Headphone Volume Limiter

tired of gettin eardrums dead for 3 days due to massive volume levels being triggered for reasons unknown.
,
is there a zener i can slap across the bones so i don't get bit?

they should have one built in,

Most of Sennheiser's 600-Ohms headphones comes with two 9V1 Zeners across the coils. Probably most for copper protection, not for the ear.

But a series resistor and a couple of zeners should do the trick?

Jakob E.

CJ said:

tired of gettin eardrums dead for 3 days due to massive volume levels being triggered for reasons unknown.
,
is there a zener i can slap across the bones so i don't get bit?

they should have one built in,

Click to expand...

You understand that there is a relationship between the voltage limit, the impedance, the h/p sensitivity and the corresponding max SPL. So you'll have to tune the zener value with each particular h/p.
You want to make the system bilateral, treating positive and negative voltages identically, so you'll need either a bridge (1.2V threshold), or a back-to-back arrangement. The lowest value zeners, at 1.2V, will limit the absolute voltage at 1.8V in b-to-b configuration, thus limiting power at ca. 100mW into 32 ohms cans (considering that under clamping, the waveform is almost square), which is enough to fry your ears. You could use just diodes, that would limit the p-p voltage to 0.6V (2 diodes) or 1.2v (4 diodes), resulting in 11 mW or 44 mW. Not too fine resolution. In addition, the onset of distortion for sinewave is at half the limiting power.
Finally, there is a danger in putting zeners or diodes across the input, since it could damage the amplifier.
I have designed dozens of studio headphone systems, I always installed a resistive network that limits the max spl, protects the amps, isolates a faulty can when it fries and ensures more or less consistent power whatever the headphones impedance.

BTW, there is a new regulation that limits the SPL produced by earbuds and headphones available to the public at 94dBspl for a voltage of 75mV rms. So h/p manufacturers are hastily modifying their products in order to comply. I've been consulting for Audio Technica on this subject, which is not utterly trivial.

I have been interested in this subject for years (decades),, ever since kids started blowing out their ears with Walkmans,

A zener clipper seems awfully crude and arbitrary. We can surely come up with a more sophisticated limiter based on THAT chipsets.

I suspect to prevent hearing fatigue is roughly like response patterns for hearing damage so some Fletcher-Munson curvature in the side chain.

OF course if listening for monitoring purposes you don't want to mess with spectral balance... for that matter you should probably normalize for loudness too.

Interesting topic...

JR

For example, the germanium diode shunt with 32 ohm input resistor and separate output resistor (apparently for tuning the clip threshold to the individual driver characteristics) is not the most fidelic method.  Also the input and output resistor forms an impedance to overcome by having to turn up the volume (huh...)

It has apparently been patented by a woman in NH yet abandoned.

Prior art from the 1980s shows a jfet (Pearson et al?) but falsely claims Zener voltage thresholds when opposing parallel Zeners prevent the Zener voltage from being reached.

Notwithstanding headphones are usually varying in their freq response when measured with a coupler and are not usually matched so well. Plus I have had $200 cans with one ear polarity swapped. C'mon.

I was thinking of power limiting by using volts * amps with a multiplier (to account for different style/impedances of cans) but one might just as well do some analog computation for RMS and get into THAT territory

but one still needs to know the driver efficiency and it's ability to convert electrical energy into mechanical/acoustic energy when impinges upon the ear drum.  Then again everyone is statistically different and my NIOSH/OSHA 85~95 dB dosimitrey thresholds over time may be way too loud for your threshold of noise induced hearing loss (remember all medical stuff is statistical group study based). Noise induced hearing loss seems to be a big problem with the kids these days especially with them earbuds.

Too bad we don't have bone conduction that work well yet. Everyone has a 2.1 port for sound conducting..

JohnRoberts said:

...
A zener clipper seems awfully crude and arbitrary. We can surely come up with a more sophisticated limiter based on THAT chipsets.
...

Click to expand...

But why? Where (in the chain)?

For inserting -between- a headphone jack and the preamp's output, a resistor-into-LED string (careful with polarity) will be usually a "good enough" thing. Or zeners.

The other "thing" I had in mind would be using voltage-controlled chip-amps (like some TDA chips) and use some circuit to sense peaks and "normalize" the amps' gain.

But that would require a "reset" button, and furthermore a lot of tinkering to which "default" gain it would "reset" the gain, because if you just set it up as an AGC which would ride gain up-and-down, that would be irritating as f+++ in a mixing/tracking situation.

Hey, diodes, you look good again, haha ...

tv said:

The other "thing" I had in mind would be using voltage-controlled chip-amps (like some TDA chips) and use some circuit to sense peaks and "normalize" the amps' gain.

Click to expand...

It's a far cry from the original idea of something that can be fitted in the jack or in the cans. This problem is a daily concern for IEM engineers. You can see that the solutions they implement in order to satisfy the basic requirement of controlling excess power are far from trivial, including several layers of analog and/or digital processing, which come in addition to the built-in limiter in the transmitters.

I've built some chipamp can-drivers for me and local studios, and one thing that all of these had in common was big headroom, low impedance drive (most of time no output resistors), for tracking rooms. (In old days, TBA800 was my favorite to build a self-contained low-profile module, it was LOUD, drummer-loud, much more than enough).

But one problem was always how to determine what was an excessive peak and how to get around it. One solution that worked to some extent was to filter the amp itself (which can be done within the NFB of a chipamp, and works, in a politically incorrect guerilla way), the other to use a pre-emphasised diode clipper and post-filter it, and third would be to "squelch" the pig (err, peak), like in amateur radio, CB etc, i.e to mute (or filter) the thing for a split-second, but that can confuse if you're in a session. At least that was what I was testing.

The "diode" things worked more reliably IMO. Low tech..

I'm biased, been a bass player for ages, hate not to use cans, also hate thin sounding cans.