Concertina Splitter as a stereo headphone driver

[Kipper4]

Is this even possible to derive a headphone level output from a concertina splitter.

Looking for a quick and dirty headphone driver for a circuit.



https://en.wikipedia.org/wiki/Phase_splitter

[mozz]

Don't know why you would want to use a phase splitter,  you can do push pull with a pair of transistors. 2n3904, 3906

[samhay]

You want to drive each headphone 180o out of phase?
Won't work very well as the output impedance is very different between the emitter and collector. You may get workable signal from the emitter, but not the collector.

[PRR]

The concertina, with equal loads both sides, WILL drive both loads equally.

It is mildly current-thrifty in this mode.

I too have NO idea why you would drive one signal to two ears out-of-phase.

The Q&D way to drive headphones is normally an LM386. The high current gain required to get from line impedance to headphone impedance discourages simple discrete plans.

[PRR]

Concertina near-optimized for two 32r loads and 9V supply, working just into clipping:


The max output is good for phones. The outputs are equal but *out of phase*. The sound will be "out there" instead of "in head". Fun but not what we usually want.

The current consumption is distressing and steady (idle or full roar). An LM386 driving 2*32r parallel will eat slightly more power at full roar but deliver 6X the max power. The LM386 will idle at 1/13th the current of this concertina. Even with the idealized bias-feed, the concertina input Z is lowish and varies a lot at high level.

[toneman]

I would like to recommend the PAIA headphone amp(see links below).
I have built several and they work surprisingly and quietly well.
Then there's the Headphone distribution amp-haven't built that.
It uses the same 5532 dual opamp as the Headphone amp.

PAIA headphone distribution amp:

https://www.paia.com/proddetail.asp?prod=9206K

PAIA (stereo) Headphone amp with dual opamp:

https://www.paia.com/proddetail.asp?prod=9605K

Schematic is there along with a good theory of operation.

Yeh, I know that it uses a bipolar supply.

Hope this helps in your project persuit.

Happy TurkeyDay!!

[GibsonGM]

If you want something simple, just for monitoring, amplifying a small signal....playing guitar into...this one is very easy, and is single supply.

I've used this many times, usually just one section.  Or build both for stereo.  It works remarkably well; I use one to jam when we have no power here for 3 to 6 days sometimes       Quick n Dirty, Rich!

[samhay]

>The concertina, with equal loads both sides, WILL drive both loads equally.

I stand corrected. Sort of.
If the load is balanced, it will work nicely. But what happens if you factor in some variance in the left vs. right speaker impedance?

[Kipper4]

Cheers guys

@Mozz you mean something like this?

http://www.electronics-diy.com/4x4.php

[mozz]

Yup, some cheap computer speakers use push a pair of push pull transistor to drive a small speaker. Your concertina will work for dual mono I guess but being out of phase you would lose bass?

[Kipper4]

So i mocked half of this up on the breadboard and it seems to work fine.
However I have added R4 47Ohms across the lh rh to simulate a load when nothing is plugged in.
Questions
Does this 47r need to be a larger wattage like 1w 47r ?
Is it needed at all.
Should i put a R to ground (dunno 100k) to help with any potential pop issues when plugging in the phones.

Tested without the 100k input pot
          the other half
          without the 47r

Otherwise seems like a good solution but might get binned off in favour of something else.

[Kipper4]

[Kipper4]

for clarity heres what im feeding into it

[PRR]

> The concertina, with equal loads both sides, WILL drive both loads equally
.... But what happens if you factor in some variance in the left vs. right speaker impedance?

Small difference hardly matters.

Ah-- if one side falls OFF (dirty plug), then things get screwy, and different for the two outputs. If the emitter side gets disconnected the collector side goes low-gain. Loss of load on collector side has less effect.

I still think it is a non-starter because why would we want OUT-of-phase in the two ears??

[anotherjim]

You could make it work if you could wire the headphone speakers in series for mono driving.
Or the full Van Gogh....


I'm not sure what effect reverse phase in stereo headphones will have with noise intended to influence mood as opposite to normal intelligable sound.

[Kipper4]

In the intrest of keeping the circuit fairly simple I wanted to add a headphones option as some sleepers may prefer to use headphones rather than an amp or speaker.

Dual mono was my thinking.
Does my add on look ok even with the dummy load 47r?

Is it worth the addition? Opinions please.

Is it worth the addition of an integrated speaker small 10w kinda thing rather than an headphones option.
What would i need to drive the speaker, new territory for me.

The wind and rain maker is doing what it should, I've contemplated a bit of mission creep with the addition of further 6 wave patterns available on the stomp lfo.

Bare in mind this circuit will be free to hopefully maker communitys and home hobbyist alike.
With the idea being that maybe autistic folks and those with sleeping issues may want to explore the possibility of building ones own circuit and further advance intrest in making and circuits.
Given that there is also the question of the enclosure.
be it a plastic margarine tub or a dedicated electrical enclosure it still needs engineering.

I hope this calrifys some stuff and am hoping for some good opinions and guidance.

[PRR]

This is a sleeper aid?

Then you do NOT want Watts. 100mW is more than you want.

If it is battery power, you want LOW power for decent life. (Which rules-out resistor-coupled amplifiers.)

[anotherjim]

I don't think the headphone driver will need any voltage gain. You will probably need fixed resistance in the output side anyway to limit the maximum volume. All the driver needs to do is buffer. So a dual op-amp wired up as unity buffers is all that is needed to drive each side of the cans. You can use the 5v for the noise chip as the DC reference for the opamp +input and the opamp runs off 9v. Apart from resistors to feed the DC reference to the +input and coupling capacitors, the opamp circuit is simple and less power hungry. Cheap duals like MC1458 or TL062 will do.

[Kipper4]

Like so?

[anotherjim]

Yes, the r/h version with a common input should work just as well.
You will need a series resistor out of each amp to limit the current - at least 100R but you may find it can go as high as 1k and still give sufficient volume. It will be important in the intended use that the user cannot set the volume so loud that it can cause hearing damage, which is easily done wearing headphones.

If there is no stereo in the job, the output resistors can be joined to mono, then only x1 output capacitor to feed both sides of the cans. The output cap will need to be uF sized to suit the load.  30ohm for stereo connection, 15ohm for mono with normal consumer headphones. So caps will be 100u to 470u (probably little difference in pricing in those sizes).

You will also need a pot of industrial-strength coffee close at hand while developing it, in case it works too well.