[OT] 3.5mm stereo to mono adapter with switch?

[Carlos E.R.]

And you think she will be able to notice the distortion be ear alone? LOL.

[micky]

Well if she can't notice it, what difference does it make?

[pamela]

How true. There many be many theoretical objections but the final test must be whether the voice recordings I play are any less intelligible when the signal paths are shorted.

[pamela]

If there are reasons other than demand for no supply of a switched adapter then why don't those reasons prevent supply of this non-switched version?

[Carlos E.R.]

That thing is very cheap. A switch is more expensive and can fail.

[Johnny B Good]

====snip====

Well spotted, that man! :-)

It only occurred to me, as I was reviewing my "posted" follow up just now, that this effect would only apply in the case of driving into a high impedance load, not when driving into a matched load where such built in impedance matching was by nature of the feedback rather than the simple addition of series connected resistors to a low Z output amp.

However, unless the designer envisions such usage cases where other signal sources can land up driving currents into the amplifier's output terminals as a matter of routine (this bridging of the left and right channels to create a 'mono mix' being just one specific case), amplifiers are usually designed to have a constant voltage characteristic (constant in that for a given input signal level, the output voltage level will remain constant over a range of loadings that extend from open circuit to some specified minimum value of load impedance).

The designer of portable music players (MP3 players and the like) is more likely to specify a low Z output amp on the basis that this will allow headphone sharing with a simple stereo headphone socket doubler without loss of volume due to halving of the headphone loading (although clipping level might drop by a dB or so along with a very slight increase in THD).

Basically, there's no advantage in designing the amplifier to tolerate having its left and right channel outputs bridged together to create a distortion free mono mix. That's a problem best left to those few who feel the need for such a facility so strongly that they can provide their own solution whether distortion ridden or free as they deem fit (there are, amazingly, people around who can cheerfully tolerate distortion levels that most of the populace would normally find intolerable).

[Johnny B Good]

That picture looks like a 2.5 to 3.5 mm stereo jack plug adapter. A bare picture of an adapter plug minus any meaningful description of its purpose is simply a picture and no more. What was this supposed to be again? :-)

[The Natural Philosopher]

I of course not. You short the amp *inputs* together and use just one channel for output.

[micky]

To nitpick, "More expensive" relates to demand, the reason she suspected. Demand would be lower because the price is higher.

Some switches are good for millions of makes and breaks, but they cost more. This seems also like a matter of price and demand.

[Martin Brown]

Ignoring the fact that shorting amplifier outputs together isn't that good for them the closest to what you want to do with standard parts would be a headphone splitter that parallels up two 3.5mm stereo sockets and a mono 3.5mm plug to short left and right together. eg

You might be able to find a mono 3.5mm adapter unswitched and carry both. But I can't see anyone making this item... no demand ;-)

[tabbypurr]

Amps for speakers are always low impedance output, barring very unusual cases. Not so for headphones, which seem to tolerate higher impedance just fine, and where the connectors used are prone to shorting on insertion.

NT

[pamela]

I originally described what I wanted as: "an adapter which goes from a 3.5mm stereo plug to 3.5mm mono socket"

I couldn't find a picture which showed exactly that so I posted a picture I said was "similar". Perhaps I should have looked harder for a more accurate picture!

[pamela]

Apart from personal tolerance towards distortion, speech sounds okay even with much higher levels of distortion than music.

For example this article finds "For detecting distortion at levels of less than 10%, the test frequencies had to be greater than 500 Hz". The fundamental frequency of human speech very rarely goes above 250 Hz although, admittedly, a lot of voice energy is in the harmonics.

All in all I think my personal need is likely to be met if I get mono from two shorted stereo channels. I will still try your earlier suggestion of using a hard panned stereo source to maximise distortion when combined and see how that affects speech intelligibility.

[Carlos E.R.]

Consider. If you do it in a single plastic block, it will be biggish and stress the socket on the device. If you do it with a wire connecting the plug and the socket, and the switch in the middle, you add complexity.

Then you have to consider that the very idea of connecting left and right amplifiers together is wrong, cabling a switch is even more wrong.

On the other hand, devices that have mono output audio may use a stereo output socket simply because stereo headphones are more common nowdays. Other devices have an internal switch to force "mono" audio. For those devices I see correct that you can use a plug converter. Placing a switch makes obvious that you want to "short" the amplifiers, which can not be correct and might cause damage... so I would not build it.

/rambling.

[Dave Plowman (News)]

Does it? Depends on personal preference, but I find mobile phone quality often very wearing to listen to.

You just try understanding speech with a steep low pass filter at 250 Hz.

Standard telephone bandwidth is given as 300 - 3000 Hz. To reproduce male speech well, you need the same sort of bandwidth as most music.

Well recorded male speech is actually a rather good test of a loudspeaker.

[pamela]

The fundamental frequencies for speech are surprisingly low but the harmonics at those frequencies you mention must have a lot of energy.

Even so, quite a lot of distortion can be tolerated when listening to speech for intelligibility especially when compared to listening to music.

[Dave Plowman (News)]

Don't think it's as simple as harmonics. The basic 'noise' might be produced by the vocal chords, but is heavily modified and added to by the bits afterwards - like tongue palate teeth and lips.

Totally different reasons. Unless just listening to both for pleasure. Plenty got pleasure out of the very first gramophone records. As there was no alternative.

I'd not want to listen to a play on R4 if it was heavily distorted. It would give no pleasure.

[Roger Hayter]

snip

Do any of these structures have a sufficiently non-elastic mechanical response to create non-harmonic frequency products? This is a genuine question, while am surprised it they do I don't actually know.

[Dave Plowman (News)]

Not sure I understand the question. Things like the 'p' sound are produced by breath being stopped by the lips. As are other of the consonants, by the tongue etc. Obviously altering the shape of the mouth changes the harmonic content of a note produced by the vocal cords.

Speech is pretty complex which is why it's difficult to synthesise. To the point where it sounds like a real voice.

[Roger Hayter]

I was just making the point that different resonators and linear elements can only enhance or suppress harmonics, they can't create different new frequencies. But, as you point out, most consonants involve highly non-linear starts and stops of various mouth movements, so these can clearly create new frequencies both by separate resonances (not from the vocal cords) and intermodulation. So you are clearly right!