State of the art DIY TUBE amp, any idea???

Look at the Woo6 modding thread, basically there are some real compromises made with that amp and yet people know it competes with a ZD. I can't help but wonder what a 6EW7 amp like this built to specs from the datasheets would sound like.

For instance a choke input powersupply for the rectifier with enough filtering that it doesn't require a $250 rectifier is the only solution.

Also start with enough voltage that the gain stage can be run at the datasheet recommended 250V rather than 70V's, with a choke on the plate. Finally the output section 150V &45mA's again per the datasheet for best performance. DC coupling is debateable as for the benefits. Think of the first section of the tube as a 12AX7, you wouldn't want to hear one of those at 70V on the plate would you?


Yet given all this, the Woo6SE is very very highly acclaimed, I think there is a lot of potential in an optimized 6EW7 SET design. I've been kicking around somethings, but nothing concrete yet. Whats holding me back is the chassis, it would have to be two big cases to do it "right."


Another option is DHT, 5842 driving a 71A, issue here is that DHTs and headphones don't play well. Most say the only solution are the expensive tentlabs DHT supplies. But I think 5 x 10m45s D-Mofsets in parralell may be a cheaper alternative?

I've been investigating the same as you, there are so many good options and the best option depends on the intended headphones, budget, time, and space.

If you read most of the meet reports, you'll find a lot of folks say so and so's DIY amp was the best at the show, you just need to have the money, time, and ability to take risks.

I would love to see a group project here like used to be done on Headwize, yet more geared toward state of the art as you are asking.

I too would love to see a "state of the art" group project here. Problem is that you're talking about a pool of maybe 20 people that are potential participants. I doubt if you could get even 2 of them to agree on exactly what "state of the art" is. This diversity is actually benificial to the community, but it pretty much means we're all going our own seperate ways.

FrankCooter FTW

I am flattered, but, that is the work of Ferrari, a really talented builder.

The real question before answering the SOA for amps would be what is SOA for headphones? Dynamic (hd800, etc), Electrostats or ortho? That is where the first divergence starts, never mind source, which will never get the same answer.

Then we must fork at pure Tube / Hybrid, where, again, little agreement will be had, and, sadly, the issue of SOA with 50 year old tubes is humorous, and SS audiofile parts become rarer by the day. Technology is not helping the cause.

Let's remember the Williamson tube circuit (circa 1937), is now over 70 years old.

Nice topic and idea, just no clear answers are even possible. If so, there would be no diversity in DIY at all.

To your credit, you are at least asking the question in the right place.

Anectdotally, I'll add the best tube amps I have ever heard was (one specific) electrostat Singlepower ES-1 (with some really pricey telefunken tubes, and before it developed 'problems') and the TTVJ 307a. I have not heard every amp available, and no one has, but it sounds like CanJam would be a requisite place to start such a journey.

Let me know when you come to a consensus on that.

Egads, you guys are aimin' to start a riot.

I think there would be several questions to ask immediately with respect to what gives the best SQ:

1. OTL or transformer O/P?
2. If the choice is OTL then OCL or not?
3. O/P stage topology. There are precious few good O/P stages that can be used effectively to drive low Z headphones directly.

After these questions are answered I bet you guys could agree easily on the rest.

I find it humorous that most of the DHT's, pre-WWII, have never been bettered (by either silicon or the later tube designs), regardless of the onward march of what we call technological progress. Amazing SQ, linear and low distortion without a shed-load of negative feedback! How I wish neg feedback had never been discovered/invented. Combined with the race for watts it resulted in a SQ nosedive.

Yes, because NFB became a simple means of covering really bad commercial designs. But I agree. There is little new under the sun, just a re-purposing of many things that were invented before even I was born.

Although we can maybe agree that the SS world has added some nice flavor and capability into the tube world that previously was not available such as easy dual rail supplies, easy low voltage bias supplies, easy CCSs (SQ affect up for debate), easy detection circuits, easy regulated DC for heaters, etc.

We are in complete agreement - my only objection was 'state of the art', meaning, IMHO, some new components to stop the devolution of circuit design.

So, I nominate Raven for such a design, only because I am not privvy to Stacker II (well, and I don't like 6sn7). The 307a (balanced) was just too costly in the end (and I couldn't build it), and the designs by both Parafeed and dsavitsk look far more interesting. FrankCooter is tempting me with the mercury vapor rectifier tubes, as is this other 6c30(?) design that keeps cropping up.

But, there will be no consensus in this rarified air, and more likely to be simply a matter of taste and texture, ie your willingness to chase a flavor of glass, and, well, iron (amorphous cores, silver wound,, yada).

Then there is the little matter of attenuation - ALPS RK50, TKD P65CS, Seiden, Shallco, Penny & Giles - not to mention signal wire, connectors and casing. A stupendous amp needs a proper presentation.

Riot, indeed.

To be fair, the "blame" has to be put on the companies designing the amplifying devices, driven by the requirement from customers for more power. What changed was that instead of inherently designing the components for maximum linearity and min distortion, other priorities dictated design, with negative feedback becoming necessary to obtain the low distortion that could be obtained from the earlier devices without using it.

Sure, I'm not against solid state or feedback per se. I just find it amusing that 80 years later we need to use negative feedback as a tool to get back to the low distortion produced by components designed 80-90 years ago. That is humorous.