Benchmark headphone amp-white paper by John S.

Status
Not open for further replies.
Jump to Latest
#2
We all know the effects of damping on a speaker, a mechanically vibrating system. Probably from just after WWII.
Now we discover that headphones are also a mechanically vibrating system, and therefor reacts similarly to damping. This is news?

The surprise for me is the apparent incompetence of many headphone amp designers. 30 ohms in series - jeez.


Jan
 
#4
headphone amplifier output Z has long been known to have audible effects - just more likely heard in the frequency response - getting a bass boost at the mass-spring resonance of a single driver headphone with those that reflect the mechanical resonance into a electrical terminal Z peak

adding switchable R series to passively EQ to taste has been a feature of some headphone amps

but it does seem like having a low output Z amp to start is the more flexible position
 
#6
jan.didden said:
We all know the effects of damping on a speaker, a mechanically vibrating system. Probably from just after WWII.
Now we discover that headphones are also a mechanically vibrating system, and therefor reacts similarly to damping. This is news?

The surprise for me is the apparent incompetence of many headphone amp designers. 30 ohms in series . . . .
Click to expand...
If I recall correctly, Neville Thiele discussed this and presented the mathematical analysis in some of his work from the early 1960's. In particular, a few folks tested his ideas and found that some commercial speaker systems of those days had measurably better performance (flatter response) with a few ohms of series resistance.

Dale

p.s. - My interests are now much different but in my mid 20's I looked up and read the papers by Thiele and Small. I believe the AES had just re-published them to make the work more accessible in North America. I was impressed that their scholarly papers were very understandable and readable for my basic engineering education, yet mathematically rigorous and thorough. Almost certainly still worth a read today by anybody interested in that topic.
 
#7
Yes that is true, I remember a speaker design by Morgan Jones that was designed with 3 ohms source resiastance in mind to get best response. There's a thread somewhere on diyaudio on it.

But the overwhelming part of commercial designs are designed to be driven by a voltage source = 0 ohms output impedance.

Jan
 
#8
jan.didden said:
The surprise for me is the apparent incompetence of many headphone amp designers. 30 ohms in series - jeez.
Click to expand...
That's not even a lot. The ISO standard that nobody in their right mind adheres to specifies 120 ohms. (Granted, it dates from the days of 600 ohm cans.) And integrated amps with primitive voltage droppers might have anything from 220 to 600 ohms. I wouldn't even be surprised if some of these old 8 ohm cans (remember them?) had been designed for current source operation.

That an added series resistor makes distortion visible if the relationship between voltage and current in the transducer isn't all that linear shouldn't be too surprising really. I can measure plenty with my trusty HD590s plus 47 ohms (loopback test with splitter cable and the lowly ALC269 in a notebook). That's not too interesting though - I'd be far more interested in how it looks on the acoustic side, aside from the obvious warped frequency response. If there are any changes in distortion, how do some Sennheisers with relatively little inherent damping fare, vs. some AKGs with high damping?
 
#9
This might become more interesting. 😉
Anyone out there with enough time to do some tests with some well known headphones ? I would have done some on my Sennheiser 580's but currently there is lack of time. Maybe a simple mic would do ( for acoustic measurements) as the relative measurements might be sufficient to see what's happening rather than an absolute measurement.

But it seems quite apparent that many manufactures used various output resistors ( when they did ) to achieve a specific end result knowing quite well what they would do to the response. They might not have been aware of the distortion variation it introduced.......or didn't care ! Some probably just introduced resistors for 'safety' and arbitrarily chose the value ......for ordinary consumer use ? In the old days the bump in the bass might have been attractive to many people with phones of that age.
 
#10
Historically speaking, most headphones have been coupled to the front panel headphone port of a receiver. The series resistor was introduced to reduce the output from the receivers power amp output to the headphone. Simple, cheap and almost effective.... no seperate headphone amp was/is used.

THx-RNMarsh
 
#11
That's correct. Most headphone sockets did come on receivers and amplifiers meant mainly for loudspeaker use.

However I think amplifiers meant specifically for headphone use can't get away easily. The manufacturers would have known , at least to some extent, what was happening. I would guess that those that addressed the issues were also expensive !😉
 
Status
Not open for further replies.
Top Bottom