Bottlehead Electrostatic Headphone Amp

Discussion in 'Headphone Amplifiers and Combo (DAC/Amp) Units' started by NationOfLaws, Feb 17, 2020.

  1. NationOfLaws

    NationOfLaws Friend

    Pyrate
    Joined:
    Nov 10, 2019
    Likes Received:
    874
    Trophy Points:
    93
    Location:
    US
    The folks over at Bottlehead have apparently teamed up with the guy at Nectar Sound to design a tube amp for electrostatic headphones. It sounds like they’re contemplating whether to sell it as a kit, as a finished product, or to shelve it altogether.

    Seems like it could be interesting at least.
     
  2. lagadu

    lagadu Facebook Friend

    Contributor
    Joined:
    Jun 11, 2019
    Likes Received:
    238
    Trophy Points:
    33
    Location:
    PT/DK
    The situation with the limited selection and information for ES amps is in large part what always put me off giving the higher end stuff a good chance at an audition because it puts me off so much. If this turns out to be good, maybe I'll finally try and audition the sr007 that I've been curious about for a long time.

    I'm definitely curious about this project.
     
  3. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,467
    Trophy Points:
    113
    Location:
    Winnipeg
    I think if they're trying to break into the DIY electrostat amp world then they'll have to make it past the church of Spritzer, for better or for worse.
     
  4. NationOfLaws

    NationOfLaws Friend

    Pyrate
    Joined:
    Nov 10, 2019
    Likes Received:
    874
    Trophy Points:
    93
    Location:
    US
    I don’t really think they’re looking for market share. Probably the type of person who is going to buy this amp is… uh… me: Bottlehead fans who want to f**k around with new ways to get electrocuted.
     
  5. dncnexus

    dncnexus Friend

    Pyrate
    Joined:
    Mar 12, 2018
    Likes Received:
    1,171
    Trophy Points:
    93
    Location:
    Virginia, USA
    Wanted to revitalize this thread with some information about this electrostatic amp. Here is a snippet from the manual

    "The Bottlehead Tube Energizer is a zero global feedback, class A tube headphone amplifier. It features both 580V Stax Pro Bias and 230V Stax Normal (legacy) bias jacks. There is a single input, a balance control, and level control.

    This circuit is rather unique as far as electrostatic headphone amplifiers are concerned, as it is a simple cascade of two differential amplifiers per channel, with the second differential amplifier directly coupled to the headphones themselves. No global feedback is used in this design, which provides a smooth, open sound. To further enhance this design, each differential stage uses our Bottlehead C4S constant current source. These provide a very high impedance under the cathodes of each differential amplifier, providing very tight AC balance in each stage. As a side benefit of using this technology, this amplifier is a tube roller's dream!

    Quality polypropylene capacitors are used in the signal path, long life high temperature electrolytic capacitors are used in the power supply, and a very generously over-sized power transformer was chosen to ensure a long lifetime. The headphone output jack is custom made and very gentle with both stock Stax plugs and the Moon Audio 5 pin plug. There jack pin contacts are recessed for added safety when plugging and unplugging headphones.

    The bias supply for this amplifier is of the Cockroft Walton voltage multiplier type with additional series resistance between each output of the supply and the headphones themselves. This supply is unregulated, as it is designed as an ultra low current power supply for safe headphone operation. A 12AL5 diode sits at the bottom of the voltage multiplier and slowly turns on as the amplifier warms up.

    The high voltage power supply in this amplifier is a bridge rectified split rail type. This provides a positive and a negative voltage rail that allow for an enormous swing at the output stage without voltages that are any higher than what you would find in a 300B amplifier. The positive side of the high voltage rail is soft started by a 6CG3 damper diode that provides a controlled application of the positive rail voltage to the amplifier to further reduce undesirable behaviors during startup.

    There is an additional small 12V DC power supply to heat the 12AL5 and the first gain stage tubes to maintain proper heater cathode voltages."


    From my email discussion with Paul Birkeland from Bottlehead who built this amplifier,

    "For the tube complement, the amplifier uses a quad of 5965s for the actual amplifying duties as well as one 12AL5 and one 6CG3 to soft start the bias supply and the positive half of the power supply. For a little bit of extra high frequency bandwidth, the rear pair of 5965s can be exchanged for 12BH7s. The circuit itself is a cascaded pair of differential amplifiers, so that is not particularly in line with the circuits we normally use, but is a zero feedback circuit and it takes advantage of our C4S active loads, so it's still very much closely related to our other products."

    Build time on this headphone is about 3-6 weeks, with mine taking right about 5 weeks from order to receiving. The price of the amp starts at $1500 which includes powder coating the chassis with textured, vein, or wrinkle powders. The amp has an upgraded option for the PEC pot upgrade for the level and balance controls for $100, and flexibility with the coupling caps for $50. There is also options for a wood side panels, although I opted for full metal chassis.

    For the internal look at the wiring here is the inside provided by Paul as he was building the amp.
    1596949324410.jpg

    As for impressions, I found this amp to be neutral, with a slight wetness and tube bloom. The mid frequency is seductive and have the tubeness. The soundstage and 3D effect is introduced with this amp. I would imagine this is somewhat similar to what a higher end tube amp for dynamics would be like. I have no experience with DNA or EC yet, but once I get my Starlett in I can see how it compares in tonality and sound signature.

    ***On a side note, I found that the 12BH7s that I received seem to be much weaker than the 5965s for the rear pair. I am not sure if this is due to the tubes being bad, but I found my gain was dramatically reduced when using those tubes. I am in the process of ordering new 12BH7s to verify, but the sound is much richer and fuller with 4 5965s in the amp. I will update when I get new tubes in to see if its consistent.

    When receiving the amp, I found many of the screws to be loose so I had to tighten them up, not a big deal, probably a result of the shipping. I do feel that this amp has improved over time, I guess that is what burn in is, but I feel it has really blossomed and opened up more now that I have put more hours on it. I will eventually attempt to roll more tubes in, but as of right now I am happy with this amp and think it will serve as a really solid corner stone for my electrostatic journey.

    If anyone has any questions let me know and I will try to answer anything anyone is curious about.

    Edit: *** Paul reached out to me and clarified that the 12BH7 does have lower gain than the 5965. The 12BH7 provides lower output impedance which should flatten out the treble response, while the 5965 provides more gain and will allow the amp to play more loudly. So please disregard my above statement!
     
    Last edited: Sep 14, 2020
  6. songmic

    songmic Gear cycler East Asia edition

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    2,102
    Trophy Points:
    93
    Location:
    Seoul, South Korea
    I don't know if I'm biased, but when I heard "Bottlehead Tube Energizer" I thought it was not really an electrostatic headphone amp but rather something like the Woo Electrostatic Energizer which needs to be plugged into a speaker amp to operate. Yes, I'm aware all electrostatic headphone amps have an energizer built in.

    https://wooaudio.com/amplifiers/wee

    But looking at the photo I see what appears to be a pair of RCA inputs. So I suppose it's a full-fledged electrostatic headphone amp?
     
  7. dncnexus

    dncnexus Friend

    Pyrate
    Joined:
    Mar 12, 2018
    Likes Received:
    1,171
    Trophy Points:
    93
    Location:
    Virginia, USA
    Correct its a full-fledged electrostatic headphone amp. The language is quite confusing, and since I am not super well versed in electrostatic gear I just assumed they were all used interchangeably, at least thats what it seemed to me at first since I agree it is confusing how everyone refers to this differently, even the makers alternate between energizer and headphone amp.
     
  8. scblock

    scblock Friend

    Pyrate
    Joined:
    Nov 11, 2019
    Likes Received:
    1,419
    Trophy Points:
    93
    Location:
    Denver
    For anyone considering this amp it’s been officially announced and is available to order now. $1,499 assembled, built to order.
     
  9. NationOfLaws

    NationOfLaws Friend

    Pyrate
    Joined:
    Nov 10, 2019
    Likes Received:
    874
    Trophy Points:
    93
    Location:
    US
    I wish I had electrostatic headphones so I could justify buying one. Alternately I’m glad I don’t have electrostatic headphones so I don’t have to buy one.
     
  10. JimL

    JimL Tongues KG's hairy starfish for fun

    Joined:
    Nov 21, 2016
    Likes Received:
    191
    Trophy Points:
    33
    Location:
    Albuquerque, NM
    Paul Birkelund must not have done much of a Google search if he thinks this is a unique circuit. There are only so many ways to arrange two tubes to get enough gain for an electrostatic headphone amp. Here is the TubeCAD stat amp circuit from 1999 - two cascaded diff amps, cap coupled with direct drive output. If you stick a 5965 in for the 12AX7 input tube, and substitute constant current sources for the cathode resistors, voila! the Bottlehead amp, or something very close to it.

    http://www.tubecad.com/november99/page8.html

    And since the 5965 has about three times the voltage gain of a 12BH7, it should sound about twice as loud for the same input signal in a no negative feedback circuit.

    A question re: safety resistors. On all modern Stax amps, the bias supply has a 5 megohm (4.7 or 5.1) safety resistor after the last cap, and 5.1k safety resistors on all the outputs. Does this amp have them? If not, I would put them in - easy to do with a DIY project.
     
    Last edited: Sep 17, 2020
  11. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    My eyes are bad, too lazy to fully trace, forgot most of the resistor color codes, and the color is off in the picture; but at glance, there looks to be x2 3.5k-ohm resistors (trace wires from STAX jack to upper left set of resistors) and physically small high M-ohm resistors on the bias (trace wires leading to the voltage multiplier circuit in lower left)? Could these be what you are talking about? I could be wrong.

    If you are going to question as if you were a SME, at least make an effort since there actually a photo of insides - don't parrot others via KG / Spritzer alarmist shit. Again, I could be totally wrong. Circuits isn't my primary expertise; but heck, if I can sort of figure it out, than perhaps people who intelligently question can also do the same.

    Another way would be current limiting somewhere else in the circuit, but I don't think this is done here.

    These claims are always somewhat familiar. I've heard the same, but in reverse from KG - vague implications that others copied his ideas. The fact is that these audio circuits are ancient tech. None of this is new. The only stuff I've read about that is new is found on tubecad, and even then, most of those ideas are half-baked. Who gives a crap what Paul claims or doesn't claim? It's entirely possible Paul may not of known. The amp designers I know don't tend to search Google all day.
     
    Last edited: Sep 17, 2020
  12. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Look closer. They are speaker terminals. It's an energizer / transformer.
     
    Last edited: Sep 17, 2020
  13. songmic

    songmic Gear cycler East Asia edition

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    2,102
    Trophy Points:
    93
    Location:
    Seoul, South Korea
    I looked closer but can't see anything that resembles speaker terminals. The stereo inputs I mentioned are the ones on the upper left corner of the photo.
     
  14. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Sorry, confused. Wee is the energizer. The BH is actually an amp, despite it also being called an energizer. I guess we can say it's an energizer in that the amp also requires a circuit to charge the diaphragm.
     
  15. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Do add, if Paul "copied" the tubecad design, did some substitutions, and "voila", then wouldn't he have also incorporated these resistors as indicated in the tubecad schematic?
     
    Last edited: Sep 17, 2020
  16. JimL

    JimL Tongues KG's hairy starfish for fun

    Joined:
    Nov 21, 2016
    Likes Received:
    191
    Trophy Points:
    33
    Location:
    Albuquerque, NM
    Sorry that you disliked my post. Those who know me know that I generally try not to upset anybody, so I apologize. I have friends who have built Bottlehead amps and are very happy with them, and the ones I have listened to were very nice. I am not, as you seem to think, a KG/spritzer fanboy - there are some things of theirs I like and some that I am less fond of.

    However, I was going directly off the description that Birkelund gave above: "...it is a simple cascade of two differential amplifiers per channel, with the second differential amplifier directly coupled to the headphones themselves." That is exactly the description of the TubeCAD circuit with the exception that the Bottlehead amp uses their C4S current source in the cathode, which I will happily point out, should be an improvement. But it is not a unique circuit, sorry. Alan Wright recommended substituting current sources for differential amp cathode resistors in his PreAmp Cookbook a number of years ago. Nor am I saying, or implying, that Paul copied the TubeCAD circuit. It is quite possible that he came up with it independently, as the topology is very common. I think you may be reading more into my comment than I meant to say.

    As to the question of the safety resistors - it was just that - a question. If the answer is affirmative then that's fine. Now that I look more closely at the photo it appears that the bias resistor reads 5004, which I think means 5 meg but I could be wrong - if that is so, great. I do not see resistors going to amp output pins but they may be under the C4S boards.

    Finally, as to the output safety resistors, with tube outputs I am not certain that they are as necessary as with transistor outputs, as a tube shorting and putting the high voltage supply directly into the output must be very rare. Stax does use them, however, even in their tube output amps, and with a DIY amp it is trivial to add them if desired, which was really my point.
     
    Last edited: Sep 17, 2020
  17. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I just think how it's funny folks from HC seem to be programmed to automatically question every estat amp that isn't a design by KG: safety issues, electrocution, design expertise, competence, credentials, sound quality, etc.

    You may not be aware of it, but this modus operadi is painfully obvious to the readers here, or at least the optics seem that way (and you know what they say, perception is reality). BTW, I didn't put that custom tag under your name - some other admin or mod did - probably years ago.
     
    Last edited: Sep 17, 2020
  18. NationOfLaws

    NationOfLaws Friend

    Pyrate
    Joined:
    Nov 10, 2019
    Likes Received:
    874
    Trophy Points:
    93
    Location:
    US
    Imagine being mad on the Internet about a marketing superlative like “unique”. Better head over to that diner in town that says they have the world’s best coffee and give them what for
     
  19. HAL9000

    HAL9000 Almost "Made"

    Contributor
    Joined:
    Apr 4, 2016
    Likes Received:
    233
    Trophy Points:
    63
    Location:
    Fairfield, IA
    Issues of provenance aside, I am really enjoying listening to the Bottlehead Energizer with the Nectar Hive headphones! D2 nervosa was creeping into the experience, so I swapped 7062s for the stock 5965 input tubes. That lifted a veil on the magic electrostatic airiness -- sooo nice! FYI, I am running 12BH7s in the output stage...
     
    Last edited: Sep 18, 2020
  20. paulbottlehead

    paulbottlehead New

    Joined:
    Sep 18, 2020
    Likes Received:
    15
    Trophy Points:
    3
    Location:
    Seattle
    I went ahead and registered so I didn't have to keep proxy answering through Hal.

    Yes, I have a 5M resistor between the bias supply outputs and the 5 pin and 6 pin Stax jacks. Even with out, you can grab the bias supply and it feels like a mild novelty bug zapper. A safe bias supply was very important!

    I was definitely aware of the directly coupled differential tube output of both the Stax SRM-T1 and the TubeCad circuit when I cooked up this design. This is part of the "R" in "R&D", it's asking and answering the questions: what is everyone else doing? What do I like about what they are doing? What would I do differently? What are people complaining about? The SRM-T1 was part of my inspiration to recommend the 12BH7 as an alternate output tube, as it gravitates more closely to the 6CG7, but the solid state front end and global feedback were things I could do without. As far as Broskie's writeup goes, it was helpful in quantifying what one actually needs out of a stat amp, but I wouldn't ever voluntarily put a 12AX7 into anything (the Miller capacitance of a 12AX7 is really high, and having it drive the Miller capacitance of the output tubes didn't seem like a great idea either). I ended up using more positive rail voltage than either design, which both use about the same voltages. Actually, they both use nearly identical output stages other than the tube choice...

    Functionally, the Stax SRM-T1 and the TubeCAD design are both cascaded differential amps, so it seemed like a starting point. Our amp integrates C4S active loads which allow for an incredibly long list of tubes that can be rolled into the circuit since the bias voltage can wander to be whatever it needs to be, but the plate voltage stay relatively fixed. They also give nearly perfect AC balance in the differential amp, and gets the directly coupled outputs very close to 0V on the output jacks.

    As far as the 5.1K resistors go, I cannot see how that would improve the safety of the amp, but it sure would screw up the high frequency response! The outputs of the amp sit very close to 0V DC, and when the amp is first powered up, there's an incredible amount of series resistance between the positive rail and the amp outputs (over 500K) to limit current flow to those headphone jack pins. Another 5K isn't going to matter there!

    There were many issues I ran into that had to be worked out to actually put this out as a finished product, and those are not covered in the TubeCAD writeup and handled very differently in the Stax amp.
     

Share This Page