Am slowly learning about system synergy. Feels like figuring out headphones and amps are the easy part. Once it comes to dacs, cables, power supply its all a mysterybut balance I feel comes down to the DAC, headphones, power supply and even cabling. Good power supply can remove a lot of noise, brightness from a system.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Aegis DIY Tube Headphone Amplifier
- Thread starter L0rdGwyn
- Start date
Commission a wood tube holder from @Toonartist. there's space next to the amp for it to fit. That way they'll always be on hand.I went on a cleaning frenzy and moved 95% of my tubes to my storage unit.
Enjoy, it still sounds great with the stock tubes! The design still shines through regardlessI'm jealous of all your fancy output tubes. I went on a cleaning frenzy and moved 95% of my tubes to my storage unit. I went down to my lair to fetch some alterative outputs (I've been listening to the ZMF Aegis with stock tubes leading up to and after CanJam for the sake of testing). To my horror, I have no output tubes left in my house
JJ tubes still sound very good so we're gonna keep going.
What are you listening to on low impedance?
mchuckp
100+ Head-Fier
Well…guess it is time to go downstairs and turn everything on for more burn in while I pack up my two other amps I just sold. Have no need for them having an Aegis.Yeah… give them time. I remember a tube I was running in on the Nirvana sounded well off for a good 20hrs but to my surprise came good. It’s why I put one new pair in at a time, run them for 60hrs before trying anything else. But, I have my amp on every day for work so it’s easy to do 60hrs
Enjoy, it still sounds great with the stock tubes! The design still shines through regardless
What are you listening to on low impedance?
Caldera, output Z on low setting for the balanced output is around 11ohms, so gets a good 5.5:1 damping ratio. For 1/4", it's around 4.5ohms, but the balanced vs. single-ended outs give a slightly different presentation IME.
BTW, I've been running my Aegis on max volume and controlling the volume using my DAC based on a recommendation from GoldenSound (when I can, on more sensitive headphones I can get occasionally too much noise).
Here's what he said about it, it applies to OTL and OTC amps:
Another thing to experiment with!
Here's what he said about it, it applies to OTL and OTC amps:
All else being equal, it's better to have the lowest gain structure possible. Which means highest starting signal, with the least amount of unnecessary gain later down the line.
You can't add dynamic range back once it's been lost, extra gain/amplification later will also amplify noise, and so having a higher starting dynamic range signal and amplifying less later is better.
BUT, amps will have a limit to what input voltage you can use, either in a hard limit that anything above will just clip, or a soft limit in that if you go higher it will distort more (this is really common in tube amps and actually most tube amps I've tested perform better with input voltage lowered to around 0.1-0.5V rather than normal 2V RCA)
Another thing to experiment with!
- Joined
- Apr 20, 2016
- Posts
- 1,780
- Likes
- 4,303
Commission a wood tube holder from @Toonartist. there's space next to the amp for it to fit. That way they'll always be on hand.
Do you have a picture of that? That sounds wonderful.
Myself, I've been using 3D printed Octal Tube stands like these:
https://www.thingiverse.com/thing:2933306
Haha. Not sure that he's commissioning them but he did make them.Do you have a picture of that? That sounds wonderful.
Myself, I've been using 3D printed Octal Tube stands like these:
https://www.thingiverse.com/thing:2933306
https://www.head-fi.org/threads/aegis-diy-tube-headphone-amplifier.965530/post-17927714
https://www.head-fi.org/threads/aegis-diy-tube-headphone-amplifier.965530/post-17927751
Toonartist
Formerly known as dAndis67
I intend to pickup some interesting wood for a couple of projects this summer so I may have a spare tube holder or two with whatever I have left if that’s the case I’ll post back here!
Pic of the current select with the two socket savers I use so the GEC KT88s can fit in when they’re not in use!
Pic of the current select with the two socket savers I use so the GEC KT88s can fit in when they’re not in use!
Attachments
BTW, I've been running my Aegis on max volume and controlling the volume using my DAC based on a recommendation from GoldenSound (when I can, on more sensitive headphones I can get occasionally too much noise).
Here's what he said about it, it applies to OTL and OTC amps:
Another thing to experiment with!
I don't really understand what he is trying to say there. The gain of the amplifier is fixed, you aren't really amplifying less but using the DAC as a signal attenuator, you are just lowering the input signal digitally at the DAC as opposed to dissipating away part of the maximum signal as heat in a potentiometer. For the same output volume, the same AC signal is going to end up at the grids of the input tube. But by putting the volume to max, you are also maximally amplifying any noise on the 6SL7 grids. Digitally attenuating the DAC signal is also going to worsen the signal to noise ratio of the DAC output. Question is which hurts the sound more (noise floor notwithstanding), dissipating part of the signal in a pot, or minimizing the output of the DAC.
Last edited:
I might have interpreted what he was saying incorrectly, I'll try and see if I can get some clarification!I don't really understand what he is trying to say there. The gain of the amplifier is fixed, you aren't really amplifying less but using the DAC as a signal attenuator, you are just lowering the input signal digitally at the DAC as opposed to dissipating away part of the maximum signal as heat in a potentiometer. For the same output volume, the same AC signal is going to end up at the grids of the input tube. But by putting the volume to max, you are also maximally amplifying any noise on the 6SL7 grids. Digitally attenuating the DAC signal is also going to worsen the signal to noise ratio of the DAC output.
- Joined
- Apr 20, 2016
- Posts
- 1,780
- Likes
- 4,303
Haha. Not sure that he's commissioning them but he did make them.
https://www.head-fi.org/threads/aegis-diy-tube-headphone-amplifier.965530/post-17927714
https://www.head-fi.org/threads/aegis-diy-tube-headphone-amplifier.965530/post-17927751
Ah yes, I remember those now. They look great. Thanks for posting.
khaine1711
500+ Head-Fier
- Joined
- Dec 12, 2011
- Posts
- 925
- Likes
- 177
Sound changes aside, it would be catastrophic if one swaps dac and forgot the volume pot is maxed ...BTW, I've been running my Aegis on max volume and controlling the volume using my DAC based on a recommendation from GoldenSound (when I can, on more sensitive headphones I can get occasionally too much noise).
Here's what he said about it, it applies to OTL and OTC amps:
Another thing to experiment with!
It also depends on the DAC but most I found sound (a tiny bit) better on fixed output (compared to variable output).
I do admit that I'd love the idea of integrated (or headphone amp) with a volume pot bypass switch to be used as an power amp so one could play with preamp/other method of volume control like tvc/ldr ...etc
Sound changes aside, it would be catastrophic if one swaps dac and forgot the volume pot is maxed ...
It also depends on the DAC but most I found sound (a tiny bit) better on fixed output (compared to variable output).
I do admit that I'd love the idea of integrated (or headphone amp) with a volume pot bypass switch to be used as an power amp so one could play with preamp/other method of volume control like tvc/ldr ...etc
Next best option is to just turn the volume to max, so there is no signal attenuation. But the input impedance of the amplifier will still be the resistance of the pot, less than the input impedance of the first amplifying stage, which would be an easier load for the source device. I've found there is a significant audible improvement using an AVC over a resistive pot, which does voltage to current conversion as opposed to loss of part of the output signal as heat in a pot.
I don't really understand what he is trying to say there. The gain of the amplifier is fixed, you aren't really amplifying less but using the DAC as a signal attenuator, you are just lowering the input signal digitally at the DAC as opposed to dissipating away part of the maximum signal as heat in a potentiometer. For the same output volume, the same AC signal is going to end up at the grids of the input tube. But by putting the volume to max, you are also maximally amplifying any noise on the 6SL7 grids. Digitally attenuating the DAC signal is also going to worsen the signal to noise ratio of the DAC output. Question is which hurts the sound more (noise floor notwithstanding), dissipating part of the signal in a pot, or minimizing the output of the DAC.
Alright here's Golden's response:
So in practice, most tube based devices I've tested seem to have a sweet spot for input voltage where distortion is best.
This can be seen by comparing the same output level from an amp but attained via different input voltages. Here's an example:
The image above is feeding 1V into a tube amp, and adjusting for 700mV out
Now this one is feeding only 70mV into the same amp, and adjusting to keep 700mV out
Distortion is way lower
This is THD vs input level for that particular amplifier
other devices I've tested behave very similarly just with the sweet spot shifted a bit
He then sent voice notes (I had dm'd him on telegram) so bear with me as I translate his thoughts:
He clarified that what you said isn't necessarily wrong but that it is assuming that a tube or transistor is going to behave identically in all situations, which they won't.
Solid state devices typically perform more linearly--- so usually you don't see this behavior on them.
But tube devices, in his experience of testing them, generally perform better with a lower line level input voltage, so that's why he recommends at least trying it to see if it improves your subjective experience. He said this is because they are not perfect devices and they don't perform linearly in all situations.
He also said you are correct about increasing the noise by turning up the volume on the amp, but that noise and distortion are separate issues that are effected differently.
He basically said only put the volume as loud as you can without hearing noise, if you hear noise turn the amp down and dac up.
He said if you can do this without getting a problematic noise floor at the output, it's probably worth doing. If your amp is noisier and you get a ton of noise by doing this, then it's not worth doing. The aegis is a quiet tube amp so I feel like it's worth giving it a shot?