[calibro]

Thank you for answers and for share your experiences!!!!  

 

[Operakid]

Prego!  Buona fortuna per ottenere un suono eccellente. La musica e' il vero obiettivo.
 
 
(So folks don't get offended and inform me that his is an English speaking forum: You are welcome, Good luck obtaining good sound.  Music is the true goal.)
 
I think this is so often forgotten in high end audio.  I have the impression that Head-Fiers are more into the music.  Certainly more people mention the music than traditional 2 channel high enders.     

 

[xaval]

I seldom visit this part of the forum but I'm glad I did today. I'm also using a tube integrated to feed my phones.
 
It's a shame I have no technical knowledge to contribute appart from the fact that since I started using the ART Headtap connected to the amps speaker tap all phones suddenly got a significant jump in quality, particularly low impedance ones. I have no idea what's inside the box as even if I would crack it open it would all be just "parts" :s
 
The fact remains that the tube speaker amp sounds wonderful not only through speakers. Hope this thread develops even more

 

[Operakid]

I am assuming you are unhooking your speakers...otherwise you would be listening to the speakers and headphones at the same time.
 
So, if the speakers are unhooked and the amp is just playing the cans through the Headtap, you are using the device in a way it is not intended.  There is no parallel resistor across the amp leads so the amp will be playing into a high impedance load.  A apeaker amp is not intended to do this so results will vary and potential for oscillation will vary.  
 
This device uses the monitor speakers as the lower impedance load the amp would like to see, and you are just tapping off to play your headpones: the load on the amp would remain  nearly identical to what it was, it would not know there was a high impedance load tapped off.  

 

[Operakid]

The more I think about it, some good thinking went into the volatage divider that Calibro linked.  This divider, through manipulation of the resistor values, would allow the amp to see a low impedance, the headpone see a higher impedance, with the desired attenuation achieved by changing the second parallel resistor.  The advantage over my setup would be that I could get my attenuation without having the headphones looking back at such a high impedance.  That may result in better sound, it may not.  But it offers the advantage of knowing.  
 
A simpler way of saying it is that with the linked voltage divider 3 variables can be set at any value one wants: load impedance on the amp, source impedance the cans see, and attenuation. 
 
I do not agree with his stated resistive choices as he has the amp looking at a higher than typical load impedance and the cans looking at a lower than typical source impedance.  However, the concept is nice and one can just choose different values to make the amp and cans happy.    

 

[StudioSound]

I have a number of extremely high quality tube and solid state amps for speakers. I am wondering about using them with Sennheiser HD800 headphones, which I do not have yet (backordered). I know that my hundred watt transformer coupled tube amps will lost a lot of power at that impedance, and it's not an efficient use of gear and electricity, but are the results typically good when doing this?

I also have 140 watt OTL (Atma-Sphere) amps I could try.

Anything to watch out for when doing this? Have folks had good results?

I would be very concerned about hooking these headphones up to a 140 watt speaker amplifier when the maximum load they can handle is 0.5 W.

And typically you want to be at the higher end of the range with most amplifiers, but with headphones connected, you will presumably need to have the volume down very low, even with some resistors connected.

I decided to try the idea I talked about earlier of loading the output of the amp at 10 ohms (could not find a high quality 8 ohm non-inductive wire wound resistor) and a dropping resistor after that and before the phones, starting with 100 ohms.

Well, the sound is phenomenal. A huge difference in width (not my most important thing), a major increase in refinement (really I'm saying major decrease in grain/distortion), beautiful sweet sweet, delicate highs with tremendous extension, deeper bass (the HD800s go amazingly low in the bass if the amp is stiff enough). The transparency is really amazing.

It sounds like what you are hearing is a change in the frequency response of the headphones caused by a damping factor that is too low. The HD800 have a 300 Ω impedance, which means that the maximum output of the amplifier should be 37.5 Ω.

Here is what happens to the frequency response when the damping factor is too low with a pair of Sony MDR-V6 for example:

That certainly seems to fit with the description you give of the change in sound - not that an HD800 will change in exactly the same way, but it will probably be similar.

And here is distortion with the HD650, which is a 300 Ω headphone just like the HD800:

It will also do a number of other bad things to the sound: http://www.benchmarkmedia.com/discuss/feedback/newsletter/2011/12/2/0-ohm-headphone-amplifier-sonic-advantages-low-impedance-headphone-amp


There is a reason headphone amplifiers are designed as they are, with as low an output impedance as possible.

 

[calibro]

Those problems occurs even with a costant impedance curve (as for orthos) or not ??

 

[StudioSound]

Those problems occurs even with a costant impedance curve (as for orthos) or not ??

I'm not sure to be honest - it looks like it shouldn't affect the frequency response, but may negatively impact other characteristics?

From looking at things, it seems that Benchmark's measurements closely match InnerFidelity's MDR-V6 impedance measurements. (listed in Ohms)

Which would indicate that as you reduce the damping factor, the frequency response of an HD800 is going to change something like this:


Which follows Operakid's comments on "deep bass" when connected up to a speaker amplifier with resistors in the path.

 

[xaval]

Quote:

I am assuming you are unhooking your speakers...otherwise you would be listening to the speakers and headphones at the same time.
 
So, if the speakers are unhooked and the amp is just playing the cans through the Headtap, you are using the device in a way it is not intended.  There is no parallel resistor across the amp leads so the amp will be playing into a high impedance load.  A apeaker amp is not intended to do this so results will vary and potential for oscillation will vary.  
 
This device uses the monitor speakers as the lower impedance load the amp would like to see, and you are just tapping off to play your headpones: the load on the amp would remain  nearly identical to what it was, it would not know there was a high impedance load tapped off.  

Hi, and thanks for your input and good reasoning. As I have a Stax energizer also connected to the amp; I have my speakers connected to the energizer and use the energizer as a switch to either listen to the speakers or the headtap. If this is not how to correctly use the device I'd appreciate some more input as it's sounding amazing... can it get better? 

 

[Operakid]

Despite above concerns by other posters, the cans are safe as I am using the same volume settings as with my speakers due to the parallel and series resistance network I am using for the cans.
 
As for the damping factor, this is a bit overblown and misunderstood.  The q of the speakes and entire system ties together with the damping factor to make damping factor alone an unpredictable thing.  For instance, I have heard a high impedance amp with low impredance speaker (damping factor = 2) have phenomenal dynamics in the bass while a very low output impedance ss amp with the same speakers be a sound soft and flat in dynamics in the bass.  It really depends on the driver, what it wants to see from the amp, and the amp and what it wants to see in the driver, and damping factor is one small thing.  Certainly this combination using the big speaker amp is so dynamic and tight on impactful bass that it puts the highly regarded headphone amps to shame in this regard.   High end 2 channel audio went through this same thing with damping factors and slew rates, finally it all fell away as it never really predicted the sonic aspects it was supposed to.

 

[xaval]

After a few questions about the Headtap, here goes a couple crappy phone pics. Not much to see 

 
 

 
 

 
 

 

[Operakid]

Regarding the Stax SRD-7 that a prior poster asked about, you need to do nothing when using on a speaker amp.  It is made for speaker amps.  
 
In the near future I'll take some input transformer impedance readings with an impedance bridge, we'll see what AC impedance load it is.      

 

[Argybargy]

Interesting topic.
I'm thinking of building a Millet DCPP Engineers Tube Amp which is a 18 wpc into 8 ohm speaker amp (transformer coupled output).
Output impedance is about 2.5 ohms.
I want to use the following 4 different headphones with this:
Denon D7000 25 ohm 108dB sensitivity
Audio Technica ATH w3000anv 40 ohm 102dB sensitivity
Audeze LCD2 50 ohm 91dB sensitivity
Hifiman HE500 45 ohm 89dB sensitivity
 
1) Would I need to set up some kind of "impedance switch" that switches between various resistors?
2) Or would an "averaged" value single resistor work?
3) Or should I set one resistor value for the most sensitive headphones that are most likely to pick up background hiss/noise and just use more volume pot movement for the other headphones?

 

[Operakid]

Quote:

Interesting topic.
I'm thinking of building a Millet DCPP Engineers Tube Amp which is a 18 wpc into 8 ohm speaker amp (transformer coupled output).
Output impedance is about 2.5 ohms.
I want to use the following 4 different headphones with this:
Denon D7000 25 ohm 108dB sensitivity
Audio Technica ATH w3000anv 40 ohm 102dB sensitivity
Audeze LCD2 50 ohm 91dB sensitivity
Hifiman HE500 45 ohm 89dB sensitivity
 
1) Would I need to set up some kind of "impedance switch" that switches between various resistors?
2) Or would an "averaged" value single resistor work?
3) Or should I set one resistor value for the most sensitive headphones that are most likely to pick up background hiss/noise and just use more volume pot movement for the other headphones?

 
 
You may not need any parallel resistor at the amp terminals as these are all relatively low impedance headphones.  With the lower sensitivity headphones you would be throwing a lot of energy away through a parallel resistor.
 
However, that is a wide range in sensitivities.  Especially for those real efficient cans you may need a series resistor at the amp, a parallel resistor after that, and another series resistor before the headphones, like that voltage divider circuit a prior poster gave a link to.  I'd be careful with those 102 and especially 108db cans, not to blow them up with a speaker amp.  I personally would not be hooking them up without resistive loading.  

 

[Mambosenior]

So happy for reading this thread. It has totally transformed my HP-listening experience. Thanks.