Re: Connecting Headphones in Parallel With a Speaker

Putting headphones on the power amp output was always kind of cheap way of doing it. I don't think the resistance is high, but it may cause the headphones to be rather quiet at normal settings. 3K is the simple resistance, because it will limit its temperature to a .3 watt @ 100 watts. I would normally use two resistors, series, parallel, to get more versitility in wattage and ohms. The 3K is not really going to get that .3 watts unless 100 watts is maintained. So with a average of 20 % wattage is only .05 watts dissipation in the resistor if the 3K is used as a simple series component.

greg

It was pretty much the ONLY way of doing it. What alternatives were there? Even today, separate headphone amps are as rare as hens' teeth. Remember, we are talking about an end- user reproduction system, not a production studio.

Which is practically the definition of "too high".

The headphones are self-limiting by virtue of their relatively high impedance. As others have observed, throwing 3K in series will emphasize frequency-related impedance variation and make the headphones sound worse.

I guess since I bought my Hitachi preamp in the late 70's, I just plugged any headphones into the headphone output, driven by an internal headphone driver chip. I guess I got used to that having been done RIGHT. I agree that most times, its done crudely. I used to think it would screw up all the damping factor, well it does, but it does not seem to matter with phones.

greg

The other main thing, its almost always problematic with a headphone plugged in, not to have level mismatches when switching from phone off to phone on, or speaker output on or off. Having an additional volume control on the headphone sure helps.

greg

Headphones are generally high enough impedance that "damping factor" is NOT a factor at all.

Headphone diaphragms move only a microscopic amount compared to speakers. They don't need "damping factor".

What you are calling "crude" is perceived as simple, functional, and even elegant from other perspectives.

I would argue damping factor is independant of levels, as long as the levels are in the linear range.

greg

Since damping factor is closely linked to source impedance, and source impedance is closely linked to the desirable sitaution where the headphones are driven by a source with reasonably flat frequency response...

I think we have lost track of what "damping factor" is.

The most compliant headphone element is extremely stiff compared (on a proportional basis) to even an average modern LF speaker driver.

I agree with Arny that throwing a (relatively) large series resistance in the line will just exacerbate the frequency-response roughness (because of varying impedance). But that is a different phenomenon than "damping factor".

I agree that it is desirable to feed headphones (or most anything else, for that matter) from a low-impedance, flat (freq-resp) source. But "damping factor" needs a low impedance load as well, else you can't "control" the position of the transducer.

I have seen plenty of speaker curves, but I'm lost as far as headphone Z curves.

greg

Well, no, on several fronts.

First, if you actually MEASURE the impedance of most headphones, you will be hard pressed to find frequency- dependent variation everyone here seems to assume exist. Most headphones, regarldess of their rated impedance, exhibit almost purely resistive impedance.

With no or at most insignificant frequency-dependent impedance to worry about, driving them from a source with a reletively high source resistance does not pose a problem.

Secondly, the reason headphones do not exhibit these sorts of variations is becuas ethey are so heavily damped mechanically and acoustically.

It has abosolutely nothing to do with how far they move. It has everything to do with how much energu is stored in mechanical reatiances like suspension compliance and diaphragm mass, and how much is dissipated in losses like electrical resistance and mechnical friction. The ratio between these two, which in essence determines the frequency-dependent electrical impedance, remains pretty much the same regardless of how far the diaphragms move.

In the case of many headphones, the loss portion of the electrical, mechanical and acoustical portions of FAR overwhelms the reactive portions, thus a nearly resistive impedance.

And technically quite justifiably correct, once you actually know what the headphone impedance actually looks like.

Oh, and lastly, "damping factor" is a crock of s*it, just more so with headphones.

I would instead argue that damping factor is a useless concept when it comes to speakers in general and to headphones especially, given their common electrical properties.

I have measured quite a few, ranging from ear buds to open-type phones to large closed-type professional varieties from a number of manufacturers spanning several decades and almost without exception, they all present a nearly flat, resistive, frequency-independent impedance that varies but a few percent over the 10 octaves commonly considered "audio."

Thanks Richard. As soon as I read your first sentence the lights turned on and I realized the 3K would be to high. I did find something available off the shelf, by the way, that I believe will suit my purposes. It's at:

If you say so.

Same reason why it is desirable to feed just about all loudspeaker systems from a low-impedance source.

Discussions of damping factor focus on bass driver performance, but the actual audible effects of driving a speaker system from a high impedance source are very often more audible in the midrange.

Whether it is headphones or a loudspeaker system, the disease and the cure are often almost exact analogs of each other.

It is more likely that you'll find headphones with a fairly flat impedance curve than speakers. I'm not sure why that is, but it is possible that the explanation could be as simple as a shunt resistor.

Unfortunately this bunch appear not to have understood the measurement they are making. They have used a dummy head with ear canals and a microphone at the bottom of them. So they are measuring the response of both the headphone and the outer auditory system - which is exactly what is not wanted. It is the ability of the phones to deliver a flat response at the outer edge of the pinna that is the measure of the quality of the headphones. The first bit of the head's auditory system has to unflatten the response to whatever degree the rest of the internal system needs.

Note that I cited this source as a reference for impedance curves. ;-)

Did they have those too? I didn't bother moving on after the amplitude nonsense.

Impedance curves was the context in which I mentioned them...