to more present vocal and richer guitar tone, making each pluck perceived like a wind blow. The OPA-Sun v.2 kept the vocalist “sane” and effortlessly singing each phrase, the scene was broad and open, everything was happening like on the stage close to you, with nice tonality and texture. The sound was believable, with no issues. The OPA-Moon got too mellow and too dark, making music sentimental but sometimes absent, not his music genre, definitely. The OPA627BP's showed all the legendary chip op-amp excellence, not even worth describing it. It's an exceptional album and want to hear that. I didn't.

Final assumptions
Well, I feel I know much enough. There should have been jazz impressions added but actually the jazz instruments appeared in classical and rock sections, with the female vocal on Ayo album evaluated, so I think I can summarize my thoughts on the discrete Audio-gd op-amps.

First, guys from Audio-gd achieved what they planned, really! They delivered a worthy chip op-amp replacement, presenting three different sonic signatures. For the studio and hi-end purposes they are probably not good enough, but for most DIY aware audiophiles they are an interesting and enjoyable upgrade, in the league that mid-end equipment will benefit from them. If you can assure the conditions below are possible to fulfill, then go ahead and try one or more of them, considering their sonic tendencies:
- efficient power supply for both rails, I guess 200mA reserve per voltage is the minimum
- enough space around the socket or near it, when the extension cable is required, you have to provide mounting of the OPA
- easy access to the ground on the original PCB, as the OPA ground wire might bring advantage when connected

This one feels best in electrical and electronic music, where raw power and aggression is an advantage. This one has got the ability to present some of the sounds live, like from your surroundings, not headphones, even on lesser models like Creative Aurvana Live! or Grado SR325i (thanks to my friend for loaning them). The bass is shaking and precise, the mids are colorful but slightly on the flat and thin side, however tonally the most complex of the three discrete op-amps. This one is doing better with the ground wire connected as you get rid of nervousness and the vocals melt in the air more naturally.

The sweetest and the most romantic one. Slightly darker than the other two, a natural step-up from the AD825 pair as I know there are lots of their fans. This one has got big but defined soundstage, the listener becomes a part of the spectacle. You feel emerged in the music flow however the musical fluid is not completely spiced for me. I don't want things to be sweet or salty all the time, I need a tiny sip of sour or bitter flavor sometimes, making the dish more refined. In other words, the music does not have to be sentimental all the time, I'd like it to be intriguing and surprising sometimes.
This one better leave ungrounded.

OPA-Sun v.2
The best first impression, losing a bit when thoroughly discovered. It simplifies the harmonic content the most. There are some tonal nuances the OPA-Sun does not deliver. The musical presentation is straightforward, positive, joyous, like a sunny day. I liked classical quite a lot with it and some people will enjoy rock because its ambiance is clear and fitting headphones well. The OPA-Sun removes the headphones from your head on most albums I tried and if you don't know, or don't think about some tonal particles missing or hidden, you'll enjoy this one. The texture is wooden thick, the midrange is on the rich side, all is brought to the listener without further excuse, honestly and friendly.
For more precise imaging, leave it ungrounded, for smoother and more airy presentation, connect the ground wire.

My favorite? OPA-Earth. Delivers the most of what was recorded but not in the best form each time. Then comes OPA-Moon and the OPA-Sun v.2. The OPA-Moon is missing something sometimes but it sounds like a young man who fell in love just forgot about something, forgivable and natural. The OPA-Sun v.2 is kind of a cottage musician. He's got a lot of willing to play and positive energy. He plays notes the only way he can do but he never forgets to smile to you, and you smile with him. Pure, simple joy.

UPDATE 11 Jan 2009
I rebuilt the power supply so that the digital and analogue section of the DAC chip gets double regulated +5V supply. I find the OPA-Moon more complete tonally and it looks like it's at least equal in terms of SQ to the OPA-Earth. All discrete op-amps improved a bit with this modified PSU but so the AD797BRZ's did. The chip op-amps got better in terms of acoustical correctness and transparency.
post #2 of 392
good review. I'm very interested as I'm about to upgrade my op-amps.

Do you reckon you could put up the costs of each of the op-amps your comparing. That way we could see some kind of price to performance comparison too.
post #3 of 392
Very nice review majkel... i'm waiting for your full review

Btw, recently I bought OPA Sun v2 from my friend. I put it on my SOUL v1 headamp (designed by my fellow friend). Here's a little spec of this headamp:
- 3 channel + buffer architecture
- class A bias for voltage-gain stage
- precision virtual ground
- Jung multiloop

Since I have only one SUN v2 (dual version), I put it on ground channel at buffer stage. For my lack of english & ability to review , I could only say that the first thing I noticed is a significant wider soundstage and fuller sound compared to my previous opamps (AD8397, 2x AD825, AD8066, OPA2134). Btw, I supplied only 19V to this headamp, so it's 9.5V rail-to-rail(?).

Next time I want to put more discrete opamps to replace my L/R channel too. Maybe I need to modify my amps power supply stage to deliver more current to feed ups these HDAMs.

Here's a photos when I tested it with DIYmod iPod, custom made LOD with 22uF BG NX Hi-Q inside, Grado SR125000 (modded like Alessandro MS1000):

post #4 of 392
^Nice! Is that sum OPA + BUFFER Three channel amplifier? Needs more details on the self etching.


I've been too absorbed by the AD825 that I did not even bother to try the AD797. If you ranked it two ranks below the AD797, that I MUST TRY.

(though i'm pretty lazy about balancing input bias).
post #5 of 392
Thread Starter 
It's easier to make the AD825 sound right than the AD797, mainly in regards of supply. As I said, the AD825 is playing its own melody but this melody is one of the best available. However, sometimes the original beauty is nicer and you'd like to hear it.
post #6 of 392
Which one would you recommend majkel? sounds like the discrete opamps aren't heaps good. Is the AD825 expensive?

when's the rest of the review coming up? too much suspense haha:P
post #7 of 392
Thread Starter 
Originally Posted by Bleemv1 View Post
Which one would you recommend majkel? sounds like the discrete opamps aren't heaps good. Is the AD825 expensive?
They are very good but not perfect. They win with most chip op-amps. Maybe higher supply voltages will make them better.
when's the rest of the review coming up? too much suspense haha:P
Probably today in the evening. I need to comment other albums and music genres, add a word regarding the RS2 and final conclusions.
post #8 of 392
Excellent job on the review man !!!

I hate to break this to you but all the audio-gd modules need at least 250+ hours to settle completely. They will all improve dramatically once you get past 175 hours with incremental further improvements to 300 hours. They take a long time to form for some reason. The original module (that LC and Burson Audio sell...also an audio-gd design) took in excess of 175 hours when I reviewed it for the Zero last spring ( I now have all the versions of HDAM both V1 and V2)...it was maddening since I was trying to get the piece completed and the dang sonic goal posts kept moving on me.....

Again...a simply outstanding review !!!!


PS Your assumption about a stiffer PSU for these is right on the money....these modules love a good PSU (best case is around + - 15-17V at 500 mV min which includes current needs for surrounding output stage of course).
post #9 of 392
250+ hours!!! that's heaps. did you find the HDAM modules sounded better than opamps after 250+ hours?
post #10 of 392
Originally Posted by Bleemv1 View Post
250+ hours!!! that's heaps. did you find the HDAM modules sounded better than opamps after 250+ hours?
For me... Sun v.2 HDAM definitely sounded better than AD8397, 2x AD825, 2x AD8065, AD8066, OPA2134, OP270 (this is all that I own) even though burn-in time is less than 10 hours (I bought it from my friend and he use it no more than 1 hour just to test if it's worked). The sound is much wider and "full" than these ICs.
post #11 of 392
Originally Posted by Bleemv1 View Post
250+ hours!!! that's heaps. did you find the HDAM modules sounded better than opamps after 250+ hours?
I wouldnt' worry too much... transistors or opamps "burn-in" is mostly related to someone's memory being unable to remember how it sounded 250 hours earlier. Not to any actual changes in the semi conductors
post #12 of 392
Thread Starter 
I haven't noticed any changes so far but this thread will be updated with my impressions. The aforementioned AD825's which I compared yesterday with the OPA-Moon got one strong kick from the discrete op-amp and went to the box ashamed. It's another league. The OPA2134 or the OPA2132 are interesting in the P series only, the PA can sound different from one production date to another. The PA series and the AD8066 is no competition for the discrete op-amps.
There is oncoming opinion on rock, electronica to add to the review in the evening.
post #13 of 392
Originally Posted by Bleemv1 View Post
250+ hours!!! that's heaps. did you find the HDAM modules sounded better than opamps after 250+ hours?
To my ears without a doubt. I do prefer the Earth and Moon modules though over the SUN (V1 at least, I have a pair of V2's cooking now).

The dynamics, texture and full freq spectrum take time to fall into place.....it's the damndest thing to think they require this time but after going through this process now for 10 HDAMs it's predictable and repeatable each time.

New ones I get automatically get a pass until at least 150 hours have passed then I get into the process from there noting the last major changes over the next 100+ hours.

post #14 of 392
Thread Starter 
OK, I will let them stay in the socket but it's only one per round possible, so this will take same time. Will I get much improvement replacing the regulators for +-15V range, or would it be just slight? I don't want to do it aimlessly as I will lose possibility of using some chips.

Peete, you can try a pair of the AD797BRZ eventually. If you don't like them, I'll take them. Maximum supply voltage is +-15V.
post #15 of 392
Originally Posted by 00940 View Post
I wouldnt' worry too much... transistors or opamps "burn-in" is mostly related to someone's memory being unable to remember how it sounded 250 hours earlier. Not to any actual changes in the semi conductors
How many of these modules (if any) do you have in your possession ?

My memory is fantastic !!! An easy test is to have 2 of the same type of HDAM, burn in one fully and then swap it for a fresh one...the differences are obvious and immediate.....you'd have to be deaf not to hear the difference....(sideways jab back at ya ).

New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Audio-gd discrete op-amps reviewed: OPA-Earth, OPA-Moon, OPA-Sun v.2