Output stage power

As usual for any naïve question, you need more information - how much voltage you are using, whether you want AB or only A, how much drive you have for the output stage, and how much heat you can dissipate. You also need to recognise that to drive 4R, regardless of low power, you will need global feedback and this entails a complete front to back design. You know enough to get yourself into trouble, but not sufficient to describe the complete problem. Look up Dunning Kruger.

Many people here have this experience, but you cannot demand this information as it's hard won and many do not have the energy or inclination to help. There is nothing in it for them...... and it's not helped by your impatience and sense of entitlement.


Paid Member
2007-09-15 8:14 am

OK, guys that's enough. We were all beginners once over... geez, when I think of some of the things that had me stumped way back yonder.

gorge. Patience :)

As AKSA has hinted at, you need to give a bit more information on what you are trying to do. So it sounds like you want to make a small headphone amp. So the first things we need to know are what power supply you have available eg mains or battery. If battery, what size and voltage.

And to answer your first question, yes, BC547 type devices will make a headphone amp that will blow your ears off.
Hi Guys

At the risk of helping Gorge:

For the low power you require, you can approach things with transistors or with ICs (integrated circuits). 20mA at 4R is only 1.6mW at 80mV swing. You could use a 5532 opamp to drive both channels. Set it up for unity gain, or whatever small gain is required. This would be the most straight-forward solution and will give good results.

To go to discrete circuitry, it is more interesting for sure and has many more options. You can build things that use feedback or not depending on how much voltage gain is required. If only a buffer is needed you can build a diamond circuit or other type of voltage follower. The diamond as a follower has the advantage of no voltage loss, where a simple complementary BJT EF has a slight voltage loss from input to output. Depending on how much signal is coming in, this loss may not be important.

The discrete simple options will not have as low THD as the 5532 wired for unity or low gain and will take up more room.

If you use the 5532, remember to tack a 100nF ceramic on the bottom side of the socket between the supply pins (pin-4 to pin-8).

Have fun
Gorge deserves an apology

Hi Guys

Again with the chastising?

Gorge deserves apologies from AKSA for certain, as it was he who misinterpreted an entirely polite and innocent question and started the "kicking a guy while he's down" festival. It is no surprise that the mercurial Fahey jumped right in. Then pe1mnk and mcd99...

That is the most vulgar of human behaviours.

This forum claims to have over 100,000 members. It is usual that a new thread has a reply within a very short number of minutes let alone hours, so is it surprising that Gorge asked a follow-up question? His first question was polite and had enough information in it for anyone of modest skill to make an intelligent reply. His second Q is a wonderment if he mis-stated things, since it is obvious English is a second language for him, or if maybe people on this forum only build higher-wattage things? Perfectly reasonable questions.

Is there some bad phase of the moon? an unfavourable conjunction of planets?

I find it embarrassing that a new member should be jumped on for absolutely no reason.

To Gorge: If you go discreet with the circuitry, the BC547/557 pair can be used for your output stage provided you keep their dissipation within the 625mW rating.

Have fun
I am an embedded software engineer. I am not familiar in analog circuits (accepting the truth). but I know soldering and circuit rigging. I am here only because I love and enjoy music . by reading this forum I came to know about legends of audio and I have great respect for them. but the problem is when junior member or technically weak people .. if they want to learn from teacher or guru there is way to ask question. only this part is missing . I respect my teachers and guru's who thought me. if my post hurts any one i apologize to them.


2016-04-27 6:59 am
So I've come to learn that the more current flowing through your output stage transistors, the lower the output impedance will be. I want to make something similar to a darlington output stage and put the largest current through them that is safe and using the largest transistors I can find. Power loss is not an issue. I just want to see for experimentation purposes, just how low the output impedance will become.
Does Gorge need to know the output impedance?
What would he do with that information?

Surely all he needs once he tells us what he is trying to do is give him a circuit/s that can drive his load.

Give us a brief. A short list of targets that your project MUST meet to be able to do it's job.

You already have some clues to what will be in the brief.
Battery or mains supply, Integrated circuits, or discrete transistors, size to fit your girl-friend's handbag, or on a table top next to your tablet/PC. What is the load - Headphones?


2006-07-28 4:35 pm
Hi Gorge, Do the ear buds that you're using really have a load impedance of 4 Ohms?

If so, that's really going to drive your design goals. For a given output level/volume, the load impedance will determine the amount of output device dissipation and current capability and thermal management that you'll need. I did a quick search on "Apple Earbud impedance" and got back a number of 42 & 45 Ohms. Of course I don't know what model you have, so your mileage will vary. But if you want ease of implementation and longer battery life, it might be a good option to select a higher impedance set of ear buds.

Many Op-amps would have no issue driving >40 Ohms directly.