This is a review and detailed measurements of the Schiit Valhalla tube headphone amplifier. It is on kind loan from a member. The unit is discontinued. I think it came out in 2011 (?). From what I can tell, the cost was US $450 or so.
The sample I have has been through some war duty but otherwise matches the look of typical Schiit products:
The volume control has a good feel.
There is nothing in the back but a set of RCA inputs, power and on/off switch so I will not be posting a picture of that.
The driver tubes were 6H1N Russian tubes. The output tubes were 6H6N.
The unit is very heavy for its size and gets incredibly warm.
Not much else to the unit so let's get into measurements.
Headphone Audio Measurements
Here is our usual dashboard:
Even though I am at just 2 volt output, distortion was through the roof. I thought something was broken by Shiit specs the unit at 0.5% distortion so we are "fine."
While Channel 1 was stable, channel 2 was all of the place. Its output level and THD+N would constantly vary. Reseating the tubes did not help so something is wrong with those set of tubes/channel.
SINAD (signal over noise+distortion) is just 43 dB, placing the Valhalla at dead last position in our headphone amplifier line up:
Signal to noise ratio at full level and 50 millivolts out is:
Let's forget about the second channel which is off the reservation (broken). The "good" channel is well, bad:
That is a second last place finish. Best not use any sensitive headphones with this amplifier.
Frequency response of the good channel is fine:
Output impedance was a very high, 75 ohm:
This will change the frequency response of many headphones and rules out using it with low impedance headphones due to high losses.
Output power vs THD+N at 300 ohm looks pretty horrific to the eye:
Forgetting again the broken channel, we see that the other channel also has very high levels of noise and distortion. The latter is mostly caused by asymmetric clipping of the negative cycle together with symmetrical rounding of the tops of the curve.
Testing with 33 ohm completely upsets the apple cart:
Stopping at -30 db, we barely have 1.5 milliwatts of power! There are dongles that do better than this.
Multitone response as you can imagine, is very poor:
I think we should stop here and not belabor the point.
Headphone Listening Tests
I thought I get the easy out of the way with testing with Hifiman HE-400i low impedance headphone. As soon as you got any decent volume out of the distortion was evident especially with bass notes. Turning the levels higher would cause massive distortion, akin go turning up a clock radio to max volume. It is basically not usable.
Switching to Sennheiser HD-650 completely transformed the experience. Now the fidelity was good and there was plenty of volume until distortion set it. Correlation was quite weak with THD+N when I compared the Valhalla with JDS Labs Atom. Unless I really paid attention where the JDS Labs had slightly less distortion in highs and tiny bit better resolution, you could not tell them apart. Indeed I had my son here who is in his 20s and has excellent high frequency acuity. He could not tell the two units apart.
The dominances of second harmonic and the fact that it masks better, explains why the audible effect is not anywhere close to what the numbers indicate.
Mind you, there was change in frequency response due to high impedance of Valhalla. I could hear the tonality changing a bit, with highs more muted with Valhalla. This also made level matching more difficult.
Of note, there was a tiny soundstage difference in favor of the Atom because it has slightly more highs. Otherwise, there was no magic to the sound of the tube other than the subtle things noted above. Mid-range was NOT more mellow. Soundstage was NOT wider. Nothing was more liquid, etc.
Conclusions
Between high distortion levels, fragility of the tubes, and very high power consumption and heat, I can't think of a reason to buy the Schiit Valhalla. This Schiit, is really Schiit.
That said, if you have an affinity for things like this, and can buy one very cheap, and have a high-impedance headphone, and like a permanent but subtle EQ applied to all of your music, I am not going to chase you around the block if you buy the Schiit Valhalla.
-------
As always, questions, comments, recommendations, etc. are welcome.
We plan to go on a road trip next week. While we have money in the bank, spending you all's money is always more fun! So please donate what you can using:
https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
The sample I have has been through some war duty but otherwise matches the look of typical Schiit products:
The volume control has a good feel.
There is nothing in the back but a set of RCA inputs, power and on/off switch so I will not be posting a picture of that.
The driver tubes were 6H1N Russian tubes. The output tubes were 6H6N.
The unit is very heavy for its size and gets incredibly warm.
Not much else to the unit so let's get into measurements.
Headphone Audio Measurements
Here is our usual dashboard:
Even though I am at just 2 volt output, distortion was through the roof. I thought something was broken by Shiit specs the unit at 0.5% distortion so we are "fine."
While Channel 1 was stable, channel 2 was all of the place. Its output level and THD+N would constantly vary. Reseating the tubes did not help so something is wrong with those set of tubes/channel.
SINAD (signal over noise+distortion) is just 43 dB, placing the Valhalla at dead last position in our headphone amplifier line up:
Signal to noise ratio at full level and 50 millivolts out is:
Let's forget about the second channel which is off the reservation (broken). The "good" channel is well, bad:
That is a second last place finish. Best not use any sensitive headphones with this amplifier.
Frequency response of the good channel is fine:
Output impedance was a very high, 75 ohm:
This will change the frequency response of many headphones and rules out using it with low impedance headphones due to high losses.
Output power vs THD+N at 300 ohm looks pretty horrific to the eye:
Forgetting again the broken channel, we see that the other channel also has very high levels of noise and distortion. The latter is mostly caused by asymmetric clipping of the negative cycle together with symmetrical rounding of the tops of the curve.
Testing with 33 ohm completely upsets the apple cart:
Stopping at -30 db, we barely have 1.5 milliwatts of power! There are dongles that do better than this.
Multitone response as you can imagine, is very poor:
I think we should stop here and not belabor the point.
Headphone Listening Tests
I thought I get the easy out of the way with testing with Hifiman HE-400i low impedance headphone. As soon as you got any decent volume out of the distortion was evident especially with bass notes. Turning the levels higher would cause massive distortion, akin go turning up a clock radio to max volume. It is basically not usable.
Switching to Sennheiser HD-650 completely transformed the experience. Now the fidelity was good and there was plenty of volume until distortion set it. Correlation was quite weak with THD+N when I compared the Valhalla with JDS Labs Atom. Unless I really paid attention where the JDS Labs had slightly less distortion in highs and tiny bit better resolution, you could not tell them apart. Indeed I had my son here who is in his 20s and has excellent high frequency acuity. He could not tell the two units apart.
The dominances of second harmonic and the fact that it masks better, explains why the audible effect is not anywhere close to what the numbers indicate.
Mind you, there was change in frequency response due to high impedance of Valhalla. I could hear the tonality changing a bit, with highs more muted with Valhalla. This also made level matching more difficult.
Of note, there was a tiny soundstage difference in favor of the Atom because it has slightly more highs. Otherwise, there was no magic to the sound of the tube other than the subtle things noted above. Mid-range was NOT more mellow. Soundstage was NOT wider. Nothing was more liquid, etc.
Conclusions
Between high distortion levels, fragility of the tubes, and very high power consumption and heat, I can't think of a reason to buy the Schiit Valhalla. This Schiit, is really Schiit.
That said, if you have an affinity for things like this, and can buy one very cheap, and have a high-impedance headphone, and like a permanent but subtle EQ applied to all of your music, I am not going to chase you around the block if you buy the Schiit Valhalla.
-------
As always, questions, comments, recommendations, etc. are welcome.
We plan to go on a road trip next week. While we have money in the bank, spending you all's money is always more fun! So please donate what you can using:
https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
