SET Asylum

Single Ended Triodes (SETs), the ultimate tube lovers dream.

Return to SET Asylum


Message Sort: Post Order or Asylum Reverse Threaded

Pondering DHTs for headphone use

152.16.191.33

Posted on January 5, 2016 at 06:06:24
KG4NEL
Audiophile

Posts: 120
Location: Hillsborough, NC
Joined: April 19, 2005
Little about where I'm at in my audiophile journey. I'm currently headphones-only, as I don't have the space in my place for the Fostex corner horns and monoblocks that are in my storage unit :(

I have a pair of high-impedance Beyerdynamics and a Decware CSP2+ OTL amp (three 6922s, tube rectified). It sounds - decent - but I really miss triodes. I'm thinking about going to 2A3s or 45s in a more traditional output transformer-based amp, so I can gain the ability to drive low impedance planars with it too.

Are there any "gotchas" I should worry about in choosing an amp design for this? I assume I'd have to run DC for the filaments, as the hum would probably be too much with normal AC. Is there a way to get OPTs to like running both planars at 32 ohms and Beyers/Sennheisers at 250?

Thanks in advance!

Jim J.

 

Hide full thread outline!
    ...
RE: Pondering DHTs for headphone use, posted on January 5, 2016 at 10:20:31
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
DC on the filaments is a must have.

Beyond that, you want some flexibility with the output transformer, but more likely the ability to switch between 2, 4, and 8 Ohms. You do not want an output transformer with a really high impedance tap for driving headphones, as none of the headphones on the market are going to want 3500mW off a 100 Ohm tap (too much power, too much noise).

An 8 ohm secondary is likely desirable for both your 250 Ohm headphones and power hungry planars.

 

RE: Pondering DHTs for headphone use, posted on January 5, 2016 at 14:47:52
mwiebe
Audiophile

Posts: 294
Joined: November 3, 2001
Contributor
  Since:
January 5, 2006
Jim,
I mostly listen to DHTs and headphones (Beyer 600ohms) and Magnequest TL-404, a 2W rated autoformer, so you have to run a DC blocking cap. The TL-404 has taps at both speaker level (4,8,16,32) and also at line level (125,250,500) and presents a 5,000 ohm load end to end, so it's ideal for a 45. I use various plate chokes, usually a Magnequest EXO-010.

I'm a pathological tube roller so I rotate through 4P1L, 1LE3, 1N3S, 1J6G, 6P21S, 01A, 12A, 26, 30, 31 and subminis like 5676, 1P24B, etc. I use batteries and DC wall warts for filament duties.

Sowther has variety of multi-tap series and parallel feed transformers. On the cheaper end of the spectrum I've had good luck with Edcor GXSE5-600-15K, GXSE5-600-5K and Carnhill VTB2291.
Matt

 

RE: Pondering DHTs for headphone use, posted on January 6, 2016 at 14:41:20
Frihed89
Audiophile

Posts: 15703
Location: Copenhagen
Joined: March 21, 2005
I can't tell if your going to DIY this or buy a commercial headphone amplifier. I have a Donald North Audio Stratus (2A3). I love it. Very quiet. Very beautiful sound. It will take a wide range of headphone impedances and senitivities

 

RE: Pondering DHTs for headphone use, posted on January 6, 2016 at 20:59:11
needtubes
Audiophile

Posts: 66
Joined: December 11, 2011
Another nice directly heated tube that can be run entirely off of batteries is the 6418. It is a pentode, but looks pretty nice when triode strapped.

 

RE: Pondering DHTs for headphone use, posted on January 7, 2016 at 07:53:00
mwiebe
Audiophile

Posts: 294
Joined: November 3, 2001
Contributor
  Since:
January 5, 2006
Thank you for the suggestion. I know they are popular sub-minis but they don't sound as good in the applications I use as do 6088s or the Russians like 1P24B.
Matt

 

Edcor Transformer, posted on January 7, 2016 at 18:23:12
Triode_Kingdom
Audiophile

Posts: 10047
Location: Central Texas
Joined: September 24, 2006
"On the cheaper end of the spectrum I've had good luck with Edcor GXSE5-600-15K..."

I had a short discussion with someone here at one time regarding this transformer. Was that you? I have it in mind to use these to transform 5K to 200, rather 15K to 600. Using them at a lower operating Z should extend their response at both ends of the spectrum. The only caveat I can see is the need to reduce their in-circuit power in order to avoid early saturation at bass frequencies. Given the miniscule power requirements of phones, I don't think that's a problem. Does any of this ring a bell?

 

What I do..., posted on January 7, 2016 at 18:51:43
Triode_Kingdom
Audiophile

Posts: 10047
Location: Central Texas
Joined: September 24, 2006
I run 32 ohm phones from a SE 6V6 amp, roughly 4.5W/ch, using the 8 ohm output. To match the amplifier, I have a 9.1 ohm resistor in parallel with the output and a 150 ohm resistor in series with the phones (Grado). This creates a load of 8.7 ohms for the amplifier, and volume is sufficient to deafen me. Maximum available voltage at the phones is slightly more than 1V RMS, or 35 mW. Based on the manufacturer's sensitivity rating for my phones, that calculates to 115 dB SPL. It seems much higher in practice, but those are the numbers. If I were to wire 250 ohm phones directly across the output of the amplifier, the equivalent load with the parallel 9.1 ohm resistor would be 8.8 ohms. Maximum available voltage at the phones would be 6V RMS, or 144 mW. I'm pretty sure that will be well beyond the threshold of pain with any headphones you're likely to use.

 

RE: Edcor Transformer, posted on January 7, 2016 at 19:22:12
mwiebe
Audiophile

Posts: 294
Joined: November 3, 2001
Contributor
  Since:
January 5, 2006
I don't know if that was me or not. I've used both these transformers since Edcor released them. The GXSE5-600-5K has been in daily use for more than five years. They are still so cheap its easier to plumb them into a circuit and then have a go at making it better.
I've never felt they were weak at either frequency extreme. I use them for preamp duties not headphones, the GXSE5-600-15K I've used with many different phones. Neither are in the same league as TL-404s, but that is not expected.
Matt

 

RE: What I do..., posted on January 7, 2016 at 20:10:23
Paul Joppa
Industry Professional

Posts: 7296
Location: Seattle, WA
Joined: April 23, 2001
Some low-impedance phones will give lumpy bass with a 150 ohm series resistor (or 120 ohms if you want to implement the IEC/IHF standard). Grados among others are said to suffer this problem. Note that I am not a headphone guy myself - but we do have a lot of customers so we get plenty of feedback on what works.

A solution is to make the 8-ohm shunt a voltage divider - for example 5.6 ohms and 2.4 ohms. for more flexibility, 8-ohm L-pads are widely available.

 

About the filaments, posted on January 8, 2016 at 01:05:55
Bas Horneman
Audiophile

Posts: 4084
Joined: March 28, 2001
I would recommend Rod Coleman regulators.


Caveat Emptor: Both papers are essentially written by the chaps who sell the regs.
Tent labs paper

Rod Coleman on why filament regulators for DHT

 

RE: Edcor Transformer, posted on January 11, 2016 at 14:14:31
Triode_Kingdom
Audiophile

Posts: 10047
Location: Central Texas
Joined: September 24, 2006
OK, thanks for your comments on this. If I test the 600/15K for the application I have in mind, I'll post results.

 

RE: What I do..., posted on January 11, 2016 at 14:18:00
Triode_Kingdom
Audiophile

Posts: 10047
Location: Central Texas
Joined: September 24, 2006
I haven't noticed any ill effects from my method, but your suggestion is certainly a reasonable alternative. I might give it a try, just to see if I can hear a difference.

 

RE: What I do..., posted on March 17, 2016 at 08:12:25
horn kid
Audiophile

Posts: 128
Joined: November 2, 2014
The best speaker amps just beat the best headphone amps in my experience, and by a lot, just use a voltage divider but one step further than what is described: a parallel, series, parallel, then series and you can make the headphones look back and see 30 ohms (or whatever you want) and still have the amp see 8 ohms. Best of all worlds.

Go on the web and look up voltage divider calculator.

 

Page processed in 0.028 seconds.