You are here

Trying to understand headphones and their amplifiers

All about the tools and techniques involved in capturing sound, in the studio or on location.

Trying to understand headphones and their amplifiers

Postby solaris » Wed Dec 04, 2019 5:06 pm

Hello there.

I have a Sonifex RB-HD6 and I am thinking to buy a Tascam MH-8 (Or a PreSonus HP60)

The Sonifex's output is 150mW into 32ohm to 600ohm headphones
The MH-8's output is 250 mW +250 mW (32ohm)
The HP60's is 150mW @ 60ohm

I love specs as much as the next man but I don't know what to make of these.
Most importantly I don't know what type of headphones should I buy.

For example I want to buy a few DT 770's and they have three versions 32ohm, 80ohm and 250 ohm. Which one is the magic number and what is the difference?

Also, is the HP60's output hotter than the MH-9's and why the Sonifex has the same 150mW for a range of impedance's? What kind of wizardry is this?

Thank you so much for your help.
User avatar
solaris
Regular
Posts: 53
Joined: Mon Oct 27, 2003 1:00 am
Location: Greece

Re: Trying to understand headphones and their amplifiers

Postby Hugh Robjohns » Wed Dec 04, 2019 5:58 pm

solaris wrote:I love specs as much as the next man but I don't know what to make of these.

The MH8 will probably go slightly louder, at maximum, than the other two -- when used with the same headphone.

Most importantly I don't know what type of headphones should I buy.

These amps will work with pretty much anything -- although note the minimum impedance rating of the one amp. Some headphone amps can't drive low-impedance headphones and will distort or even be seriously damaged if you try!

So buy headphones on basis of open/closed for application, comfort and sound quality (and see below about impedance).

I want to buy a few DT 770's and they have three versions 32ohm, 80ohm and 250 ohm. Which one is the magic number and what is the difference?

As a rough rule of thumb, low-impedance versions (80 Ohms and below) are better for use with battery-powered devices where they are effectively current-driven, and high-impedances (typically around 150-600 Ohms) for mains-powered where they are essentially voltage-driven.

So you could hedge your bets and opt for the 80 Ohm version if you want to be able to use them on your phone or iPod occasionally... But any variant would work perfectly well with those amplifiers. Personally, I'd go for the 250 Ohm model for use with mains-powered headphone amps.

Also, is the HP60's output hotter than the MH-9's and why the Sonifex has the same 150mW for a range of impedance's? What kind of wizardry is this?

It's economy of information... They would need to provide a graph to illustrate the way the power delivery varies with different load impedances... so it's easier just to give a nominal figure. The issue would be headphones with too low or two high an impedance... but for all typical values, it will just work fine.

In fact, the DT770s are rated for a maximum input power of 100mW, so all of those headphone amps are theoretically overpowered for them anyway... Which just means you have useful headroom and they will go plenty loud enough...
User avatar
Hugh Robjohns
Moderator
Posts: 25417
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Thu Dec 05, 2019 12:20 am

Just to confound us further Solaris there has been a growing 'fashion' in the last couple of years to make headphone amps with a very low, near zero output resistance. This is claimed to give benefits in terms of a flatter frequency response and lower distortion. I have not seen any decent research that backs this up.

Providing a near zero output R is also fraught with certain problems . Conventional amps have a physical resistor of between 30 and 120 Ohms in series with each output. This serves two useful purposes..
1) It tends to compensate for the widely different headphone impedances.
2) It is a cheap, virtually bombroof way to protect the headphone amplifier. There is at least one high level headphone amp on the market that warns against shorting the output. That is not acceptable these days, to me at least.

I cannot see how a 'zero' resistance drive can make things any better when the cans have a DC resistance close to or even equal to their stated impedance?

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Trying to understand headphones and their amplifiers

Postby Hugh Robjohns » Thu Dec 05, 2019 11:42 am

There was an article in the magazine recently about near-zero Ohm output impednaces in headphone amps: https://www.soundonsound.com/techniques/phones-ohms

There's also a lot of useful information on the BenchmarkMedia website, including some AES papers with proper evidence... such as:

https://cdn.shopify.com/s/files/1/0321/7609/files/Headphone_Amplifier_Performance_-_Part_2.pdf?1361

And There's also a useful article about choosing headphones and the significance of impedance and sensitivity:

https://benchmarkmedia.com/blogs/application_notes/14017381-headphone-impedance-and-sensitivity
User avatar
Hugh Robjohns
Moderator
Posts: 25417
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Thu Dec 05, 2019 12:42 pm

Thanks Hugh. I did read that article with great care and did not think it gave any really solid evidence for the benefits very low Z drive?

My main concern is that surely, headphones have been designed for decades to be driven from a few tens, maybe a hundred or so source resistance whereas loudspeaker designers do of course assume a very low drive impedance. Do we tamper with that at our peril?

I shall read the other links avidly.

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Thu Dec 05, 2019 1:03 pm

Whilst I have the highest regard for Bnechmark product Hugh, based entirely on SOS reviews, that paper IS produced by Benchmark.
I would like to see a diagram of the test setup.

I also cannot understand how the electromagetic damping can be 3000 better when there is still a 60 Ohm resistance in circuit, that of the phones themselves? Perhaps I do not understand how a series resistance does not limit current in some circuits?

I am also at a loss to understand how vast numbers of talented studio people (such as yourself Hugh) and audiophiles (of the proper kind!) have 'missed' a distortion 50dB greater than that emerging from the monitors all these years?

Please believe me, I do find this all very confusing!

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Trying to understand headphones and their amplifiers

Postby Hugh Robjohns » Thu Dec 05, 2019 1:36 pm

ef37a wrote:... that paper IS produced by Benchmark.

Yes, it is... but a version of it was presented at AES for peer review, and John Siau is a very clever and genuine bloke.

But the SOS article -- which says much the same thing -- is from an independent source; the man behind Sonarworks which specialises in software correction to optimise the response of headphones... so he'd know a thing or two about the practical effect of output impedance on performance.

When I get a moment I'll try and track down the full AES paper for you.

H
User avatar
Hugh Robjohns
Moderator
Posts: 25417
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Thu Dec 05, 2019 3:12 pm

Hugh Robjohns wrote:
ef37a wrote:... that paper IS produced by Benchmark.

Yes, it is... but a version of it was presented at AES for peer review, and John Siau is a very clever and genuine bloke.

But the SOS article -- which says much the same thing -- is from an independent source; the man behind Sonarworks which specialises in software correction to optimise the response of headphones... so he'd know a thing or two about the practical effect of output impedance on performance.

When I get a moment I'll try and track down the full AES paper for you.

H

Ok, thanks Hugh.
It really is those two things that confuse me. The 'crazy' 6000 damping factor and the fact that no one has noticed that much distortion until quite recently.

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Trying to understand headphones and their amplifiers

Postby solaris » Thu Dec 05, 2019 8:05 pm

Hugh Robjohns wrote:But any variant would work perfectly well with those amplifiers. Personally, I'd go for the 250 Ohm model for use with mains-powered headphone amps.

Hugh Robjohns wrote:And There's also a useful article about choosing headphones and the significance of impedance and sensitivity:

https://benchmarkmedia.com/blogs/applic ... ensitivity

These will do it. I will get the 250 Ohm 770's. Thank you very much for your help.

ef37a wrote:Just to confound us further Solaris there has been a growing 'fashion' in the last couple of years to make headphone amps with a very low, near zero output resistance. This is claimed to give benefits in terms of a flatter frequency response and lower distortion. I have not seen any decent research that backs this up.

Hopefully I won't have to buy another pre-amp and headphones for another 5-10 years. (If I am lucky) So there's plenty of time for research. Hope I am not missing out on the greatest invention of all time. Probably not.

Again thank you for the help. We are all very fortunate to have this forum. :thumbup:
User avatar
solaris
Regular
Posts: 53
Joined: Mon Oct 27, 2003 1:00 am
Location: Greece

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Thu Dec 05, 2019 9:16 pm

solaris wrote:
Hugh Robjohns wrote:But any variant would work perfectly well with those amplifiers. Personally, I'd go for the 250 Ohm model for use with mains-powered headphone amps.

Hugh Robjohns wrote:And There's also a useful article about choosing headphones and the significance of impedance and sensitivity:

https://benchmarkmedia.com/blogs/applic ... ensitivity

These will do it. I will get the 250 Ohm 770's. Thank you very much for your help.

ef37a wrote:Just to confound us further Solaris there has been a growing 'fashion' in the last couple of years to make headphone amps with a very low, near zero output resistance. This is claimed to give benefits in terms of a flatter frequency response and lower distortion. I have not seen any decent research that backs this up.

Hopefully I won't have to buy another pre-amp and headphones for another 5-10 years. (If I am lucky) So there's plenty of time for research. Hope I am not missing out on the greatest invention of all time. Probably not.

Again thank you for the help. We are all very fortunate to have this forum. :thumbup:

We are very lucky indeed to have this repository of wisdom and technical brilliance (I am amazed they have put up with ME for so long!)

Your choice of the 250 Ohm phones at last makes me realize the 'wrongness' about that damping factor. By chosing the higher impedance cans you have increased the DF by a factor of over 4 compared to 60R jobs. WITH THE SAME AMPLIFIER! Obviously any distortions will be reduced by a further 12dB if used with a 'zero'Z amplifier.

Damping factor is thus at best just advertizing puff. Yes quote the ouptut resistance of the amp but not that measured on the bare circuit in the lab. By the time PCB tracks and jack contact resistances have been accounted for I doubt the figure is as low as 0.01 Ohms and it will be higher again at the actual terminals of the headphones. Curly leads have a lot to answer for it seems!

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Trying to understand headphones and their amplifiers

Postby n o i s e f l e ur » Fri Dec 06, 2019 9:41 pm

ef37a wrote:
Your choice of the 250 Ohm phones at last makes me realize the 'wrongness' about that damping factor. By chosing the higher impedance cans you have increased the DF by a factor of over 4 compared to 60R jobs. WITH THE SAME AMPLIFIER! Obviously any distortions will be reduced by a further 12dB if used with a 'zero'Z amplifier.

Damping factor is thus at best just advertizing puff.

Dave.


D'ya think? I'd have thought it was a reasonable factor in control of the phones?

This is the bit I'm most interested to ear Hugh chime in on - the "output impedance 1/8 of phones impedance" rule of thumb.
n o i s e f l e ur
Regular
Posts: 119
Joined: Sat Aug 04, 2018 6:20 am

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Fri Dec 06, 2019 10:24 pm

n o i s e f l e ur wrote:
ef37a wrote:
Your choice of the 250 Ohm phones at last makes me realize the 'wrongness' about that damping factor. By chosing the higher impedance cans you have increased the DF by a factor of over 4 compared to 60R jobs. WITH THE SAME AMPLIFIER! Obviously any distortions will be reduced by a further 12dB if used with a 'zero'Z amplifier.

Damping factor is thus at best just advertizing puff.

Dave.


D'ya think? I'd have thought it was a reasonable factor in control of the phones?

This is the bit I'm most interested to ear Hugh chime in on - the "output impedance 1/8 of phones impedance" rule of thumb.

Ok I'll try another tack? As far as I know, electromagnetic damping is proportional to magnetic flux and current. The flux is of course constant for any particular phone motor.

Let us have a headphone amplifier with a 30 Ohm feed resitor feeding 60 Ohm cans? For 1V rms the current is 11.1mA.

Now short the 30R for a 'zero' Z amp. The current increases to 16.7mA, an increase of just 1.5X. So how in the name of (insert diety of choice) does the damping factor suddenly become several thousands?

I am quite willing to have my arithmetic dismissed as not applicatble in this case but I would like to know how?

Of course, you can have very clever, negative resistance amplifiers using a combination of positive and negative feedback but to cancel out 60 Ohms, never mind 250 I think you would be pushing stability and distortion?

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Trying to understand headphones and their amplifiers

Postby Folderol » Fri Dec 06, 2019 11:26 pm

I must say I have exactly the same misgivings as Dave does. It doesn't matter where the resistance is. Damping is going to be related to the total series value. My AKG240s have a stated impedance of 55 ohms. They have a measured DC resistance of 54.2 ohms. The (protected) headphone output of my amp has a resistance of just under 3 ohms at the setting that suits those phones. Should I worry?
User avatar
Folderol
Jedi Poster
Posts: 9495
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Sat Dec 07, 2019 8:26 am

Folderol wrote:I must say I have exactly the same misgivings as Dave does. It doesn't matter where the resistance is. Damping is going to be related to the total series value. My AKG240s have a stated impedance of 55 ohms. They have a measured DC resistance of 54.2 ohms. The (protected) headphone output of my amp has a resistance of just under 3 ohms at the setting that suits those phones. Should I worry?

No Will but I would be very interested to know how that amplifier is 'protected' ? How did you measure the output resistance?

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Trying to understand headphones and their amplifiers

Postby Folderol » Sat Dec 07, 2019 11:03 am

The amp is 60W per channel. This is fed to a resistor ladder, high enough to not waste too much power (and also keep the output stage in it's most linear region), low enough to be insignificant against the phones impedance. The ladder is switched so I can get a close approximation to the same apparent sound level as the speakers. Calculating the value of the position that works for those phones gives 3 ohms. Measuring it by applying a signal, and then shunting so the level halves, needs 2.8(something)
User avatar
Folderol
Jedi Poster
Posts: 9495
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Sat Dec 07, 2019 11:27 am

Folderol wrote:The amp is 60W per channel. This is fed to a resistor ladder, high enough to not waste too much power (and also keep the output stage in it's most linear region), low enough to be insignificant against the phones impedance. The ladder is switched so I can get a close approximation to the same apparent sound level as the speakers. Calculating the value of the position that works for those phones gives 3 ohms. Measuring it by applying a signal, and then shunting so the level halves, needs 2.8(something)

Ok, might have guessed you'd do it right!

BTW, please excuse the memory of this old, med ridden bugger but...Did I send you my circuit idea for a near zero OP headphone amp that would not self destruct?

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Trying to understand headphones and their amplifiers

Postby Folderol » Sat Dec 07, 2019 1:55 pm

ef37a wrote:
Folderol wrote:The amp is 60W per channel. This is fed to a resistor ladder, high enough to not waste too much power (and also keep the output stage in it's most linear region), low enough to be insignificant against the phones impedance. The ladder is switched so I can get a close approximation to the same apparent sound level as the speakers. Calculating the value of the position that works for those phones gives 3 ohms. Measuring it by applying a signal, and then shunting so the level halves, needs 2.8(something)

Ok, might have guessed you'd do it right!

BTW, please excuse the memory of this old, med ridden bugger but...Did I send you my circuit idea for a near zero OP headphone amp that would not self destruct?

Dave.

I remember you mentioning it, but I don't think you actually sent it - leastways I can't find it here :?
User avatar
Folderol
Jedi Poster
Posts: 9495
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Trying to understand headphones and their amplifiers

Postby ef37a » Sat Dec 07, 2019 2:35 pm

Folderol wrote:
ef37a wrote:
Folderol wrote:The amp is 60W per channel. This is fed to a resistor ladder, high enough to not waste too much power (and also keep the output stage in it's most linear region), low enough to be insignificant against the phones impedance. The ladder is switched so I can get a close approximation to the same apparent sound level as the speakers. Calculating the value of the position that works for those phones gives 3 ohms. Measuring it by applying a signal, and then shunting so the level halves, needs 2.8(something)

Ok, might have guessed you'd do it right!

BTW, please excuse the memory of this old, med ridden bugger but...Did I send you my circuit idea for a near zero OP headphone amp that would not self destruct?

Dave.

I remember you mentioning it, but I don't think you actually sent it - leastways I can't find it here :?

Ok, (getting worse...coats coming soon) I have a scribble.

Dave.
ef37a
Jedi Poster
Posts: 11050
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk


Who is online

Users browsing this forum: No registered users