Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Op-amp to headphones: Forget Matching?

Status
Not open for further replies.
DigiTan said:
Would it be possible to cancel that by putting a small capacitor parallel with the feedback resistance?
No, but if you over sized the bias voltage filter capacitor so the bias voltage rises slow enough (2-3 seconds) on power up, you will reduce the amount of power up POP to almost nothing.
 
Last edited:
kchriste said:
No, but if you over sized the bias voltage filter capacitor so the bias voltage rises slow enough (2-3 seconds) on power up, you will reduce the amount of power up POP to almost nothing.

Not that this hasn't been a thread worth reading, but that hint caps it. Going to have to remember that one.


Torben
 
I guess I'll have to try that last suggestion and see how it goes. Spikes and noise I could probably deal with, but that POP issue would've been a problem for another project.
 
kchriste said:
No, but if you over sized the bias voltage filter capacitor so the bias voltage rises slow enough (2-3 seconds) on power up, you will reduce the amount of power up POP to almost nothing.
It depends on the input common mode voltage range of the amplifier. Most amplifiers don't work when the bias voltage is within a few volts from ground. As the capacitor charges, when the voltage reaches the allowed input common mode voltage then the amplifier suddenly jumps awake and makes a POP.

You can do it wirh a lousy old LM324 and LM358.
You can also do it with a good MV3317x, MC3407x or TLE214x because they have an allowed input common mode voltage down to 0V. But they are opamps, not power amps.
 
The Maxim thing is just a charge pump converter + amp in the same package. You can get the same thing with an external charge pump(s) and any power op amp. The Maxim is essentially making a V- rail and running as an amp would with a dual supply.

The integration is good, but if you prefer to use a different amp you can still do this with another charge pump for dual supplies.
 
The TL084 is a dual rail op-amp but (like all op-amps) it's possible to bias the inputs at half the supply and use AC coupling capacitors so it runs from a single supply.
 
speakerguy79 said:
The Maxim thing is just a charge pump converter + amp in the same package. You can get the same thing with an external charge pump(s) and any power op amp. The Maxim is essentially making a V- rail and running as an amp would with a dual supply.

The integration is good, but if you prefer to use a different amp you can still do this with another charge pump for dual supplies.
This is probably the route I'll take because I already have their 12/15V inverters. I just happened to pick up the TL084 from a surplus vendor hoping it would be single rail so I could keep my 2 inverters. But from the sound of it, TL084's dual nature isn't so bad.

Since the TL084 won't drive the 32Ohms I need, I'll probably take the bipolor transistors I have and get a B or AB amp do all the heavy lifting. Before I can start lowering the crossover distortion though, I need to know what voltage typical headphones are driven to.

What's a typically peak-to-peak voltage for the cheap headphone that might come with a Walkman or CD player? Something like 1, 2, or 3 volts?
 
Last edited:
40mW is pretty loud in headphones. Then the voltage for a 32 ohm load is 1.1V RMS or 3.2V p-p. The peak current is only 50mA but it is too high for an opamp by itself.

A TL074 is a TL084 selected for low noise. The TL074 costs less because it is used more for audio.
 
I don't see what's so good about Maxim's ICs, surely a bridged amplifier would be a more compact solution than a charge pump.
 
Some headphones IC have 3 amplifiers. Left channel, right channel and a DC output at half the supply voltage so output capacitors are not needed.

Bridged amplifiers are not needed and are not wanted because headphones have one common wire from each channel connected together.
 
Thanks audioguru, that's pretty obviously but I didn't realise it.
 
Some people connect stereo headphones with only the two live wires. Then the ears are out-of-phase.
 
Since there are 3 audio sources, I want have a summing op-amp mix the signals and then feed the output into my speaker driver.

The thing is, I want to have the 3 channels isolated from each other to a large extent. The incoming signals are already made for headphones, so I'm okay with less-than-unity gain as long as all my amplifiers are still stable.

Any good pointers for this isolation stage? I was thinking of having each signal go through an emitter follower, then the summing amp, with decoupling capacitors throughout.
 
If you have three stereo sources then you need a dual three-position switch to select one for the headphones stereo amplifier.

You don't want to sum the signals.
 
What I want to do is be able to do listen to 2 or possibly 3 channels simultaneously. Be it phone & PC, phone & television, PC and weather radio, or some other combination. Presently, to do this I have to swap headphones. With this thing, I could answer the phone without missing a beat, or disturbing the caller with lots of background audio.

Yes, my room is a noisy place.
 
Well, yeah. I mean every place I checked suggested the summing amp will be my most direct option. And apart from a few specific resistors, I should have all the parts I need already in storage.

Here's one of the circuits I'm looking at putting on the op-amps output. I wanted to go with a B amplifier. The reference said the diodes will lower crossover distortion if I put enough of a bias current through them through Q3. Does it seem to a reasonable approach?
 

Attachments

  • driver.gif
    driver.gif
    9 KB · Views: 212
DigiTan said:
Here's one of the circuits I'm looking at putting on the op-amps output. I wanted to go with a B amplifier. The reference said the diodes will lower crossover distortion if I put enough of a bias current through them through Q3. Does it seem to a reasonable approach?

No, it looks completely lacking in anything resembling a working amplifier! :(

There are loads of small opamp designs with additional transistors out there on the net, I suggest you try looking at some.
 
The transistor circuit will be very distorted without any negative feedback and without boostrapping.

Your first transistor is turned off so the circuit doesn't do anything. If the signal is fed to the input of first transistor, it is biased and it has a collector load resistor then you have a very poor quality amplifier.

Here is the simulation of a similar circuit except I used an NPN first transistor instead of a PNP. Both would have severe distortion. But it has a high voltage gain.
It might make a good fuzz circuit for a geetar. It would sustain almost all day.
 

Attachments

  • DigiTan's amplifier.PNG
    DigiTan's amplifier.PNG
    24.1 KB · Views: 199
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top