HiFiman HE-500 (HE as in High End) please post your impressions. Proving to be a great headphone... - Page 181

Quote:

Originally Posted by davidsh 

I don't really get why a +1 watt amp is recommended when you will never use any more than 20mW in peaks (i doubt). Is it due to the low impedance, and the resulting distortion from the amp at say a couple of milliwatts?

Does anybody know in the first place, I have never found a reasonable explanation.

It's not about impedance as much as it's about sensitivity.

This should help a bit: http://support.audeze.com/entries/20866002-selecting-an-amp

Quote:

Originally Posted by h4mm3r 0f th0r 

maybe you are right... although i always thought gain= more power... 11.32 is +12db i think... louder volume ain't everything then...

Don't confuse gain with power.  An amp that caps out at say 5V, can have a gain of 20dB(10x) which will only serve to clip long before you run out of pot sweep ;)  My old Peak was like this.  6V output but 14x(23dB) gain.  It would have taken a really weak source of 400mV to use the whole sweep /facepalm

A lot of it is marketing 'no feedback' and the mind trick of being way down on the sweep making you think you have lots of power left ;)

Quote:

Originally Posted by MalVeauX 

 

Because Dr Fang himself stated about 1 watt should do it.

 

(though in an experiment with several different amps, I found the HE500 sounded fine from a tiny little Fiio E10 that outputs a 10th of that power, plenty of bass, mids were great, sounded pretty dynamic to me... so I don't know... and it clipped like a limping horse from an OTL LDMK III ... awful.. and I currently power from a 50 watt amp... so again, who really knows what's truly needed here...)

 

 

Very best,

You don't really need 1W, I think Fang was just simplifying and accounting for margin. An amp that can do a half watt (500mw) into the HE-500 38 ohms and 89dB/mW sensitivity can drive the HE-500 to 116db peaks.

89 + 10 × Log(500) = 116db.

That's very loud and you might only hit it listening to very dynamic orchestra music. Most people probably won't need nearly 116db peaks since most non orchestra music dynamics is compressed anyway. Not a lot of portable amps can do 500mw cleanly however (the O2 is one that can) and many amps mis-spec or spec with only driving one channel so it's probably just easier for Fang to say 1W.

Just want to chime in and say that I drive my HE-500 with the ODAC/O2 combo, and it's sounds really great. I had an Audio gd amp/dac (NFB-10SE) which put out 3W per channel, but I sold it after a few days with AB'ing the two. The O2 was better in every way possible.

I won't say that power don't matter, but I will say that how that power is used matters more. Quality > quantity.
 

The O2 tops out at 180mW into the HE-500 and that's being generous since at that level distortion is off the charts ;) 110dB or so.

This is my first post at Head-fi so I put in some effort. I am not a native speaker so please forgive my spelling mistakes.

I have given the power demands for headphones  some thought lately, because I am considering buying the Hifiman HE500/HE6 as well the Sennheiser HD800. At first, I was a bit skeptical about the power demands for the Hifiman HE-500/HE 6. The reason is that headphone sensitivity is measured at a miniwatt scale (1/1000 watt). Now let’s use  a speaker analogy to put the power demands of 1Watt (e.g. 1000miniwatt) at 38 Ohm to drive the Hifiman HE500 (sensitivity 89dB/1miniWatt) into perspective. A speaker with a 89 dB sensitivity/1 watt at one meter then needs 1000 watts to be properly driven. I have never ever heard anybody tell me that to properly drive a speaker with a 89db sensitivity you need at least a 1000 watt amplifier. And speakers are not placed near you ear drums, but need to power a whole room….

So why such demands on headphones like the Hifiman HE500 and even more so for the Hifiman HE6?

Now to know where I am coming from.  At this moment,  I use the AKG K1000 (74dB/mW, 120 Ohm) on a 8watt 300B speaker amplifier at the 8 ohm speaker outputs. This means that I have a maximum of about 0,5 watt to drive the 120 ohm load of the AKG K1000. This results in a  maximum output of 101dB for the AKG K1000. For most music and most cds that is OK because most cds compress dynamic peaks with peak levels often much less than 10db above average level (pop, rock). With the AKG K1000 I seldom get into trouble with pop and rock music and I do like to listen loud…

However there are exceptions especially in classical music. Well recorded symphonic music can reach peaks of + 16-20db. If you want to record that on cd without distortion, you have to lower the average recording level. This means that classical cds at a given volume setting generally result in less dB on your headphones than pop and rock cds. To listen to classical music with 20db peaks above average listening,  your amp first needs to have enough gain (e.g. the volume setting for listening to classical music will generally be  at a higher level than pop cds). At the same time your amp has to have enough power (watts) to deal with the 20dB peaks.

I my case with the very insensitive AKG K1000 I notice with some classical cds (symphonic work of Mahler, Bruckner etc)  compression and distortion on peaks. I personally find that tiresome and not relaxing, because while listening to music at a relatively high average level (what I like with headphones), I know that compression and distortion will occur at musical peaks. For me headphones are the only option to really enjoy orchestral music because  if I listen to this type of music through speakers I scare my wife and kids on musical peaks with also causes me not to relax.

That is why I think that you should not underestimate the importance of headroom in your amplifier. For instance, (single ended) tube speaker amps generally perform best (e.g. distort less) if they operate within their first watt. So ideally, and with headphones this is possible,  the single ended amplifier should stay below an output of 1 watt even on musical peaks. Now let’s calculate the power demands for the Hifiman He500 . I use a speaker amp and need to connect  the headphone at the 8 Ohm speaker outputs. My 300B amp puts out 8 watts at 8 ohm which equals 8 volt;  8 volt into the measured impedance of 38 Ohm for the Hifiman HE500 results into a maximum output of 1.7watts which corresponds with 119dB . 1 Watt at 8Ohm equals 2.83 volt; and 2.83 volt at the measured impedance of 38Ohm for the Hifiman HE500 results in 211miniWatts. The Hifiman HE500 has a measured sensitivity of 87 dB/1mwatt. As such it can reach levels of 110db with just 211miniWatts. Hence, again to translate that to speakers, it means that you use a 8 Watt amplifier to drive a (horn) speaker with a sensitivity of 110dB/1watt. I used to have huge hornspeakers with even higher sensitivity and I can tell you, it relaxes you knowing that no matter what happens there will be no compression and no distortion on musical peaks.

I further want to emphasize the importance of the power supply of the valve amplifier. Valve amplifiers are primarily concerned with voltage regulation which means the ability to keep a fixed voltage despite a varying current draw. Poor voltage regulation  results into poor bass control and extension. The effect of the power supply falling apart reveals itself as the sound falling apart, and all too often this happens before maximum power is achieved. It is my impression that speaker valve amplifiers need to pay more attention to the power supply than headphone amplifiers, because they more often reach their maximum power to drive speakers at reasonable outputs in large rooms. I think that this is an important reason, besides power demands, to prefer a speaker amplifier over a headphone amplifier. Speaker amplifier are maybe a bit better in providing sufficient current in demanding situations.

So want does this all mean. First, if you do not listen loud and you are not listening to demanding material you do not need much power. But if you want to relax no matter what music and at what levels you listen please go for enough headroom and with about 0,5-1 watt at 38 Ohms to drive the Hifiman HE500 you have just that. My two cents…. Hope this is helpful.

Look again ;)

http://www.innerfidelity.com/images/121105_blog_update_FirstAmpTests.pdf

If the O2 is current limited, highly likely, it could be as high as 380mW.

Quote:

Originally Posted by playitloud 

This is my first post at Head-fi so I put in some effort. I am not a native speaker so please forgive my spelling mistakes.

I have given the power demands for headphones  some thought lately, because I am considering buying the Hifiman HE500/HE6 as well the Sennheiser HD800. At first, I was a bit skeptical about the power demands for the Hifiman HE-500/HE 6. The reason is that headphone sensitivity is measured at a miniwatt scale (1/1000 watt). Now let’s use  a speaker analogy to put the power demands of 1Watt (e.g. 1000miniwatt) at 38 Ohm to drive the Hifiman HE500 (sensitivity 89dB/1miniWatt) into perspective. A speaker with a 89 dB sensitivity/1 watt at one meter then needs 1000 watts to be properly driven. I have never ever heard anybody tell me that to properly drive a speaker with a 89db sensitivity you need at least a 1000 watt amplifier. And speakers are not placed near you ear drums, but need to power a whole room….

 

So why such demands on headphones like the Hifiman HE500 and even more so for the Hifiman HE6?

Now to know where I am coming from.  At this moment,  I use the AKG K1000 (74dB/mW, 120 Ohm) on a 8watt 300B speaker amplifier at the 8 ohm speaker outputs. This means that I have a maximum of about 0,5 watt to drive the 120 ohm load of the AKG K1000. This results in a  maximum output of 101dB for the AKG K1000. For most music and most cds that is OK because most cds compress dynamic peaks with peak levels often much less than 10db above average level (pop, rock). With the AKG K1000 I seldom get into trouble with pop and rock music and I do like to listen loud…

 

However there are exceptions especially in classical music. Well recorded symphonic music can reach peaks of + 16-20db. If you want to record that on cd without distortion, you have to lower the average recording level. This means that classical cds at a given volume setting generally result in less dB on your headphones than pop and rock cds. To listen to classical music with 20db peaks above average listening,  your amp first needs to have enough gain (e.g. the volume setting for listening to classical music will generally be  at a higher level than pop cds). At the same time your amp has to have enough power (watts) to deal with the 20dB peaks.

 

I my case with the very insensitive AKG K1000 I notice with some classical cds (symphonic work of Mahler, Bruckner etc)  compression and distortion on peaks. I personally find that tiresome and not relaxing, because while listening to music at a relatively high average level (what I like with headphones), I know that compression and distortion will occur at musical peaks. For me headphones are the only option to really enjoy orchestral music because  if I listen to this type of music through speakers I scare my wife and kids on musical peaks with also causes me not to relax.

 

That is why I think that you should not underestimate the importance of headroom in your amplifier. For instance, (single ended) tube speaker amps generally perform best (e.g. distort less) if they operate within their first watt. So ideally, and with headphones this is possible,  the single ended amplifier should stay below an output of 1 watt even on musical peaks. Now let’s calculate the power demands for the Hifiman He500 . I use a speaker amp and need to connect  the headphone at the 8 Ohm speaker outputs. My 300B amp puts out 8 watts at 8 ohm which equals 8 volt;  8 volt into the measured impedance of 38 Ohm for the Hifiman HE500 results into a maximum output of 1.7watts which corresponds with 119dB . 1 Watt at 8Ohm equals 2.83 volt; and 2.83 volt at the measured impedance of 38Ohm for the Hifiman HE500 results in 211miniWatts. The Hifiman HE500 has a measured sensitivity of 87 dB/1mwatt. As such it can reach levels of 110db with just 211miniWatts. Hence, again to translate that to speakers, it means that you use a 8 Watt amplifier to drive a (horn) speaker with a sensitivity of 110dB/1watt. I used to have huge hornspeakers with even higher sensitivity and I can tell you, it relaxes you knowing that no matter what happens there will be no compression and no distortion on musical peaks.

 

I further want to emphasize the importance of the power supply of the valve amplifier. Valve amplifiers are primarily concerned with voltage regulation which means the ability to keep a fixed voltage despite a varying current draw. Poor voltage regulation  results into poor bass control and extension. The effect of the power supply falling apart reveals itself as the sound falling apart, and all too often this happens before maximum power is achieved. It is my impression that speaker valve amplifiers need to pay more attention to the power supply than headphone amplifiers, because they more often reach their maximum power to drive speakers at reasonable outputs in large rooms. I think that this is an important reason, besides power demands, to prefer a speaker amplifier over a headphone amplifier. Speaker amplifier are maybe a bit better in providing sufficient current in demanding situations.

 

So want does this all mean. First, if you do not listen loud and do not listening to demanding material you do not need much power. But if you want to relax no matter what music and at what levels you listen please go for enough headroom and with about 0,5-1 watt at 38 Ohms to drive the Hifiman HE500 you have just that. My two cents…. Hope this is helpful.

Excellent!

I, indeed, did not mentioned the notion of distortion! That's a valid argument since the rated power is given for x% of distortion. The farther from the maximum power rating, the farther from the maximum distortion.

The same goes for every other technical measurement (crosstalk, SNR, ...) that get worse when power demand increase.

Joys of independent measurements ;)  I believe the original article is NwAvGuy provided numbers while the second are Tyll's results.  Or more to the point, the second set can be fairly compared to the rest of the group while the first can't.