HiFiMan Susvara
Jan 13, 2021 at 4:39 AM Post #6,076 of 25,550
Basically everyone of us knows that headphones won't use the full 8W, 10W or whatever an amp is capable of.

However for Sound it often is the best if the amp is bored and doesn't use more than ~10-20% of it's Max power because it's most linear in that range depending on design

10% of an 8W amplifier, based on the Susvara's specs, would be at a loudness of 112dB, which will cause permanent hearing damage at prolonged listening. 10% of a 0.5W amplifier would be at a loudness of 100dB, which is still far louder than any person should be listening.
 
Jan 13, 2021 at 4:42 AM Post #6,077 of 25,550
You're not really wrong. When I first bought my Susvara, I was driving it with a Rogue Audio RH-5 integrated headphone amp (3.5W into 32 ohm) and the sound was entirely satisfactory, even excellent. Did I "need" more power? Not really, any more than a commuter "needs" a 600 horsepower Ferrari. But if you want the Ferrari experience (or the Hypex Class D sound, in my case), and 600 horses (300 watts) is the minimum they offer, then that's what you buy. You don't really need that much power, but it's not a detriment either.

Right, more power can't hurt anything, other than your ears, but it isn't going to be detrimental to the sound quality. One might consider it irresponsible to sell an amplifier capable of high power output should the user accidentally turn the volume up beyond safe listening levels and inadvertently damage their hearing, or God forbid, their fancy headphones!
 
Jan 13, 2021 at 6:25 AM Post #6,078 of 25,550
If you can afford a Susvara and all that Chord stuff you surely can afford "a pair of alligator clips and a 60ohm resistor" and do the testing instead of being so dismissive?
As I said, I have done my testing and am happy with my results. How am I being dismissive.
 
Jan 13, 2021 at 6:28 AM Post #6,079 of 25,550
@Bonesy Jonesy if you wanted to measure how much power is being used by the Susvara at your listening volume, I would be more than happy to tell you how to do it. I'm not trying to be rude, but there is a misconception on Head-Fi about how much power a headphone needs, fueled I think by what is being marketed to Head-Fiers. In reality, even the most demanding headphones don't need anywhere near the amount of power some of these amps are capable of providing. Quality is much more important than quantity! It just so happens the two go hand-in-hand because that is what is being sold - quality amplifiers with high power output, but it doesn't necessarily need to be that way.
L0rdGwyn, Thank you for the offer, but I don't need to be shown how to test electronic equipment as I already know how to do it !

There is also a massive misconception on loudness levels and amplifier outputs and correlating this to improvement in sound quality.
 
Jan 13, 2021 at 6:32 AM Post #6,080 of 25,550
L0rdGwyn, Thank you for the offer, but I don't need to be shown how to test electronic equipment as I already know how to do it !

There is also a massive misconception on loudness levels and amplifier outputs and correlating this to improvement in sound quality.

So you are saying you measured the Susvara taking 15W of power?
 
Jan 13, 2021 at 6:33 AM Post #6,081 of 25,550
Jan 13, 2021 at 6:35 AM Post #6,082 of 25,550
How did you come to that conclusion?

Why don't you go and build a 0.5 Watts and let us know your findings or if you have already (to back up your information and claims) send us some nice photos of it and test measurements.

In the meantime, I am very much enjoying listening beautiful music with my wonderful Susvara's connected direct to my Chord Mezzo 140 Speaker Amp and giving them circa 15 Watts, and looking forward to demoing a Hifiman E-1000 with my Susvara's (and giving them circa 11 Watts) hopefully in February (UK COVID lockdown restrictions allowed).
 
Jan 13, 2021 at 6:38 AM Post #6,083 of 25,550
Obvious to me that he meant, that those amps are capable of outputting that much power into the susvara load, not that they actually draw that much power

Those values are clearly the max output values given by the manufacturer and not measured via gear
 
Jan 13, 2021 at 7:25 AM Post #6,084 of 25,550
Obvious to me that he meant, that those amps are capable of outputting that much power into the susvara load, not that they actually draw that much power

Those values are clearly the max output values given by the manufacturer and not measured via gear

Fair enough, I did not check the spec sheets, my mistake if that is what @Bonesy Jonesy meant, but he has not specified exactly what he measured, only that he is satisfied with the results. The Susvara, and any headphone for that matter, does not need more than 0.5-1W of amplifier power to be driven to ear-destroying volumes and can be done with a high degree of linearity. This is not an opinion, this is physics. As I said, the point I made to you specifically, is that even 10% of a 0.5W amplifier into the Susvara will drive them to volumes beyond what could be considered healthy. The argument that an amplifier needs to be spec'd to 8W or more to operate in an area of high linearity is not valid. The 0.5W amplifier that I stated I am building has distortion figures at 0.05% THD at 1mW into a 300ohm load, second harmonic dominant, which is quite good for a single-ended tube amp, very linear despite being a "measly" 0.5W amplifier, I have no doubt the Susvara would sound excellent on it.

Here is an FFT as proof, 1mW into a 300ohm dummy load. The equivalent voltage swing into the Susvara would generate 5mW of power and a volume of ~90dB with 20dB of headroom. You'll have to excuse the noise floor as this was a prototype built on a piece of plywood, not exactly ideal as far as grounding and noise rejection go.

841 Spud 1mW into 300ohm Left Dedicated.png

And a photo from the prototype itself.

DSCF6225.jpg

I'll leave it at that, maybe we will see each other at a post-COVID Head-Fi meet and I can show you all in person :)
 
Last edited:
Jan 13, 2021 at 7:38 AM Post #6,085 of 25,550
Fair enough, I did not check the spec sheets, my mistake if that is what @Bonesy Jonesy meant, but he has not specified exactly what he measured, only that he is satisfied with the results. The Susvara, and any headphone for that matter, does not need more than 0.5-1W of amplifier power to be driven to ear-destroying volumes and can be done with a high degree of linearity. This is not an opinion, this is physics. As I said, the point I made to you specifically, is that even 10% of a 0.5W amplifier into the Susvara will drive them to volumes beyond what could be considered healthy. The argument that an amplifier needs to be spec'd to 8W or more to operate in an area of high linearity is not valid. The 0.5W amplifier that I stated I am building has distortion figures at 0.05% THD at 1mW into a 300ohm load, second harmonic dominant, which is quite good for a single-ended tube amp, very linear despite being a "measly" 0.5W amplifier, I have no doubt the Susvara would sound excellent on it.

Here is an FFT as proof, 1mW into a 300ohm dummy load. The equivalent voltage swing into the Susvara would generate 5mW of power and a volume of ~90dB. You'll have to excuse the noise floor as this was a prototype built on a piece of plywood, not exactly ideal as far as grounding and noise rejection go.

841 Spud 1mW into 300ohm Left Dedicated.png

And a photo from the prototype itself.

DSCF6225.jpg

I'll leave it at that, maybe we will see each other at a post-COVID Head-Fi meet and I can show you all in person :)
I totally agree, that headphones don't need that much power but that it's about the amplifier Design and quality.

There are plenty of examples in both ways, i.e. the Formula S has comparatively low power specs but certainly is one of the best sounding amps for Susvara and Abyss.

Or the Pass Labs XA25 speaker amp has "only" 25W into 8 ohm, which would be like 3,5 W into Susvara load compared to other speaker amps.(fair enough Tests have shown that the manufacturer horribly underrated those figures, testers measured up to 80W into 8ohm on this)
Still the sound of this amp was superior to many others with higher power ratings

Nonetheless there are improvements in sound when changing from most headphone amps to a speaker amp with the Susvara, which has to have a reason.
For a Lack of knowledge most of us Attribute this improvement to the bigger power reserves the amps provide.

Probably output impedance has also an important factor here, since e.g. my tube amp has 3 different impedance settings and while he can only output 1.2W in the lowest setting it can provide 4.5W on the highest.
With my Empyrean the lowest setting sounds Best, probably due to the better impedance matching
 
Jan 13, 2021 at 7:43 AM Post #6,086 of 25,550
I totally agree, that headphones don't need that much power but that it's about the amplifier Design and quality.

There are plenty of examples in both ways, i.e. the Formula S has comparatively low power specs but certainly is one of the best sounding amps for Susvara and Abyss.

Or the Pass Labs XA25 speaker amp has "only" 25W into 8 ohm, which would be like 3,5 W into Susvara load compared to other speaker amps.(fair enough Tests have shown that the manufacturer horribly underrated those figures, testers measured up to 80W into 8ohm on this)
Still the sound of this amp was superior to many others with higher power ratings

Nonetheless there are improvements in sound when changing from most headphone amps to a speaker amp with the Susvara, which has to have a reason.
For a Lack of knowledge most of us Attribute this improvement to the bigger power reserves the amps provide.

Probably output impedance has also an important factor here, since e.g. my tube amp has 3 different impedance settings and while he can only output 1.2W in the lowest setting it can provide 4.5W on the highest.
With my Empyrean the lowest setting sounds Best, probably due to the better impedance matching

My guess is you are right, it has more to do with the output impedance than it does with the power reserve. A speaker amplifier, by design, must have a low output impedance in order to properly damp the much more difficult load of an 8ohm speaker. The lower the impedance of the load, the higher its current demand. The powering device must have a low output impedance across the audio band to properly meet these high current demands, especially at low frequencies where output impedance tends to increase. Planar magnetic headphones are current-hungry loads, as far as headphones go, so would definitely benefit from the lower output impedance of a speaker amplifier, but these needs can also be met with a low-output-impedance headphone amplifier without the extra power reserve.
 
Jan 13, 2021 at 9:44 AM Post #6,087 of 25,550
Being an automotive engineer I like to think it like this (amp designers / experts, sorry for over-simplification :flushed:).

Even a very low-powered engine will allow any car to reach and maintain, say, 150km/h on a flat road, allowing sufficient time to get to that speed.
That's the analogy to having 0.5W into a Susvara and playing it 110dB on a stationary 1kHz signal.

But, if you want to drive your car through a curvy uphill road with a lot of accelerations, overtaking other cars etc. with ease, then you need not only power but an engine with a lot of torque.

Musical signals are abruptly variable ones, much more difficulto to cope with than a stationary sinewave, and in order to keep up with them we don't only need power, but also the closest analogy to torque that comes to my mind, i.e. current delivery, which is in turn related to amplifier topology (e.g. class A is usually good at that), power supply design, damping factor etc..

Of course, if we get a speaker amp with 100W/8ohm or a class A 25W like the XA25 we can safely assume that the dimensioning of the amp itself is plenty sufficient by default to provide more than enough current to satisfy the Susvara hungerness.

On the flipside, I believe it is not true that you strictly NEED 5W or more on the Susvara load if the amplifier is designed to deliver even a much modest power (say 1-1.5W minimum) with a beefy peak current output reserve.

In actuality, I have listened to the Susvara through the Formula S / Powerman, an Air Tight ATM 300B (8W/8ohm), my Riviera AIC-10 and very strong speaker amplifiers (Pass XA25, Viva Solista, Krell KSA100) and I have been always satisfied with 'quantity' (grunt, dynamics, slam, ...).

Rather, the quality of the amp - and its synergy with Susvara signature - is what stands out the most, though, in terms of tonal accuracy, nuanced presentation, transparency, refinement, soundstage layering etc., which was not - in my experience - correlated to raw power (would rather say the opposite).

I agree. I have also attempt to explain why we want to have powerful amps a few times by using the analogy of a car engine; which to be strong and powerful need both sufficient amount of horsepower and torque on tap. And also that current is often the limiting factor then driving hard to drive planers. Despite often repeated (wrongly) it is current that actually makes planar membrane move and not power/watts. My last post regarding this.

https://www.head-fi.org/threads/hifiman-susvara.853031/page-389#post-16060265
 
Last edited:
Jan 13, 2021 at 9:54 AM Post #6,088 of 25,550
My guess is you are right, it has more to do with the output impedance than it does with the power reserve. A speaker amplifier, by design, must have a low output impedance in order to properly damp the much more difficult load of an 8ohm speaker. The lower the impedance of the load, the higher its current demand. The powering device must have a low output impedance across the audio band to properly meet these high current demands, especially at low frequencies where output impedance tends to increase. Planar magnetic headphones are current-hungry loads, as far as headphones go, so would definitely benefit from the lower output impedance of a speaker amplifier, but these needs can also be met with a low-output-impedance headphone amplifier without the extra power reserve.

Well the output impedance, current delivering and power reserve are all important. Chord DAVE for example have very low output impedance (a few milliohm) but that alone is not enough for making it a champ on driving hard to drive headphones.
 
Jan 13, 2021 at 10:05 AM Post #6,089 of 25,550
Well the output impedance, current delivering and power reserve are all important. Chord DAVE for example have very low output impedance (a few milliohm) but that alone is not enough for making it a champ on driving hard to drive headphones.

What makes the Dave a champ? The low output impedance ensures it will be able to respond quickly to instantaneous current demands from the load.
 
Jan 13, 2021 at 10:05 AM Post #6,090 of 25,550
I tried, some don't understand the technical side of it, hearing 0.5W they think underpowered yet fail to realise the impedance this is into, 60 Ω, and without the experience of the innards of tube amps it's hard to grasp why this will work. Wishing 10 + 10 needs to be 100 isn't how this goes.


I have, 0.3W and 2W, told about it a few posts back, it works because of maths.

I agree that many doesn’t understand the technical side of hifi. I can’t say that I understand everything about how digital processers actually works and affects sound. Many times it’s better to just listen to which gear and combinations of gear that sounds good and report it to others. Empirical lists with suitable amps for headphones like the Susvara may be better than a general technical explanation for most people.
 

Users who are viewing this thread

Back
Top