Massively overbuilt 45 headphone amp all noise no signal

So currently the plans are to have no, nada, none, not any, silicones running inside nor contained in tiny plastic packages, anywhere in this amp.

Thus far the LED's we used for the bias on the 6J5's have been yanked out in favor of using batteries.
After the next round of transformers go in, we will replace the HEXFREDs used for the diode bridge with a rectifier tube.

This also means no voltage regulators, of any type or sort are used in this amp.
The tubes filaments run on ac, and the B+ is 'free floating' and is directly tied to the ac mains line voltage, which in turn has been 'tamed' by using 2 ac mains transformers, the last one being a step down xfmr.

And the SQ has been most intriguing and enticing up and down the entire audio bandwidth.
And it's no where near dialed in and I'm using my 'development set" of tubes and not the vintage set.

And we are also now working on a preamp to drive this large, 10 transformer amp.
Only it's going to be usable as a secondary HP amp as well.
This will make a nifty meet rig as it will, in essence be, a mini-me version of the BIG amp.

So it's actually a HP/pre-amp combo designed to use the same set of tubes as the BIG irennn amp, but with enough design flexibility built in to use a variety of other DHT tubes.
This will come in handy as we figure which set of tubes work best in the pre-amp mode, as a compliment to the 6J5's/45's used in the power amp.


And it may just be we wind up using the same set of tubes in both, but that remains to be determined.

But for me the sure sign we are on the right track is SDSG (SuperDuperSuperGlue) has risen to new levels and the MitM (Magic in the Mids) is in full evidence, yet it has not fully blossomed.

There remains much tweaking and fussing yet to do.

JJ

 

I'm so confused.

 

confused about what?

JJ

 

After the last post describing the preamp, my thought was “massively over engineered” past the point of rediculous and about to pass the point of insanity. But hey, what do I know?

 

@johnjen would you mind making a glossary of the acronyms ? I think that's what's confusing Bazelio

 

I feel like batteries are a terrible bias reference. All batteries drop in voltage as they empty out and depending on which formulation, the drop can be extremely nonlinear.

 

Well I would but apparently the High Council has decided to continue to put words in my mouth so really there is no point.

Fin.

 

May have something to do with shit talking this site on Head-Fi recently... hmmmmmm...

 

Meh, I've found it entertaining and interesting.

 

It's like watching Hunter S. Thompson work with Oscar Acosta to build an amplifier.

 

He still has a ways to go to get past the point of insanity

 

Okay, what would you like to know?

 

This guy gets it.

 

Lets see if I can solve some of that. Fair warning, this is going to be a giant wall of text and I will probably need to read through it a few times for grammar and clarity. If there is some part that needs further clarification just let me know. I am pretty much an open book.

1) Planing the amp.

Before even starting down the path of building the amplifier, we needed to set some design goals. Now I have heard johns system already. I knew the general characteristics and sound signature of his system. I wanted to build something that sort of reflected the same sort of sound. Or in other words, I did not want something that was going to sound so completely different that we would risk disappointment if that makes sense.

John also described that he was very keen to have full frequency response (down to below 20hz), extremely quiet noise floor, and as low distortion as possible.

These might seem like simple and obvious requests, but these goals are non trivial accomplishments for a tube amp.

2) Coming up with a design

Given the previous criteria, and having heard johns system before, I knew I wanted to use the Type 45 tube as a power tube. Its perfect for headphones because 45 amps usually have a very wide, clear, and airy sound stage. The max power output is only about 2ish watts. So it just sort of works for our application.

However the 45 does not have low enough output impedance to be able to use it in an OTL amp like the schiit valhalla. So we have to use an output transformer. This is where things get tricky. Most output transformers will actually roll off frequency response at the low end. Most affordable and easily obtainable output transformers will actually have a -3db down point of around 25-30hz. This works really well for speakers because most speakers will roll off before the -3db down point of a transformer. However it sucks for headphones.

The parafeed design allows up to optimize the output transformer section which theoretically should give us a -3db down point of around 10hz.

The next step was to figure out a driver tube for the 45. I knew I wanted a single triode driving the 45 so that we could have a very pure 2 triode design. My mind went to the 6sn7 which is a very nice low distortion tube with a respectable amount of gain. The 6j5 is literally 1/2 of a 6sn7 and by splitting the tube apart, you get better channel separation and all that jazz.

The problem with the 6j5 was that the only way it would really work is if we really made sure the tube was achieving the maximum gain it possibly could. Uses a load and bias resistor will actually detract a bit from the maximum gain. Using a bias LED and plate chokes helps improve the overall efficiency. Again, if you want me to go into more detail, just ask.

Lastly the power supply is actually just a very basic power supply. We used a choke input for better power regulation, and secondary choke filters over resistors to keep the DCR as low as possible. Splitting up the power supply into a right and left channel simply helps with channel separation. The temco caps are nice cheap film caps. They have low ESR low ESL and quicker charge times compared to electrolytic capacitors but they have lower capacitance values. Hence why we have so many of the suckers in parallel.

3) implementing the design.

This is currently where we sit. We are making incremental improvements. Most of which are very basic like rewiring the amp for cleaner wire runs.

The modification that I think will need some explaining are the input and interstage transformers. John wanted to be able to use XLR connectors for this amp and wanted something that was sort of pseudo balanced. Part of the reason is because the dac he has has a higher voltage output from the XLR output and we honestly need all the signal voltage we can get.

Now in a normal tube circuit, the tube needs something called a grid leak resistor. For simplicity lets just say that without some sort of grid leak resistor or path to ground, the tube could potentially melt. Grid leak resistors are not ideal because they let some of your audio signal slip through to ground. Using a signal transformer still gives your tube a path to ground, but it has high AC resistance and very low DC resistance. So all the positive DC ions can flow straight to ground, but the AC audio signal will be less likely to go to ground and will ideally all end up at the next tube. There is a lot more to it, but hopefully that suffices.

Another side benefit is having the input and interstage transformer directly tied to the grid, we can also apply fixed bias voltage directly to tube girds through the secondary of the transformer via a battery. Batteries are very low noise and inherently stable since the tube grids draw next to no current what so ever. Plus fixed bias helps raise the efficiency of the circuit and will allow it to pump out a hair more power which is kind of nice.

------------------------------------------

Hopefully that gives you all some clarification or information about the amp.

Yeah, the amp truly is a big iron amp. The reason john keeps saying it in the way he does is because the thing is literally comical and hes trying to be funny about the fact that this thing is such ridiculous overkill for a headphone amp, that you can't help but chuckle at it.

I literally designed the thing (granted with a lot of inspiration and suggestions from others) and even I don't have anything this ridiculous for myself. Thats the level of insanity we currently are sitting at.

Now as far as the rest of johns vocabulary (leading edge dynamic impact and what not), I think he has write ups on all of that.

 

What casework are you using for that? I assume a two box amp. Sounds way too big & heavy for everything to be crammed inside one box?

 

Case work is sort of yet to be determined. Right now the amp is literally on a piece of sheet steel. Once we get the last of the hum issues solved, we will think about chassis design.

But a single chassis is not out of the question. This thing is heavy, but the biggest issue is that it is awkward to carry. If the weight were more balanced and the unit had handles, then its totally possible to carry it.

The other thing is now that we are working on this preamp/headphone amp, he will be able to use that to take to meets and it will serve the purpose of portability.

 

@Tristan Jones Thank you for your detailed and well thought out post. It's obvious you know what you're talking about! Lots of info like that is appreciated and enjoyable to read - please keep it up. Cheers.