Crack tube adapters and high-end OTL question

[adamct]

My noobie questions continue...I apologize for hogging the boards lately, and I also apologize for starting a new thread each time, but I figure that makes it easier for others to find answers to the same questions later.

So, with that out of the way, I have a couple of questions...a few practical, and one of curiosity. First the practical:

1. I have a two sets of adapters from my Woo Audio WA6. The first adapter lets you run 6SN7 tube in a 6DE7 socket. The second adapter lets you run 6GL7/6EM7 tubes in a 6DE7 socket. Can I use these adapters to run 6SN7s and 6GL7/6EM7s with my Crack+Speedball with the stock 9-pin socket and wiring?

2. I have 6AS7 and 7236 tubes from my WA22, where they are interchangeable with 6080s. Can I use these in my Crack+Speedball with the stock octal socket and wiring?

3. How do I figure out the answer to these kinds of questions on my own in the future, so I don't waste your time?

Now out of curiosity: As you can gather from the above, I have a WA6 and a WA22. I think I prefer the Crack to the WA6, and I consider the Crack to be roughly equal to the WA22. Quite apart from the price difference between the two amps themselves, please note that the WA22 is running tubes that cost $500-600, just by themselves. There are things that favor the WA22, but also elements that favor the Crack+Speedball. Both the WA6 and the WA22 are transformer-coupled. I'm sure the great sound of the Crack isn't just due to the fact that it is OTL (a good design and good parts obviously matter, too). But given that I assume the Crack was designed with a rough price range and a target level of assembly difficulty  in mind, I wonder what compromises were made. Has there has ever been any consideration given to creating an even better OTL design/kit (albeit at a higher price)? Would having a greater budget or creating a more difficult to assemble kit permit greater flexibility in design, allow you to use better parts, or...just be a waste of money? Is it possible to SUBSTANTIALY improve upon the performance of the Crack+Speedball by using a different design, or is the design already pretty close to some theoretical ideal when considered in terms of real-world performance? (To clarify what I mean by that, you could have a design that was was radically different, more complex, used more tubes, and was theoretically perfect from an engineering perspective, but which only improved upon the performance by, say, 3-5%. Alternatively, you could have a design that was fairly similar to that of the Crack, but which improved upon the Crack's performance by, say, 20%.). Or would bigger gains be had by using more expensive parts in the Crack, in which case the best idea would be to just keep upgrading the Crack? In short: what would be needed to create a substantially better OTL amp?

Any advice is greatly appreciated!

Best regards,
Adam

[Paul Birkeland]

1. I have a two sets of adapters from my Woo Audio WA6. The first adapter lets you run 6SN7 tube in a 6DE7 socket. The second adapter lets you run 6GL7/6EM7 tubes in a 6DE7 socket. Can I use these adapters to run 6SN7s and 6GL7/6EM7s with my Crack+Speedball with the stock 9-pin socket and wiring?

2. I have 6AS7 and 7236 tubes from my WA22, where they are interchangeable with 6080s. Can I use these in my Crack+Speedball with the stock octal socket and wiring?

3. How do I figure out the answer to these kinds of questions on my own in the future, so I don't waste your time?

1.  No, that adapter won't work, it will actually not function at all in the Crack socket.
2.  Yes, you can use those tubes.
3.  Fly out to Washington and take some classes   This is what the forum is for, don't sweat it.

Hmm, creating a better OTL amp, yeah, it's possible.  One way to drastically increase performance would be to shunt regulate the high voltage supply.  This would require two power transformers and two 6080's (or maybe a 6080 and some other tubes to serve as a shunt regulator).  This would also require a rather large chassis plate.  We could also throw in the attenuator design of the Bee Pre, and use film coupling capacitors, etc. etc.

The cost for a kit of this nature would be well over $1000, it's much more than twice the parts of a stock Crack, and we still haven't addressed lowering the output impedance. 

I am currently doing some experiments with some different triodes other than the 6080, mostly to look at the impact of the local feedback in the output stage while attempting to hold all other variables constant.

I also have a chassis that has a differential Crack built onto it (from 2008, when I was originally messing around with the circuit), and this would be a nice test bed for building a shunt regulated Crack at some point.

I guess the take home message from all this is that yes, we can improve the circuit, but at this point there are still a bunch of experiments to do so we know what is worthwhile and what is not.  I'm not aware of any other tube headphone amplifier manufacturer that would offer anything nearly this complex. 

-PB

[earwaxxer]

These are great questions that we all wonder about, and I am not qualified to answer, except for some basic observations that have helped guide me in this, at times, insane hobby. First and for-most, we need to focus on what we want to 'improve' on, in the sound that we have now. In other words, fix what is broken, and dont fix it if its not broken.

In terms of circuitry of a given component there will be benefits gained from using 'X' technique, ex. feedback, at the expense of other aspects of sound. I'm not sure that this approach will get you the most bang for your buck in solving your sonic dilemmas. Bob Carver, for example,  has been in the business of tube amp design for decades. He is still designing 'new' schematics, or a rehash of old ideas from a different perspective. Will his new amps 'solve' our problems. Probably not. They may help, but just as likely may not.

In summary - Bottleheaders are tweakers by nature, but there comes a point when we need to look at the bigger picture. Sometimes it means a shuffling of the component deck. Some things may need to go, and that may be the best decision on how to move forward....  (Ok Greg - I know what you are thinking - I'm actually fairly lucid, and I am not wedged between the toilet!)

[adamct]

Paul,

I assume you are saying that neither adapter will work with the Crack, is that right (there are two different kinds)?

And not that it matters, but I'm not interested in lowering output impedence...in my experience, if you are using low impedance cans, solid state is better. I love the Crack with my Sennheiser HD800s, HD600s and Beyer DT990s (600 Ohm) and that is all I would be interested in using with this "improved" amp. So we can keep the price of your (theoretical) kit to only "well over $1,000", no need to jump to "many thousands of dollars", LOL...

Eric,

You make excellent points. My problem, of course, is that I don't know what components have what effect on sound, so I'm flying completely blind. And to be clear, I don't really have any complaints about the Crack sound, other than the noise floor being higher than I would like, even with the Speedball installed. I wouldn't mind if the gain were lower, and I never object to more clarity, but that's about it in terms of wants. I sometimes find treble a little piercing (high drum strikes), but it is tough to point the finger at the Crack for that...could just as easily be my HD800s or the recording.

So at this stage, I'm just casting about for upgrades that are generally considered to be improvements. As I start swapping out components, I figure I will learn what effects they have on the sound. But at some point I need to start somewhere (not that I'm in any rush, since I'm very happy with the Crack as-is, and I haven't even done any tube rolling yet).

Best regards,
Adam

[Paul Birkeland]

Yes, both adapters won't work.

Can you describe the nature of the noise you're referring to?  We haven't noticed anything in any of our tests.

One possible way to mitigate this might be to remove the black jumper shown on the bottom of page 19 (PT-4 to 22L), and replace it with a pair of 100 ohm 3W resistors from PT-4 to 22L and PT-5 to 22U.

[earwaxxer]

Eric,

You make excellent points. My problem, of course, is that I don't know what components have what effect on sound, so I'm flying completely blind. And to be clear, I don't really have any complaints about the Crack sound, other than the noise floor being higher than I would like, even with the Speedball installed. I wouldn't mind if the gain were lower, and I never object to more clarity, but that's about it in terms of wants. I sometimes find treble a little piercing (high drum strikes), but it is tough to point the finger at the Crack for that...could just as easily be my HD800s or the recording.

So at this stage, I'm just casting about for upgrades that are generally considered to be improvements. As I start swapping out components, I figure I will learn what effects they have on the sound. But at some point I need to start somewhere (not that I'm in any rush, since I'm very happy with the Crack as-is, and I haven't even done any tube rolling yet).

Best regards,
Adam

Now I get it a little better - I dont know the Crack circuit, but some boutique caps/resistors in the right spots could improve those symptoms. Also volume pot. I'm sure others can chime in on that. From my experience with the Quickie, what you describe as 'noise floor' was improved with high end parts. Not sure how much that would cost for the Crack, but I spent several hundred bucks on the Quickie. Specifically, I noticed a 'maturing' of the sound, opening up between instruments, bigger and better depth. A 'fleshing' out if you will. Tube rolling was also very noticeable. There is much you can do to 'tweak' those problems.

good luck my friend - god speed - piece, love, tubes - Eric

[Doc B.]

Try different tubes if there is a perceptible noise floor. HD800s have a reputation for that treble. I don't find it piercing, but it is definitely a bit brighter than HD600s or 650s.

[adamct]

Which tubes have a greater impact on the sound, the 12AU7s or the 6080/6AS7s? I assume the latter?

Best regards,
Adam

[Jim R.]

Hey Adam,

Not sure where you stand on the whole cable thing, but replacing my stock hd-800 cable with a Q-Audio silk headphone cable turned the hd-800s into my absolute, all-time and forever fave cans. :-).  Also, the Crack circuit is so simple, that upstream components (dac/phono stage) will show all their warts, especially with the 800s.  The good news is that there are now some really incredible dacs for not a lot of money.

HTH,

Jim

[adamct]

Jim,

Here is a picture of my HD800 cable: http://www.bottlehead.com/smf/index.php?action=post;topic=3702.0;num_replies=8

And here is a picture of my 8-wire HE-6 cable:  http://jeaudiodesigns.wix.com/jeaudiodesigns#!HE-6-Blue-and-Purple-8-wire/zoom/c13ay/imageyxg

One of my HD600 cables is pictured here: http://jeaudiodesigns.wix.com/jeaudiodesigns#!Sennheiser-Walnut-and-Oak/zoom/c13ay/imageg9a

All of these cables were made by Justin from JE Audio Designs. I can't recommend him highly enough.

The HD800s are already my favorite headphones... 

As for my DAC, I use an Audio-Gd Ref 5.32. I'm very happy with it. I actually think that the irritating drum strikes I mentioned above were just part of the recording I've been listening to lately.

Best regards,
Adam

[Jim R.]

Adam,

Well, thanks for the pics, but I'm totally blind so they don't do me much good :-).

Yes, sounds like a really nice setup there and really improvements you may make from here on out are probably going to be incremental, but hey, that's part of the fun too.

-- Jim

[Paul Joppa]

In terms of a "better Crack" for high impedance phones, I think the following are the biggest potential improvements:

1) Speedball!

2) Better output capacitor

3) Shunt regulation

I'd put direct output (not cathode follower) as #4, but it's quite impractical. Something like a 6C33 running 500mA plate current would be needed. You could grill steaks on it!

I'd put a DHT output as #5, but there aren't any that have enough transconductance to compete with the 6080 for output impedance. And we have not been satisfied with paralleled tubes to get more transconductance.

Incidentally, the BeePre has a ~600 ohm output impedance. I don't know whether that is too much for the best sound from 600 ohm phones, but I hope someone tries it and posts... The BeePre has shunt regulation and current source loads, plus a directly-heated cathode and no cathode follower.

[Paul Birkeland]

Hey PJ,

We should talk more about a 500mA shunt regulated 6C33 amp!

I have some plate transformers that can deliver the current, and at least 10 6C33 tubes hanging around...

-PB

[adamct]

Paul,

The BeePre looks like a pretty cool kit! Since I don't need a pre-amp, I wouldn't be willing to lay out >$1,000 only to find out that it doesn't work, but I would definitely be interested in building/modifying one if there was any way to be assured that it would work, or if someone else were able/willing to test it.

Ideally, though, I would want to get the impedance down a bit. While my DT990s are 600 Ohm, my most frequently used cans are 300 Ohm Sennheisers (HD800 and HD600).

Any chance you want to develop a headphone version of the BeePre? You could call it the Marie Antoinette!*

Best regards,
Adam

* Get it? The "BeeHead"! 

[adamct]

Also, I should have mentioned that I've already installed the Speedball... I have to admit, that was kind of fun to assemble. Actually, the whole thing was fun. I found assembling the kit was exceedingly relaxing...

What is a good source of information for learning the very, very basic basics of this stuff? Right now I'm just following instructions (solder 3" black wire to Terminal A), but don't really understand what I'm doing. Imagine a not-very-bright 4th grade student wanted to learn about this, what would you have him/her read? (OK, maybe slightly above that level...)