The Aum Acoustics
2A3 Tube Power Amplifier
One of the finest sounding statement quality stereo 2A3 tube amplifiers money can buy . . . available only on a custom build to order basis (but only to a super qualified audiophile!). I don’t make enough money to build this amp so I’m very careful choosing who I will accept an order from.
Please realize, that building this amp takes up about 3 months of my attention and several hundred hours of actual build time. I could get paid more by working at minimum wage. The result I get, however, is so worthwhile that I will continue to offer my expertise on a very selective basis.
I do want my LIO and Harbeth client’s to know that I not only care intensely about creating high audio quality but have gained my knowledge from direct experience where I’ve achieved my goals through my own hands on success – at a very high level.
My own 2A3 amp (paired with my own LDR based volume control solution) is the only other system that I will recommend in addition to the Vinnie Rossi LIO. This 2A3 amp isn’t for everyone, however, and does not allow for playing all genre’s of music. It is specialized in what it does but what it does do is performed at the highest audio level possible. If you want the finest sound quality for small scale music genre’s, this is your “beast”. If you would like to enjoy all music and achieve an extremely stunning audio quality please consider the LIO as a better overall solution.
(17″ wide x 9.5″ deep x 2.25″ high w/o feet)
It takes about three months to build this incredible amplifier and the two slowest items involved are the chrome plating for the top plate and obtaining the Hashimoto transformer set from Hashimoto in Japan. This is definitely not a production item nor will there ever be a lot of them built. Instead, it is a “cost no object” design and my goal is to create the absolute best sounding 2A3 tube amplifier built in the best way that is possible. I’ve accomplished that goal and am willing to sell this unique “one-of-a-kind” statement quality amplifier for actual parts cost plus a very fair and reasonable markup for my labor.
My raw parts cost for this amp is over $5,000. The tube selection that I’ve tuned this amp for costs around $1,200. It is imperative that a pair of Emission Labs 2A3 power tubes is used along with a NOS Mullard GZ32 rectifier tube. The pair of 6SN7 tubes are selected to match each particular audio system depending upon whether you need complete linearity or an emphasis in a particular frequency spectrum area.
My labor cost will be negotiated with a new prospective client and is definitely a serious value when you find out how many hours goes into building this amp. That’s also without consideration of the research and development needed to create the entire amplifier design. I refuse to give my time away as I have in the past and if no one orders this amp, I’m fine with that.
I have one, love it, and know that it provides the absolute best sound quality possible in audio today. I do not believe that there is a higher quality level obtainable. Staying as close to the first watt of power and using a world class high efficiency speakers system results in a truly fulfilling and extremely unique sound quality. However, everything in audio is a trade off and a better purchase value would be that of a fully loaded Vinnie Rossi LIO paired with the Harbeth Super HL5+ speakers. An alternative would be adding the Vinnie Rossi VR120 power amp and moving up to the Harbeth Monitor 40.2 speakers. However, my vote is for the HL5+ combination.
The LIO plays all genres of music extremely well (particularly when paired with Harbeth) whereas this 2A3 amp is very specialized in what it can do and speaker matching is extremely difficult. Nothing can touch it, however, where it shines if you pair it with a high quality full range speaker system like that made by Rethm in India. I used to be a Rethm dealer and absolutely respect what Jacob George at Rethm has accomplished for my kind of high-end tube amp design.
My additional requirement (before accepting an order) is that the new owner is capable of and will provide a serious and honest review of this amp. I’m looking for a “seasoned” audiophile who already owns a super quality pair of high-efficiency speakers (97db or higher). I’m also looking for someone who knows audio well and is a serious tube amp lover from owning more than one. I want that new client to have owned and listened to other amps to be able to appreciate how good this 2A3 amp sounds in comparison. It does but that will not be obvious without prior tube amp ownership experience and hence my unique requirement.
If I was able to efficiently allocate my time and not waste a single moment, it would take at least 2 full time weeks to build this amplifier and probably well over 100 hours. However, CNC machining, chrome plating, hard coat anodizing, custom painting, parts ordering is a waiting game and it literally takes about 3 months to complete all of the tasks to accumulate the necessary finished parts and to build this amplifier. It’s a hurry and wait game with many hurdles to traverse through.
It takes a huge amount of patience and the labor investment is intense. Also, there is considerable testing that is done once the build is complete and then a proper listening and burn-in process that follows. I cannot imagine any company building this high quality an amplifier and expecting to make a lot of money. It just takes too much time and is the reason this will remain a very exclusive product and one that I’m not hurrying to build. I’m also not willing to give my time away and consequently, this will remain to be a very exclusive project. I have no interest in doing this in any other way.
My original goal was to build what I call an “archival” reference amplifier, one that will be able to endow many generations of people for a long period of time. I’m very serious about this and feel that including this intangible benefit is something that I must do. This amp’s sound quality is so good that I would like many others to be able to appreciate it.
The Hashimoto transformers that I use are able to be used for at least 50 years before needing repair, re-building, or replacement. In a world of “throwaway” short-lived audio gear, this (for me) is a worthwhile goal and one that allows me to endow other people beyond myself. The market for ultimate quality is small. It also is not one where money making and profitability can take the front seat. Instead, money should be the least consideration. Important? Yes, but not the proper way to create this kind of archival vehicle. Money is only required to purchase the proper grade of parts and to pay for (even though not enough) a reasonable amount of time involved. Without your having some “skin in this game” I’m not willing to offer a gift of my time.
CHECK OUT MY DISCUSSION OF THE DESIGN OF THIS AMP USING THE ACCORDIAN STYLE SECTIONS LISTED BELOW:
There are a number of ingredients that must be present and included to make a 2A3 tube amplifier one of the best reference examples available:
My foundational goal for this custom 2A3 amplifier is to provide decades of use for as many generations of people as is possible. The Hashimoto transformers are built to last at least for 50 years so why shouldn’t the rest of this amplifier also be able to span that time frame?
I’ve built this amplifier for at least 50 years of use. To make that possible, however, there will be times during this period when electrolytic capacitors will need to be exchanged for new ones. That will be easy where I’ve used point to point construction. I have, however, used a printed circuit board to wire the two 6SN7 tubes. Printed circuit boards don’t do well when original parts need to be de-soldered and replaced. A good technician can probably do this at least once. But to achieve a 50 year minimum usage, I’m also including 2 unpopulated printed circuit boards so the entire original printed circuit board can be replaced with a newly populated board at least two more times during this 50 year period.
Also included, will be a detailed PDF format layout showing all connections for this amplifier. Anyone can follow the wire connections and parts connections and can in fact use this layout to build a new 2A3 amplifier from scratch. I have no problem sharing this information. If someone wants to copy my work, they should be given the chance to do it correctly rather than messing something up.
In all honesty, to build this amplifier as I’m building it, there is no or very little money to be made by copying my approach because it takes a huge amount of time to build this amp. Go right ahead if you like. It is extremely time consuming to build this amp and the parts quality and chassis design are such that just at raw parts cost, this amplifier is already maxed out from a price standpoint. There is little room remaining to charge for labor and definitely no room available to add on incredibly expensive distributor/dealer sales margin. If you want to make money, you would be better off using the cheapest parts and having the amplifier built in China but even their labor rates are going up. Even then, it would be tough to make any serious money if you approached building a 2A3 amp like I’m doing. It just doesn’t make economic sense. It is, however, exactly how to build the best possible sounding 2A3 tube amplifier and is why I’m doing it.
I’ve provided more information regarding the pluses and minuses for using printed circuit boards and point to point wiring in an accordion link below. By including two blank printed circuit boards with this amplifier, I have built in a serious amount of longevity and will have accomplished my original goal: to build an amplifier that will last for at least 50 years of use and provide the highest possible sound quality.
I will provide the new owner of this amplifier with a detailed layout of all wiring and parts used for this amplifier including the parts manufacturer and all values. This amplifier has been designed around all Hashimoto transformers and they are the foundation on top of which everything else depends. These transformers are some of the best sounding ever made and are constructed to provide at least 50 years of useful life. Yes they are expensive, but they are so worth it! So I don’t see needing to replace any of these transformers over the useful life of this amplifier. Only when these transformers finally wear out will the lifespan of this amplifier have reached its ultimate goal. Even then, the transformers could be re-wound if you could find an “old grandpa” who is still around (unlikely) and who is capable of such a highly skilled re-build.
If you just use the circuit as designed, you’ll obtain about 4 watts per channel of super clean, exotic sounding, seductive tube developed power. If there ever is a problem with this amp, it will probably be when electrolytic capacitors have reached their life expectancy and need to be replaced (probably 10+ years). All resistors are super high quality so should last much longer (probably as long as the transformers). Electrolytic capacitors can easily be replaced and I include two extra unpopulated printed circuit boards for the pair of 6SN7 tubes to make capacitor replacement a viable alternative for the next 50+ years. I suggest replacing the printed circuit board after one set of electrolytic capacitors has already been replaced. I’m estimating that you’ll need to use the first of these two pcb’s in about 20 years.
I challenge anyone to compare this amplifier’s sound quality to anything available anywhere else. Other than raw output power, the sound quality of this amplifier will not be able to be exceeded. If power is increased (by parallel circuit design, etc), sound quality will suffer. This 2A3 amp puts out 4w/ch of power and is just enough to provide good headroom for a high quality high efficiency speaker system. It, however, isn’t far enough away from the 1st watt where all of the original sound quality is retained. 4w/ch is perfect to provide the highest quality of sound reproduction and adequate power for high efficiency speakers. It hits this “sweet spot” just perfectly and is one of the reasons why 2A3 amps (particular this circuit design) are more popular than other tube designs.
When a tube amp builder mounts all tube sockets on a printed circuit board, there is no doubt that this is for the sake of lowering cost and increasing profit margin. The rectifier tube and the 2A3 output tubes all put out a huge amount of heat. When they heat up, the solder joints are expanded and when the amp is shut off, the tubes cool, and the solder joints are contracted. This heat on, heat off, situation gradually destroys the solder joints and before they fail, the sound quality gradually deteriorates.
This is a known fact and is the main reason I will not use a printed circuit board and even worse, solder tubes sockets to a printed circuit board for the rectifier and the two 2A3 output tubes. This just doesn’t provide any reasonable longevity for the amp since heat becomes your greatest enemy with a printed circuit board design.
The two 6SN7 input tubes, however, don’t get hot and their tube sockets can be soldered to a printed circuit board without adverse effect. The circuitry around the 6SN7 tubes is the most complex in this amplifier and actually it is here where a printed circuit board will “trump” point to point wiring. This is the only place in this amplifier where I recommend using a printed circuit board and it makes good sense, but only here.
In addition to using a printed circuit board for this one application, I don’t use the normal cheap printed circuit board style ceramic tube sockets like everyone else uses. I’ve designed my printed circuit board (after much trial and error) for mounting super high quality teflon sockets that are constructed using gold plated bronze phosphor pins. The strength of these pins is substantial and the care and soldering procedure that I use to solder these sockets to the printed circuit board provides a superior socket connection and superb longevity for this section of the amps circuit.
I use all point to point wiring everywhere else in this amplifier! In fact, I also install expensive Yamamoto teflon tube sockets for the three “hot” tubes and mount these incredibly built sockets on special ventilation “hanging brackets” that attach the sockets just below the surface of the top plate of the amp. This provides superb ventilation around these “hot” tubes and the gold plating on the Yamamoto sockets is heavier than any other brand and more robust than any other brand. If you’re interested in using your amplifier beyond 5 to 10 years of usage, these tube sockets will help you accomplish that goal. They definitely are worth the extra cost.
Here’s a photo of the 8 pin and 4 pin teflon Yamamoto sockets & the 8 pin teflon CMC socket that I use:
And . . . the highest quality audio transformers for tube amplifiers have always been Tango and Hashimoto. There are others that are good too but those two build the best output transformers as well as power transformers and chokes. When you select a transformer set for an amplifier, it is best to use transformers all from the same company. There is a synergy that is designed by Hashimoto when all transformers are built by them.
Tango is no longer in business and Hashimoto still has some of the “old grandpas’ still hand wiring their exotic tube transformers. However, these guys are disappearing fast and in the not too distant future, you will no longer be able to order new tube transformers from Hashimoto. This is a unique and dying art form and if you own a set of Hashimoto transformers, consider yourself blessed.
The output transformers in this amplifier are the 20W, H-20-3.5U single ended transformers. You might consider that it’s overkill to use a 20W rated transformer in an amp that puts out only 4w/ch. However, it makes a huge and positive difference. Prior to using this 20W output transformer, I was using the Hashimoto 7w single ended transformer. Even though these transformers were well over spec’d, they sounded a bit thin and a little bright when using my reference quality Rethm speakers from India. When I installed the 20W version, I was immediately amazed at how well these bigger transformers provide better body and an extremely realistic sound quality. I can no longer go back to the 7W units even though they are very high quality. The 20W are hand wound by the “old grandpas” and the 7w are machine wound. There is no comparison. The 20W, H-20-3.5U is magnificent with this amp. The 7W just doesn’t do it.
The PT-160 power transformer is perfectly sized for this amplifier and is ideal.
The C-10-130W choke is probably a better made and better sounding choke than what other manufacturers offer. There are manufacturers in Russia that will only use the Hashimoto chokes in their amplifiers. They consider them the best and who am I to question that? The choke and the power transformers are ideal together and both provide exactly the spec’s needed for this amp. In fact, this amplifier ends up putting out 4w/ch instead of 3.5w/ch than would be available with the 7W Hashimoto transformers and the same power transformer and choke.
Prior to using Hashimoto transformers, I was using custom wound Electraprint transformers. The Electraprint were too bass heavy for me. The Hashimoto are extremely well balanced. The frequency spectrum for the H-20-3.5U is an amazing: 25Hz to 90,000Hz (+-2db). That’s incredible!
Try to do that with a Hammond chassis or other typical audio component chassis!
The Aum Acoustics chassis is extremely robust, precision CNC milled down to the last 1000th of an inch, and provides a long-term value unlike the typical thin cheap chassis used for most audio components. The majority of components use thin cheap chassis material. They try to make the front plates look nice by over-engineering them but the overall chassis is still cheap, thin, and vibrates like crazy thus reducing overall audio quality.
Even my beautiful looking Auralic Vega has a cheap chassis hidden by an elegantly designed front plate. Unlike these typical vibration prone audio boxes, the Aum Acoustics 2A3 amplifier chassis is extremely well but and beautifully executed, every single millimeter! All side walls are 1/8″ reinforced ribbed aluminum.
A cross section of a side wall from this chassis is shown below. The height of this side wall extrusion is 2.25″ and there are two 1/2″ wide ribs integrated into the top and bottom portions of every sidewall. This makes this extrusion extremely strong and much better than even 1/8″ thickness typically provides on its own (without the ribs).
The Duelund CAST PIO copper capacitors simply break the mold in terms of what a PIO capacitor is expected to sound like. Gone is the top-end roll-off and bottom-end bloat, and the level of detail resolution is unheard of in a PIO. In fact, the sheer amount of clarity and detail resolution the Duelund brings to the table is quite uncanny, not “for a PIO” but for any capacitor. Thankfully, the special naturalness and ease of a good PIO remains intact. Actually even more so, since there are no strange frequency response aberrations or quirks in the sound to distract the listener. What’s even more interesting is how the harmonic richness and bloom is maintained while avoiding extraneous warmth, bloat, or opaqueness. The overtones and textures seem to be physical properties of the themselves, not a fuzzy, resonant ghost riding on top of the music like a halo, a problem often heard in very old PIO’s that were not manufactured to exacting standards or perhaps those that have not aged well.
The way the Duelund presents strings is especially something to cherish. It is very difficult to get the balance correct between the resin of the bow and cavity of the body, but here there is the perfect amount of friction and glide, which presents the right amount of glorious texture without undue etch, all resonating from a warm, palpable body of the instrument. Those who lament the death of glorious strings in the digital era may have hope left yet. Speaking of strings and classical music, this capacitor is capable of localizing and layering the various sections of the orchestra in lateral, vertical, as well as depth dimensions, leading to a very engaging experience.
The Duelund distinguishes itself from some of the state-of-the-art teflon capacitors by the very way the sonic background gives the impression of perfectly calm lake without a single ripple, a calmness that permeates through the space between instruments, which in turn holds and supports the instruments firmly in place in the soundstage. There has to be a correlation between this sense of serenity and calmness and the extreme mechanical stability and damping present in the construction of the CAST PIO capacitor. There is no trace of blurred images or inconsistent tonal balance that can happen when parts are allowed to resonate or ring, and reading about how these capacitors are constructed, one can begin to understand why.
The Duelund CAST Capacitor is a no holds barred flagship capacitor. The construction builds on the foundation of the Virtual Stack Foil design but does so with the addition of a proprietary WPIOdielectricum which necessitates several days of impregnation under very high pressure. This creates a form with incredible damping properties giving a mechanical stability hitherto unheard of. On top of this casting process, a ring of hard pressed paper strengthens the damping properties of the design helping music flow effortlessly.”
Duelund CAST PIO Copper CapacitorPerhaps aided by the mechanical integrity, this capacitor’s instrument separation and localization is top-notch, and while the presentation is beguilingly smooth and organic, when called for, there is plenty of metallic sheen and ring when such instruments are struck. In fact, this is one of the areas where CAST PIO seems to outdo most Teflon capacitors, which tend to lean a little more in the direction of metallic sheen and crackle versus the smooth and organic, to the point where sibilance and stridency can be a little more distracting, at least on less-than-perfect recordings.
There seems to be an interesting phenomenon that happens to the perception of time with the Duelund in place; time almost seems to slow down a fraction. This is not the same thing as if a musician makes an error and rhythm actually slows down; there just seems to be a minute fraction of a second more available between notes that allows more time for the listener to relax and sink into the music. The best teflon capacitors subjectively seem to present the “accepted” time flow with more distinct and harder-hitting transients and beats, but the effect is akin to a speed boat rocketting down the river energetically, splitting the water producing power waves. Duelund’s effect is more like a beautiful sail boat with strong wind in its sails, still propelling forward but with more of a graceful glide. There are certainly types of music that will favor one type of presentation over the other, but both approaches are quite satisfying in their own way.
Here’s a link to a review that compares the Dueland CAST PIO to two other top class capacitors, one from Audionote and the other the V-Cap CU:
I install a 12W Ohmite resistor on the “cooler” dropping resistor that is installed on the 6SN7 printed circuit board. A 2W resistor would be fine but the extra size of this resistor and the others I install provide more “air” and open-ness in the sound quality. Large resistors do make a positive difference even though the circuit specs could use smaller wattage sizes.
Every other resistor and capacitor has been chosen to contribute to sound quality and longevity. The excellent AmTrans metal oxide resistors are some of the best anywhere. The Takman Rey metal film is more than adequate for where they are positioned. The electrolytic capacitors are rated for 10,000 hours instead of the typical 2,000 wherever possible.
In total, everything counts including making the fewest solder connections possible. Even the Dueland Cast PIO CU film capacitors have leads in excess of 6″ in length and are able to reach the printed circuit boards without needing to solder to a solder terminal first. This eliminates two solder connections and provides the best possible connection and of course, a boost for sound quality.
Even the wire is the best I’ve used and I’ve spent years experimenting with all kinds of wire (solid platinum, solid gold, solid silver, solid copper, and stranded copper). Every single piece of wire has been chosen for its ability to improve sound quality.
I’m very pleased with all of the selections made and you would too if you spent the many hours doing the R&D and research that I’ve gone through. This amplifier has the absolute best sounding parts that I’ve been able to accumulate. I love the presentation of this amplifier and am stunned by its sonic beauty!
This time around, I decided to use the very thin silver internal connectors provided by the WBT 0710-Ag binding posts. After listening to them and through connecting my Furutech rhodium speaker cable spade connectors to them, I’ve fallen in love with these posts. These are the absolute best I’ve ever used and I like them better than my previous Furutech binding posts.
However, I still vastly prefer the Furutech FP-901 rhodium plated chassis jacks and of course, the Furutech FI-03 rhodium plated IEC module. This combination of rear panel connectors is the best I’ve used to date. Plus, they all look good too! What a bonus.
For this amplifier, I’ve chosen 18 gauge and 22 gauge Neotech teflon coated copper. Yes, I’ve used much more exotic wire than this, but this wire sounds the best where I use it.
Neotech Continuous Cast UP-OCC 99.99998% Pure Wire
I’ve chosen Jena Labs 22 gauge uniquely cryo’d copper wire for the RCA input wire in this amplifier. I tried all of my runs of exotic wire and this wire sounds the absolute best. It has the speed of solid silver but the warmth of copper without the copper bloating. This is, without a doubt, my favorite input wire for tube amps.
The final combination that I’ve chosen (Neotech and Jena Labs) allows this 2A3 amplifier to really sing and its voice is beautiful!
The chrome top plate screws are chrome plated and have hex nut heads. This is a very unique screw but looks better than any other alternative. I chose these screws for their cosmetic beauty.
All of the other screws were chosen for their combination of strength and beauty. All screws have lock nuts or lock washers and all thread tips have been sealed with a special clear sealer to help prevent any of the screws loosening over time. I’ve very pleased with all of the screws and connectors that were chosen. I do not believe I could have made a better selection.
The only way that I strip any wire in any of my projects is with my Teledyne TW-1 thermal stripper with thermostat control. This provides a much finer end product and I couldn’t go back to using a mechanical stripper if I tried. No need to anymore since I own my thermal stripper. In fact, this is my 2nd thermal stripper in the last 10 years. The first one was a used unit that eventually fell apart. I replaced it a year ago with a brand new unit.
I also use one of the best soldering stations, the Hako FX951.
Also, extremely important, is the solder that is used. Jena Labs has done substantial research and they’ve found that Cardas Quad Eutectic solder is the absolute best. That’s what I’ve used throughout this entire amplifier. Here’s Jena Labs comment from their website:
“WE DO NOT RECOMMEND CONVENTIONAL SILVER BEARING ‘AUDIO’ SOLDER BECAUSE OF AGGRESSIVE HIGH FREQUENCY DISTORTION, SOUNDSTAGE PINCHING, GENERALLY POOR MECHANICAL WORKING BEHAVIOR, AND POOR LONG TERM STABILITY. !!! WARNING: A solder joint made with ANY non-eutectic solder will fail in from 3 to 8 years !!!
Make sure that whatever solder you choose to use is guaranteed to be a Eutectic Alloy blend.
We are so convinced of the superior performance of Cardas Quad Eutectic Solder we include a usable amount of it in every wire order (approx 5 to 10 feet depending on wire order size… ”
CLICK BELOW FOR PHOTOS:
I have voiced this amplifier around a particular set of 5 tubes that in my opinion sound better than other alternatives:
One, rectifier tube – rare New Old Stock (NOS) Mullard GZ32 in red, white, and blue original Mullard box. This tube is the ultimate for this 2A3 amplifier and I will provide one with the purchase of this amp. (Thereafter, I will sell one of these NOS GZ32 tubes for $250 each. I maintain a stash of these rare tubes and will provide the additional $250 NOS GZ32 in the original Mullard red white and blue box)
A 2A3 matched output tube pair – New Stock, Emission Labs 2A3. I can provide a new matched pair if you desire. I have them in stock. My purchase price is $500 with the purchase of this amplifier. You can also purchase your own for $629 per matched pair (plus PayPal charge plus shipping cost) at tubeusa.com.
Two, CV181-TII (aka 6SN7) matched input tube pair – New Stock, Pavane CV181-TII ($290 per Grade A pair at Grant Fidelity) – This is the most balanced sounding 6SN7 tube. I own many NOS rare and expensive 6SN7 tubes and the Pavane sounds the absolute best for my Rethm single driver speaker systems. I can help you determine whether these would be best for your system or another choice. Once again, I can provide a new matched pair of Pavane CV181-Tii tubes if you like. I have them in stock.
I also stock rare NOS MELZ 1578 with the holes in the plates. These are some of the best 6SN7 type tubes available anywhere. They are slightly bright with my Rethm speakers and I prefer the CV-181TII. However, if your system can handle the extra high end extension, these would be “killer” tubes. I paid $283 per Amplitrex matched pair and will sell them at the same price to whoever purchases this amp.
Another rare tube that I stock are the NOS Sylvania 6SN7 “Bad Boys”. Once again, the selection of the pair of tubes in your 6SN7 position is really dependent upon the speakers you intend to use. All three of these types that I’m recommending have the best possible sound quality and these are the type of tubes that make this amplifier sing.
You can use other tubes if you desire but please consult with me before trying anything other than what I recommend above!