Questions about headphone amps vs. range of headphone impedances and sensitivities - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 18 Old 02-20-2013, 01:18 PM - Thread Starter
Senior Member
 
jimshowalter's Avatar
 
Join Date: Nov 2005
Posts: 316
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 43
I'm in the design phase for a DIY headphone-amplifier project, and am trying to understand the requirements.

In some ways, it seems harder to design a dinky little headphone amplifier than to design a power amp.

Loudspeaker impedances range from 2 Ohms to 16 Ohms, clustering around 8 Ohms. The max is only 8x the min (and the range is more typically 4-10 Ohms).

Headphone impedances range from 12 Ohms to > 600 Ohms, for normal headphones. The max is > 50x the min. (Electrostatics are way higher, but they require special amplifiers and so my questions don't apply to them.)

Headphone sensitivities range from < 80 dB/V to > 140 dB/V, which is a gigantic difference.

For a headphone amplifier to be able to drive high-impedance/low-sensitivity headphones and low-impedance/high-sensitivity headphones, without under-driving the former and without over-driving the latter seems to be pretty much impossible with fixed gain.

Yet most headphone amplifiers have fixed gain. Do they just not drive some headphones, and users select the headphone amps that match their particular headphones? What if they have more than one pair, with different impedances and/or sensitivities?

Some headphone amplifiers have switchable gain, but even the ones I found online with three gain options only cover about 20 dB of variation, well short of the 60 dB needed to cover the range of headphone sensitivities.

What am I missing?

My system? Google for: Martin Logan 420 CLX Descent Stage Summit
jimshowalter is offline  
Sponsored Links
Advertisement
 
post #2 of 18 Old 02-21-2013, 12:11 AM
AVS Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 3,645
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked: 250
There are so many things to take into consideration, i would suggest you start to read up on how to.

Here is a link that might start you in the right path. http://headwize.com/?page_id=79

Kevin Gilmore knows his stuff so you might want to start reading what he says about head amps.

ss
sillysally is offline  
post #3 of 18 Old 02-21-2013, 07:33 AM - Thread Starter
Senior Member
 
jimshowalter's Avatar
 
Join Date: Nov 2005
Posts: 316
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 43
Mr. Gilmore seems confused about feedback. He should read up on it here: http://www.linearaudio.nl/Documents/Volume_1_BP.pdf.

http://nwavguy.blogspot.com/2011/07/o2-headphone-amp.html

My question was not about circuit philosophy, it is a question with quantifiable values that needs a quantifiable answer.

My system? Google for: Martin Logan 420 CLX Descent Stage Summit
jimshowalter is offline  
post #4 of 18 Old 02-21-2013, 08:15 AM
Senior Member
 
Tazishere's Avatar
 
Join Date: Apr 2008
Location: Broadus, Montana
Posts: 250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 22
http://www.mav-audio.com/

These guys seem to have solved all the problems that you cite.
Tazishere is offline  
post #5 of 18 Old 02-21-2013, 08:45 AM - Thread Starter
Senior Member
 
jimshowalter's Avatar
 
Join Date: Nov 2005
Posts: 316
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 43
Quote:
Originally Posted by Tazishere View Post

http://www.mav-audio.com/

These guys seem to have solved all the problems that you cite.

.1% THD. I'm shooting for less than .001 %.

My system? Google for: Martin Logan 420 CLX Descent Stage Summit
jimshowalter is offline  
post #6 of 18 Old 02-21-2013, 11:17 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by jimshowalter View Post

I'm in the design phase for a DIY headphone-amplifier project, and am trying to understand the requirements.

In some ways, it seems harder to design a dinky little headphone amplifier than to design a power amp.

Loudspeaker impedances range from 2 Ohms to 16 Ohms, clustering around 8 Ohms. The max is only 8x the min (and the range is more typically 4-10 Ohms).

Headphone impedances range from 12 Ohms to > 600 Ohms, for normal headphones. The max is > 50x the min. (Electrostatics are way higher, but they require special amplifiers and so my questions don't apply to them.)

Headphone sensitivities range from < 80 dB/V to > 140 dB/V, which is a gigantic difference.

For a headphone amplifier to be able to drive high-impedance/low-sensitivity headphones and low-impedance/high-sensitivity headphones, without under-driving the former and without over-driving the latter seems to be pretty much impossible with fixed gain.

Yet most headphone amplifiers have fixed gain.

I question this because so many of the headphone amplifiers that I own, some consumer (FIIO), some pro (Rane) have gain controls. I have one (Boostaroo) that does not have a gain control. Then there are the portable digital players (Sansa Clip+ and Fuze)) that of course have a gain control.
Quote:
Some headphone amplifiers have switchable gain, but even the ones I found online with three gain options only cover about 20 dB of variation, well short of the 60 dB needed to cover the range of headphone sensitivities.

What am I missing?

Apparently, all of the many headphone amps that have gain controls???

Much of the angst about headphone amps mystifies me. This used to be a simple, common sense product category. Most headphone amps lack the one feature that I cherish the most - a built in equalizer. I don't know of any with a built-in 5 band full parametric or 30 band graphic equalizer.
arnyk is offline  
post #7 of 18 Old 02-21-2013, 11:21 AM
AVS Special Member
 
MarkHotchkiss's Avatar
 
Join Date: Mar 2007
Location: Long Beach, California
Posts: 1,476
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 82
Hi Tazishere,
Quote:
Originally Posted by Tazishere View Post

http://www.mav-audio.com/

These guys seem to have solved all the problems that you cite.
Maybe I missed it, but I don't see how they have addressed Jim's headphone impedance issue any differently than the cheap headphone amplifiers. They have enough voltage to push one watt through a 600 ohm load, which means they have a high voltage gain , but it looks like you would still need to turn the volume-control way down for an 8 ohm load.

The issue that Jim brought up boils down to this: How can you make a headphone amplifier that can produce different voltage gains in order to match different headphone impedances and still keep the sound-quality high? So far, nobody has addressed that issue.
MarkHotchkiss is offline  
post #8 of 18 Old 02-21-2013, 12:12 PM - Thread Starter
Senior Member
 
jimshowalter's Avatar
 
Join Date: Nov 2005
Posts: 316
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 43
Quote:
Originally Posted by MarkHotchkiss View Post

Hi Tazishere,
Maybe I missed it, but I don't see how they have addressed Jim's headphone impedance issue any differently than the cheap headphone amplifiers. They have enough voltage to push one watt through a 600 ohm load, which means they have a high voltage gain , but it looks like you would still need to turn the volume-control way down for an 8 ohm load.

The issue that Jim brought up boils down to this: How can you make a headphone amplifier that can produce different voltage gains in order to match different headphone impedances and still keep the sound-quality high? So far, nobody has addressed that issue.

Thank you. That is exactly what I am asking!

My system? Google for: Martin Logan 420 CLX Descent Stage Summit
jimshowalter is offline  
post #9 of 18 Old 02-21-2013, 03:01 PM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,148
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 41
Quote:
Originally Posted by MarkHotchkiss View Post

The issue that Jim brought up boils down to this: How can you make a headphone amplifier that can produce different voltage gains in order to match different headphone impedances and still keep the sound-quality high? So far, nobody has addressed that issue.
Vary the feedback as well as the input attenuation to produce the gain structure desired for the specific headphones. Make it as low noise as possible to keep SNR high in high gain application and design it well with good open loop linearity so that with low feedback distortion performance is still good. Multiple output devices in parallel to keep output Z low, and high standing current to keep it in class A all (or most) of the time, irrespective of the load. Use the highest voltage rails you can to allow it to swing as much as needed for high Z cans. Cascode at least the first diff pair.
A9X-308 is offline  
post #10 of 18 Old 02-21-2013, 03:42 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by A9X-308 View Post

Quote:
Originally Posted by MarkHotchkiss View Post

The issue that Jim brought up boils down to this: How can you make a headphone amplifier that can produce different voltage gains in order to match different headphone impedances and still keep the sound-quality high? So far, nobody has addressed that issue.
Vary the feedback as well as the input attenuation to produce the gain structure desired for the specific headphones. Make it as low noise as possible to keep SNR high in high gain application and design it well with good open loop linearity so that with low feedback distortion performance is still good. Multiple output devices in parallel to keep output Z low, and high standing current to keep it in class A all (or most) of the time, irrespective of the load. Use the highest voltage rails you can to allow it to swing as much as needed for high Z cans. Cascode at least the first diff pair.

An existing circuit style that meets many of these goals is known as a "Professional Microphone Amplifier". To turn a pro mic preamp into a headphone amp, just beef up the output stage.
arnyk is offline  
post #11 of 18 Old 02-21-2013, 04:53 PM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,148
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 41
Quote:
Originally Posted by arnyk View Post

An existing circuit style that meets many of these goals is known as a "Professional Microphone Amplifier". To turn a pro mic preamp into a headphone amp, just beef up the output stage.
A diamond buffer is easy to implement for the output, and I don't really see much need for gain of 20dB assuming a 0dBV input. Most of the time think it would be closer to unity.

The point of my post was that it's not really a difficult design brief.
A9X-308 is offline  
post #12 of 18 Old 02-22-2013, 09:09 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by A9X-308 View Post

Quote:
Originally Posted by arnyk View Post

An existing circuit style that meets many of these goals is known as a "Professional Microphone Amplifier". To turn a pro mic preamp into a headphone amp, just beef up the output stage.
A diamond buffer is easy to implement for the output

Seems overly complex when a simple complementary emitter follower would do the job.
Quote:
and I don't really see much need for gain of 20dB assuming a 0dBV input. Most of the time think it would be closer to unity.

The OP seems to think otherwise - he sees the need for a 60 dB variation in gain.
Quote:
The point of my post was that it's not really a difficult design brief.

Not difficult, and also probably not needed in the real world.
arnyk is offline  
post #13 of 18 Old 02-22-2013, 09:45 AM - Thread Starter
Senior Member
 
jimshowalter's Avatar
 
Join Date: Nov 2005
Posts: 316
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 43
Quote:
Originally Posted by arnyk View Post

Seems overly complex when a simple complementary emitter follower would do the job.
The OP seems to think otherwise - he sees the need for a 60 dB variation in gain.
Not difficult, and also probably not needed in the real world.

The spreadsheet I based the impedance and sensitivity ranges on is taken directly from manufacturers' specifications for their headphones.

My system? Google for: Martin Logan 420 CLX Descent Stage Summit
jimshowalter is offline  
post #14 of 18 Old 02-22-2013, 10:24 AM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,148
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 41
Quote:
Originally Posted by arnyk View Post

Seems overly complex when a simple complementary emitter follower would do the job..
4 transitors and seven resitors, plus the opportunity to use OPs with the ability to deliver some real current eg BD139/140 - almost clean enough to be used stand alone. Not complicated. Add a modern opamp like LM4562, include the BD inside the gNFB and you're done.

Quote:
Originally Posted by jimshowalter View Post

The spreadsheet I based the impedance and sensitivity ranges on is taken directly from manufacturers' specifications for their headphones.
Work out what voltage is required to give each headphone a given SPL, say 110. Number of Vrms gives about the required gain as you can safely assume most modern sources can supply 1Vrms.

Why do you need to be able to drive every headphone made?
A9X-308 is offline  
post #15 of 18 Old 02-22-2013, 10:53 AM - Thread Starter
Senior Member
 
jimshowalter's Avatar
 
Join Date: Nov 2005
Posts: 316
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 43
Quote:
Originally Posted by A9X-308 View Post

Why do you need to be able to drive every headphone made?

Well, it won't drive STAX, or other electrostatics like the Koss that has 100k impedance.

But ignoring electrostatics, there are other headphones that present a challenging load, and it seems kind of goofy to have a headphone amp that can't drive them.

My system? Google for: Martin Logan 420 CLX Descent Stage Summit
jimshowalter is offline  
post #16 of 18 Old 02-22-2013, 11:12 AM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,148
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 41
Quote:
Originally Posted by jimshowalter View Post

Well, it won't drive STAX, or other electrostatics like the Koss that has 100k impedance.
I know it won't because it won't be able to swing the hundreds of volts required to drive an electrostatic without a transformer. Apart from on the computer, I've used nothing but STAX (directly driven) since 85.
Quote:
Originally Posted by jimshowalter View Post

But ignoring electrostatics, there are other headphones that present a challenging load, and it seems kind of goofy to have a headphone amp that can't drive them.
Then run the numbers and work out what you actually need. For instance, there are no really low impedance phones that interest me, so I would not bother designing for 8R.
A9X-308 is offline  
post #17 of 18 Old 02-22-2013, 01:30 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 13,654
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 288 Post(s)
Liked: 997
Quote:
Originally Posted by A9X-308 View Post

Quote:
Originally Posted by jimshowalter View Post

Well, it won't drive STAX, or other electrostatics like the Koss that has 100k impedance.
I know it won't because it won't be able to swing the hundreds of volts required to drive an electrostatic without a transformer. Apart from on the computer, I've used nothing but STAX (directly driven) since 85.

Isn't a high voltage bias in addition to the high voltage signal required?
arnyk is offline  
post #18 of 18 Old 02-22-2013, 03:03 PM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,148
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 41
Quote:
Originally Posted by arnyk View Post

Isn't a high voltage bias in addition to the high voltage signal required?
Of course, 560Vdc on the Pro from memory. It's applied to separate pins on the STAX connector. Signal is balanced.
A9X-308 is offline  
Reply 2 Channel Audio

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off