I am casting about for ideas to build a DHT headphone amp for my current-hungry Audeze LCD-2 planar headphones. I like the sound of the 45 very much and I was thinking about that, but then someone suggested the 46 tube (more gain, cheaper) in Jack Elano's DRD circuit driven by something like the C3G or D3A pentodes with DC heaters. I am worried about the hum level, even with DC heaters, in a direct coupled circuit, but have no actual experience to base that on, just discussions with DHT-"users".
Is this worth following up on? The transformers to even breadboard it would be quite expensive and I am also told that it takes a lot of experience building DC DHT amps to get the sound perfect (even though the circuits are all deceptively simple on paper), so you may actually have to do the final build before you can be sure what it sounds like.
Does anyone know if the "old" Moth Audio 2A3 headphone amp used this circuit? I hear it does sound outstanding.
Is this worth following up on? The transformers to even breadboard it would be quite expensive and I am also told that it takes a lot of experience building DC DHT amps to get the sound perfect (even though the circuits are all deceptively simple on paper), so you may actually have to do the final build before you can be sure what it sounds like.
Does anyone know if the "old" Moth Audio 2A3 headphone amp used this circuit? I hear it does sound outstanding.
My suggestion;
27 / 6C45P mu-follower direct coupled to 71A. Use Coleman type heater supply, and there will be no hum. Also tube regulated PS (I have 2 6V6GT's as pass tubes and 6SH7 as an error amplifier). OT is the lowly Hammond 125DSE, but with only 20mA of DC thru them, you'd be surprised as how good they sound. If you want all DHT, replace 27 with 26, although I feel the 27 is extremely holographic and all that as well. I have a LOT of LEDs on the 71A cathodes (88 per channel now); the bass lightens up some (as there is some degeneration) but the sound is much clearer now that there are no electrolytics in the signal chain. I should add another, perhaps a third, string of LEDs and then the bass would suffer no more.
Just plug your headphones to the 4 ohm tap (10k primary) and it'll work magnificently!
I've been meaning to draw a schematic for some time.
(If you do build a 71A SET, please don't tell many people about it. I still haven't got enough spares in my stash. 😉
27 / 6C45P mu-follower direct coupled to 71A. Use Coleman type heater supply, and there will be no hum. Also tube regulated PS (I have 2 6V6GT's as pass tubes and 6SH7 as an error amplifier). OT is the lowly Hammond 125DSE, but with only 20mA of DC thru them, you'd be surprised as how good they sound. If you want all DHT, replace 27 with 26, although I feel the 27 is extremely holographic and all that as well. I have a LOT of LEDs on the 71A cathodes (88 per channel now); the bass lightens up some (as there is some degeneration) but the sound is much clearer now that there are no electrolytics in the signal chain. I should add another, perhaps a third, string of LEDs and then the bass would suffer no more.
Just plug your headphones to the 4 ohm tap (10k primary) and it'll work magnificently!
I've been meaning to draw a schematic for some time.
(If you do build a 71A SET, please don't tell many people about it. I still haven't got enough spares in my stash. 😉
I designed a headphone amplifier around the 71A driven by 5842 some years ago, used a choke load on the driver and fixed bias on the 71A - no problems with noise at all. The noise floor was below the limit of my M Audio 2496 based audio analyzer. I thought the headphone amp was "broken" when I saw that..
I'd be somewhat concerned about the dynamic resistance and potential linearity issues with that many LEDs in series. Obviously that resistance would raise rp significantly and might also result in significantly more output stage distortion than fixed bias. Not sure I would have the patience to solder two clusters of 88 LEDs either. 😛
I'd be somewhat concerned about the dynamic resistance and potential linearity issues with that many LEDs in series. Obviously that resistance would raise rp significantly and might also result in significantly more output stage distortion than fixed bias. Not sure I would have the patience to solder two clusters of 88 LEDs either. 😛
- Status
- Not open for further replies.