Making an Amplifier in order to better understand transistors

Thread Starter

BrainFog

Joined Jan 24, 2011
122
I am currently learning about electronics in my spare time and one thing I am trying to get a better understanding of is transistors. My plan is to build an amplifier using transistors, most likely just using pnp or npn, that will be good enough for me to want to use.

The schematic I am thinking of using: http://www.zen22142.zen.co.uk/Circuits/Audio/headphone.htm what do you think of the design?

This will also be a good opportunity to design a dual rail power supply. I bought a transformer that can be wired up to work as a dual rail a while ago. Would I be able to use a zener diode controlling a transistor, can't remember the technical term for this, or would this cause to much noise? I have read that amps are insanely sensitive to distortions in a power supply.

Would it be best if I use metal film resistors or will carbon ones suffice?

I know that the wattage of an amplifier does not refer to the power drain at the plug, how would I find out how much power the above schematic would require?

This may seem daft but I am new to this, what does +Vcc mean? I read that symbols like it should only be connected to identical symbols but it is alone on the schematic.

Thank you
 

Thread Starter

BrainFog

Joined Jan 24, 2011
122
Thanks

Oops I seem to have mixed this one up with one that uses a dual rail power power supply, that is embarrassing. I remember reading that half the circuits of an amp use half the voltage of the other so I would imagine dual power supplies would be perfect as you can easily access full or half voltage.

I find it difficult to read large blocks of text on the computer screen so I never managed to effectively get through this sites guides.
 

Audioguru

Joined Dec 20, 2007
11,248
The headphones amplifier circuit you found might be 50 years old. Back then maybe "medium impedance" headphones were used. Today Sennheiser still make them but they are extremely expensive. That old circuit will not drive modern low cost but good quality 32 ohm headphones.

In an amplifier circuit, metal film resistors cause less hiss than carbon film resistors.
 

Thread Starter

BrainFog

Joined Jan 24, 2011
122
So carbon it is.

I chose this amp partly because it had similarities to a design from one of my electronics books. If I remember correctly the first transistor boosts voltage and the second boosts current. And because it was complex but not too complex for a beginner.

Do you have any recommendations for how I would be able to boost the power if it lacks the power I was hoping for? It will be interesting to see how small modifications to the circuit can make changes to its outputs.
 

Audioguru

Joined Dec 20, 2007
11,248
So carbon it is.
Read about the hiss a carbon film resistor causes. A metal film resistor causes less hiss.

I chose this amp partly because it had similarities to a design from one of my electronics books. If I remember correctly the first transistor boosts voltage and the second boosts current. And because it was complex but not too complex for a beginner.
Yes, it is an extremely simple circuit.

Do you have any recommendations for how I would be able to boost the power if it lacks the power I was hoping for? It will be interesting to see how small modifications to the circuit can make changes to its outputs.
The circuit you found is not suitable to drive modern low impedance headphones. Most headphones amplifiers today use an opamp or an opamp that drives transistors.
First you must look at how much impedance your headphones have (32 ohms?) then find an amplifier circuit that will drive them.
 

Thread Starter

BrainFog

Joined Jan 24, 2011
122
Thanks for the information. I have been reading a bit about speakers and impedance although it seems to suffer from a few ambiguities.

I have found conflicting information. One site states that you can use equal or higher impedance speakers on an amp of a set impedance level however I have found another that states the opposite. It seems as though it is best to match the levels though.

The project is for no specific purpose other than understanding the technology, but I was considering using it for either headphones or speakers. After not finding any marking stating ohms I measured the headphones I use at just over 16ohm.

I am really interested from a design point of view how the schematic could be modified to boost or lower impedance.
 

Audioguru

Joined Dec 20, 2007
11,248
I have found conflicting information. One site states that you can use equal or higher impedance speakers on an amp of a set impedance level however I have found another that states the opposite. It seems as though it is best to match the levels though.
Old fashioned vacuum tube amplifiers matched their output impedance to a speaker's impedance with taps on their output transformer.
But modern solid state amplifiers do not have an output transformer. They have an extremely low output impedance of 0.04 ohms or less so they certainly do not match the impedance of a speaker. The minimum impedance they can drive is determined by the resulting max current and the resulting heat in the output transistors.
Most amplifiers today can drive 4 ohm or 8 ohm speakers. Usually the power is more in 4 ohms than in 8 ohms because the current is higher.

The project is for no specific purpose other than understanding the technology, but I was considering using it for either headphones or speakers. After not finding any marking stating ohms I measured the headphones I use at just over 16ohm.
I think the circuit you found cannot drive 16 ohm headphones. You need a little power amplifier to do it.

I am really interested from a design point of view how the schematic could be modified to boost or lower impedance.
You need a power amplifier circuit, not just a common emitter transistor driving an emitter-follower.

I simulated the circuit with American transistors that are almost the same as the European ones to see if it will work. It won't. As I suspected when the output is "shorted" by the very low impedance of your 16 ohm headphone then the bottom half of the waveform became so distorted that it nearly disappeared.

Rectified audio sounds awful.

EDIT: I nearly forgot to say that the polarity of the input capacitor in the original project is shown backwards.
 

Attachments

Thread Starter

BrainFog

Joined Jan 24, 2011
122
Thank you for your reply, I would have replied sooner but my exams are coming up.

I recently downloaded the software you used in the screenshot to simulate the circuit, LTspice IV, as I wish to have a pay with the circuit so to speak. Sadly It did not go well. What settings did you use for the sine wave? Also how did you set up the run settings? I have been overwhelmed by unfamiliar options and the guides I have found were ambiguous at best.

Hehe rectified sound would not be great would it.

That is interesting, I did not realise there were differences between American and European transistors. I know that different companies produce slightly different versions of the same item, hmmmm, is it to do with different standardisations?

Maybe I have read things incorrectly but very low impedance would be 1 or 2 ohm and 16 or 32 ohm be classed as high impedance for speakers? I suspect that I am too closely comparing them to resistors.

Thanks
 

Audioguru

Joined Dec 20, 2007
11,248
TLspice is American so it has only a few European transistors in its library. I selected American transistors that have nearly the same spec's as the European transistors in the original circuit. For transistors in a TO-92 case the emitter and collector are reversed across the Atlantic ocean.
The settings of the input is shown under the input. 100mV peak level and 3kHz (3000) frequency. I leave it at 3kHz for most of my simulations.

40 years ago headphones were 8 ohms per ear but I don't think they are available today. Today they are 32 ohms and 600 ohms.

As I showed, the circuit you found is garbage and does not work.
 

CDRIVE

Joined Jul 1, 2008
2,219
Back in the day I remember earphones being much higher Z, in the order of ~1K. I still have a drawer full of the single earplug type that was sold with pocket transistor radios. I have a set of military headphones I acquired in the 60's. They were surplus when I bought them, so they may have been 40's or 50's era. One of these days I'll dig them out but I remember them being 600 to 1K also. Either way I don't think it's proper or correct to call that man's circuit garbage. It's suitable for which it was designed.

A 40 ton crane is not garbage because it can't lift 100 tons.
 

Audioguru

Joined Dec 20, 2007
11,248
I don't think it's proper or correct to call that man's circuit garbage. It's suitable for which it was designed.
BrainFog posted the very old headphones amplifier circuit and said his headphones measure 16 ohms.
I showed in a simulation that it does not work with a load as low as 16 ohms so the circuit is garbage for his application.
 

CDRIVE

Joined Jul 1, 2008
2,219
BrainFog posted the very old headphones amplifier circuit and said his headphones measure 16 ohms.
I showed in a simulation that it does not work with a load as low as 16 ohms so the circuit is garbage for his application.
If you think happy thoughts your statement can be politely rephrased as.. "Your circuit is unsuitable for low impedance headphones". ;)
 

Audioguru

Joined Dec 20, 2007
11,248
Nothing an audio transformer can't fix. <ducking>
No.
A high input impedance to low output impedance transformer has a fairly high voltage loss. A headphones amplifier circuit usually has voltage gain.
This old circuit has a voltage gain of 10 for high impedance headphones.
 

CDRIVE

Joined Jul 1, 2008
2,219
No.
A high input impedance to low output impedance transformer has a fairly high voltage loss. A headphones amplifier circuit usually has voltage gain.
This old circuit has a voltage gain of 10 for high impedance headphones.
This is the first time I've ever read that the turns ratio or input/output Z had anything (worth mentioning) to do with a transformer's efficiency. Are you referring to resistive losses? In terms of mW or dB loss, what are we talking about here. I know you said "Voltage loss" but power is really whats relevant here when dealing with a 30 ohm load.
 

Audioguru

Joined Dec 20, 2007
11,248
A transformer has a very low loss (I should have said it has a Voltage Stepdown instead of a loss) therefore its efficiency is very high.
A transformer's input power is equal to the output power.

1V into 1k ohms is 1.0mW.
1.0mW into 8 ohms is 0.0894V.
1V/0.0894V= 11.186 times.
11.186 times squared is 125.
1k ohms/8 ohms= 125 (same as the voltage stepup squared).

Somebody said their 32 ohm headphones are rated at 110dB at 100mW. So 90dB is made with a power of 1mW. Then a 1k/8 ohm transformer will produce 90dB in 32 ohm headphones if the input is 1V into 4k ohms. But many people like the peak sound of music to be 100dB so the input should be 3.2V RMS which is much higher than line level.
 

CDRIVE

Joined Jul 1, 2008
2,219
Somebody said their 32 ohm headphones are rated at 110dB at 100mW. So 90dB is made with a power of 1mW. Then a 1k/8 ohm transformer will produce 90dB in 32 ohm headphones if the input is 1V into 4k ohms. But many people like the peak sound of music to be 100dB so the input should be 3.2V RMS which is much higher than line level.
That's funny AG. I've seen some of the kids walking around with those things cranked up so loud that what little they had between their ears was being turned to mush. :D
 

Audioguru

Joined Dec 20, 2007
11,248
That's funny AG. I've seen some of the kids walking around with those things cranked up so loud that what little they had between their ears was being turned to mush. :D
I know.
When I am walking in a mall and a group of kids are 30m away I can hear their earbuds.
 

Thread Starter

BrainFog

Joined Jan 24, 2011
122
Some interesting banter. :p

I will pretend to read "Your circuit is unsuitable for low impedance headphones" rather than "garbage".

One thing that has clearly been forgotten is that the purpose of this project is to gain a better understanding of transistors. Although I have a strong understanding of how an individual transistor works I am having trouble grasping how they work as a circuit. One example is how do they turn DC into AC. If I understand how the circuit works, even if it does not perform the function correctly I will be happy. Maybe that was a bad way of saying knowledge building experiment.

I am beginning to look at this circuit http://electronics-diy.com/4x4.php Just another simple low current headphone amp. I was wondering if you could confirm or deny if some of my analysis of the circuit.

Capacitors are used to remove unwanted frequencies. P1 is volume and simply acts as a voltage divider. R1, R2, D1 and D2 are used to supply a bias current so the transistors work at any voltage rather than only activating at above 0.7v, I assume the diodes are in place as a voltage drop but not sure. R3 may be another voltage divider. As for the placement of Q2 and Q3 I simply cant see the logic in their placement, I cant work out how energy flows through this circuit to give a boosted output.

I would be grateful for any help you can offer. Thanks
 
Top