Headphone Outputs: Build-Out vs. No Build-Out

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by mediatechnology »

That's the cool thing about resistors: They come in every flavor of the rainbow.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by mediatechnology »

We have 3 sets of K240 "Monitor" that were bought in 2001... 600 Ohms

I just bought a pair of the new K240 "Studio" that are 50 Ohms.
I have to think that the days of headphones with impedances >100 Ohms or maybe >63 Ohms will diminish in a low-voltage battery-powered single-supply world.
Sure there are going to be niche products, but how many manufacturers are going to want to sell product that can't be used with an iPod or "USB-pod."

I have to think that drove AKG's decision to change the K-240 to 50 Ohms.
In the past they were notoriously difficult to drive because they required so much voltage.
Maybe I'm biased because I got tired or repairing passive Cue Boxes and replacing 600 Ohm K-240 transducers.

Low-voltage supply-limited applications are what motivated this train of thought: viewtopic.php?f=6&t=499
Driving both ends of the transducer makes lots more sense when running off batteries.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by JR. »

I suspect the old 600 ohm professional headphones were driven by the ever present 600 ohm common equipment interfaces,,, just plug up the cans to a 600 ohm line and your golden.

Consumer cans were always all over the place. When I last researched this a few decades ago before designing my one commercial headphone amp I found cans as low as 3.2 ohms. Probably for early battery powered units, while I recall seeing a battery powered RCA tube portable radio back in the 50's (wish I still had that).

JR
Cancel the "cancel culture", do not support mob hatred.
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by ricardo »

JR. wrote:I suspect the old 600 ohm professional headphones were driven by the ever present 600 ohm common equipment interfaces,,, just plug up the cans to a 600 ohm line and your golden.

Consumer cans were always all over the place. When I last researched this a few decades ago before designing my one commercial headphone amp I found cans as low as 3.2 ohms. Probably for early battery powered units, while I recall seeing a battery powered RCA tube portable radio back in the 50's (wish I still had that).
Don't forget the Senheisser HD414 @ 2k. The best crystal set phones in the last, this or the next millenium. That's a clue as to when I had a pair of those. :mrgreen:
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by mediatechnology »

I suspect the old 600 ohm professional headphones were driven by the ever present 600 ohm common equipment interfaces,,, just plug up the cans to a 600 ohm line and your golden.


I think you nailed it.
I've often wondered if there was some old broadcast standard that dictated phones would terminate a line just like a transformer...in a "+8" broadcast world.

There have been multiple occasions where I would clip-lead AKG K-240s directly to the line output of a Scientific Atlanta SEDAT receiver's line output to find a feed.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by mediatechnology »

I should also point out that current drive (bigger Build-Outs) reduce xtalk AT the phone transducers; ie where it matters. If you measure at the amp just after the Build-0uts, before the leads & shields, then bigger Build-Outs will show increased xtalk.
You are indeed correct ricardo.

Lowering the build-out makes it look really good at the connector.
I didn't see that from the beginning.
At the transducer there's no improvement and perhaps a worsening.
Both you and John S. set me straight on that.

Now as to which is better for the transducer, 0R build-out , infinite or somewhere in-between I remain curious. viewtopic.php?f=6&t=500

I think I'll fire up the HPA and do some LF sweeps of the MDR-7506 and listen for buzz, resonance etc both with and without a simple Rbuild-out.
User avatar
JohnSiau
Posts: 2
Joined: Tue Jan 17, 2012 2:50 pm
Location: Benchmark Media Systems, Inc., Syracuse, NY, USA
Contact:

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by JohnSiau »

mediatechnology wrote:I'm not sure the extreme "far more is better" on the damping factor holds since the headphone cable resistance appears in series and limits it several orders of magnitude over what is available at the amp output.
Damping factor is a very misleading measurement. The peak current created by back EMF is limited by the source impedance in series with the transducer impedance. 50% of the benefit is reached when the source impedance is equal to the load impedance (a damping factor of 1). 91% of the benefit is reached when the source impedance is 10% of the load impedance (a damping factor of 10). 99% of the benefit is reached at a damping factor of 100. Damping factors above 100 produce diminishing returns and are probably unnecessary.

The effective damping is really given by (Damping Factor)/(1+Damping Factor)

In the above examples:
1/(1+1) = 0.5
10/(1+10) = 0.909
100/(1+100) = 0.990

Damping current is:

(Back EMF)/(Rtransducer+Rsource)

Damping power is:

((Back EMF)^2)/(Rtransducer+Rsource)

But, a high-impedance transducer will start at a higher back EMF, so we need to normalize this equation to achieve a constant initial power (following a transient).

For constant power, the initial voltage needs to be scaled according to (Rtransducer)^0.5

For a constant initial power, the Damping power is:

(((Normalized Voltage)*(Rtransducer^0.5))^2)/(Rtransducer+Rsource)

The effectiveness of the damping is given by:

Rtransducer/(Rtransducer+Rsource)

which is the same as:

(Damping Factor)/(1+Damping Factor)


Bottom line, we should have a "damping effectiveness" specification which would be defined as:

Rtransducer/(Rtransducer+Rsource)
John Siau
V.P.
Benchmark Media Systems, Inc.

(315)437-6300
www.benchmarkmedia.com
User avatar
JohnSiau
Posts: 2
Joined: Tue Jan 17, 2012 2:50 pm
Location: Benchmark Media Systems, Inc., Syracuse, NY, USA
Contact:

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by JohnSiau »

Electrical damping can be applied to a resonating mechanical system to remove energy from the system. The power applied will determine how fast energy can be removed from the system. The more electrical power we apply, the faster we can change the mechanical motion of the system.

If we have no way of measuring unwanted motions in real time, and we have no models for the unwanted motion, then the best damping we can achieve will occur when we short the input terminals of the transducer. This implies that maximum damping is achieved when the drive impedance is 0-Ohms. As shown in my prior post, we can achieve 91% of the maximum damping when the source impedance is 1/10th of the transducer impedance.

Therefore a 3-Ohm drive impedance should be acceptable for a 30-Ohm headphone transducer (91% of maximum damping). A 0.3-Ohm drive should give excellent results with a 30-Ohm transducer (99% of maximum damping). Nothing significant would be gained by reducing the drive impedance below 0.3 Ohms.

If we want to improve upon a 0-Ohm drive system, we have several options:
1) Use a feedback system to actively control the position of the transducer cone
2) Use a feed-forward system to pre-distort the drive signal based upon the measured impulse response of the transducer
3) Improve the transducer
_a) reduce(moving mass^2 x velocity)
_b) reduce mechanical resonances in the driver and enclosure mechanical system
4) Add mechanical damping

Option 4 can only be applied in moderation. Mechanical damping can adversely affect the leading edge of transients. In contrast, electrical damping (low drive impedance) improves the leading edge of transients.

The energy stored in a resonating mechanical system is the square of the moving mass times the peak velocity. In our case, the moving mass is the mass of the cone + the mass of air moved by cone. Smaller cones must reach higher peak velocities to deliver the same power as a larger cone. But smaller cones have less mass. Exotic cone materials help, but at some point the mass of the air becomes a limiting factor. These tradeoffs are in the hands of driver manufacturers.
John Siau
V.P.
Benchmark Media Systems, Inc.

(315)437-6300
www.benchmarkmedia.com
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by mediatechnology »

Therefore a 3-Ohm drive impedance should be acceptable for a 30-Ohm headphone transducer (91% of maximum damping). A 0.3-Ohm drive should give excellent results with a 30-Ohm transducer (99% of maximum damping). Nothing significant would be gained by reducing the drive impedance below 0.3 Ohms.
Thanks John. Now we know where the point of diminishing returns is.
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: Headphone Outputs: Build-Out vs. No Build-Out

Post by ricardo »

JohnSiau wrote:If we want to improve upon a 0-Ohm drive system, we have several options:
1) Use a feedback system to actively control the position of the transducer cone
...
4) Add mechanical damping

Option 4 can only be applied in moderation.
In most phones, this is already applied to excess. The impedance curve tells you the relative importance of electromagnetic and mechanical damping.

If we take the Sony MDR-V6 on your page, http://www.benchmarkmedia.com/discuss/f ... dphone-amp I estimate the impedance peak at the main resonance as rmax = 74R7 @ 75Hz from a rDC = 60R (don't have exact figures but good enough for this illustration)

rmax - rDC = 74R7 - 60 = 14R7

The relative amount of Electromagnetic & Mechanical Damping is given by

Electromagnetic Damping / Mechanical Damping = (rmax - rDC) / rDC = 14R7 / 60R = 0.245

ie the Electromagnetic Damping (at Ro = 0) is less than 1/4 the Mechanical Damping. The Sony has more (relative) Electromagnetic Damping than most.

This relationship between the impedance curve & damping is EXACT for dynamic transducers including ribbons.
______________________

Mr. Siau, you don't by any chance have a B&K or GRAS 1/2" mike? If you're only interested in distortion, you could stick it in a store dummy head and do some measurements.

This millenium, it is possible to do good measurements of speakers w/o loadsa dosh cos the computing power now available and cheap electret capsules. But for distortion, you still need a B&K capsule.

Response of phones is another matter. The jury is still out on the most "accurate" artificial ear but there is more concensus today than in the early 80's when I was playing.
______________________

Also IF current feed or build outs reduce distortion, the measured electrical distortion is the MAXIMUM distortion reduction available.

So from John's distortion curves, the electrical distortion with Ro=30R on the Sony is -47dB (0.785%) at 30Hz.

So IF current feed does reduce the "electro-acoustic" distortion of the Sony, the maximum benefit at 30Hz with Ro=30R is a 0.785% reduction. Loadsa caveats cos 30Hz isn't where I'd expect the best benefit from current feed
Last edited by ricardo on Fri Jan 27, 2012 3:10 pm, edited 2 times in total.
Post Reply