Simplest 10W mono audio amplifier?

Status
Not open for further replies.

neazoi

Advanced Member level 6
Joined
Jan 5, 2008
Messages
4,157
Helped
13
Reputation
26
Reaction score
15
Trophy points
1,318
Location
Greece
www.microwave.gr
Activity points
37,198
Hi I would like to build a small monophonic audio amplifier that will take audio from the cd player and will be able to drive a transformer, which in turn will AM modulate the anode of a small 8W transmitter.

Since audio quality is not too important I wonder if a class-C amplifier is ok, or it has to be AB?

I am looking for the simplest circuit (in the world) that can do it.
I guess I should look for these IC's but I have no idea.
The less external components the better.

- - - Updated - - -

TDA7360 seems good and no output caps at all. The onl;y thing I do not like if the 18V max input voltage. I would prefer it to be 28v or a bit higher.
any other one?
 
Last edited:

The supply voltage is almost irrelevant, use whatever audio amp you have available because the transformer will isolate the audio from the anode circuit anyway. Forget class-C audio! That is, if you want it to be recognized as voice or music. Class-C output will be a worse than half-wave rectified version of the input, distortion will be close to 100%!

You can use Class-C for the RF as it doesn't really matter what wave shape the carrier is as long as you use a good LPF to remove all the harmonics it creates.

Don't forget that for a single supply audio amp driving a transformer you either have to use a bridged output stage or a coupling capacitor.

Brian.
 
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
The supply voltage is irrelevant, but I want to power the transmitter and the audio amplifier from the same PSU. Obviously making a voltage divider for the audio amplifier using power resistors will cause a lot of heat.

The TDA7360 datasheet has a bridged schematic for mono. I believe I could replace the speaker with a 4R primary of a modulator transformer, what do you think?

Any guess about the secondary resistance for AM modulation?
 

Your thread reminded me of AM radio transmitters I worked with, used to page doctors in a hospital 50 years ago. The power supply voltage for the RF output tube was modulated.
A doctor was paged in Toronto, Canada and a doctor responded in Buffalo, New York, USA pretty far away across Lake Ontario.
 


Yes, the power supply voltage is modulated. The output power of the single stage transistorized power oscillator depends on the input voltage. So I thought a simple modulator transformer in series can vary the amplitude of the output signal modulated by the audio. It worked nicely using an external amplifier, but I wanted to embed the modulator inside the transmitter, so a cheap and dirty circuit was needed.
I have not found any simpler than the TDA7360 (in bridged mode), it does not even require output capacitors, which would make the circuit big and expensive (yes the TDA7360 cost a few euros, but the output caps would cost much more).

The only thing I do not like if the 18v max voltage, I would prefer it to be up to 30v or so, so that I can drive it with the same voltage as the transmitter.

UPDATE
Now that I see it the TDA7240 or TDA7241 seems to have a 28V supply voltage, am I reading the datasheet right?

- - - Updated - - -

What about this LM3886? does it need an output capacitor?
 
Last edited:

ALL single ended, single supply amplifiers need an output capacitor because their output pins have half supply voltage on them. The options are to use both positive and negative supplies so the mid-rail voltage is zero, or to use two amplifiers in bridge configuration so the difference between their outputs is zero. A bridge configuration sounds best for your application because it doesn't need a capacitor and also delivers twice the effective output voltage (= 4 times the power).

I'm a bit worried that you first mentioned anode modulation, then a supply voltage too low for a vacuum tube, then a 'power oscillator'. Which is it? If you are modulating a single stage oscillator you will produce all sorts of unwanted FM effects and you will have to keep the modulation level low. At 100% modulation, the oscillator would have no power so it would be stopping and starting in sympathy with the modulation.

Brian.
 
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating

I cannot give the link directly, but if you go to my page **broken link removed** and search for "easy40", this is the oscillator I modulate.
I have tested it with a small 8R to 120R transformer and my home hifi amplifier and the volume on the AM radio is very accepted. However as I increase the VCC more AF amplification is needed to modulate it, I think this is expected.

With the 120R winding in series the oscillator can start at 3.3v. Without it it can start at 2.5v. Max voltage is 29v or so. So I think I won't have significant cut-off problems.
There might be more effective ways that this can be modulated and since the output voltage is max 29V a transformer might not be required. If an active voltage divider could be used that has a resistor and an active device (driven by audio) one could regulate the voltage to the transmitter directly without a transformer. I think of a power mosfet that coule be directly driven by low volume audio and this could eliminate the need for an amplifier and a transformer? Eliminating the transformer is of course highly desirable, if it happens to know such a technique please let me know.
 

chuckey said:
Transformer modulation means that the RF PA, runs with its own Vcc. Using a series transistor, the RF PA must run with 1/2 Vcc, so the modulation can swing it from Vcc to 0v (almost).
Frank
Click to expand...
Thank you Frank. I just saw the series transistor AM modulator scheme. It was also used to AM modulate some TV hobby transmitters.
So the use of the transformer eliminates this problem of reduced average output power?

Previously I proposed another scheme with a Mosfet to be used as a variable resistor, in one of the two elements of a voltage divider (the other being a power resistor). Could this scheme avoid this problem you mention with the series BJT?
I propose a mosfet because I suspect it could be driven by fery low voltage signals and vary high current DC, so that no big audio amplifier is needed. Just a guess.
See for example this




He claims that 0.5W can modulate 10W AM. This is a series circuit though not a voltage divider. I have no idea how this circuit works but it looks interesting!
 
Last edited by a moderator:

Because of the problem matching the Rf PA collector impedance to the amplifier, you will need something like a 3:1 AUDIO transformer able to run with DC through its secondary and rated at 10W. You will never find one. Its best to use choke modulation. Connect a LF choke in series with the PA feed and its Vcc, feed your amplifier to the junction via a suitable capacitor. If you feed a current of about 1/40 of the RF PA's current through the primary in the correct phase, it will cancel the flux caused by the PA's DC current, so leading to lower distortion.
Frank
 


Where shall the capacitor be connected? at the point where the choke is connected to the collector of the transistor RF amplifier, or at the VCC point before the choke?

So in that way you saturate the choke with the signal from the power amplifier?
 

That design looks 'dodgy'. No bias on the third transistor. Poor matching to the PA MOSFET and no DC isolation at the modulation input which has about 24V DC on it!
Are they really using a PC fan as a voltage stabilizer???
How can 0.5W be quoted as the audio input when the impedance can be anything from a direct short to the supply line or ~1K in mid control setting?

Brian.
 
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating

I have no idea. The circuit looks too "suspicious" to me to work at all, as well.
I was wondering if somehow one can exploit the low gate loading of a mosfet to make a powerful modulator in a single stage, that can be driven from a low level audio signal.
The way I think it, the small audio signal variations at the mosfet gate should be able to somehow make it off or saturate it for good.
My experiemce is so limited about mosfets though.

I have seen the choke way proposed in the previous post in an old tube transmitter, but never applied to a transistor final stage using a coupling capacitor.
 

The modulation is applied to the PA end of the choke. The whole point of putting a small DC current through the primary is to reduce the core saturation.
I think the "5W" audio input in that circuit , should be 5V. There should be some bias around the audio FETs to get them to drop around 1/2 Vcc.
Frank
 


What kind of choke would you use in your proposition Frank?
Does it have to be an big and bulky audio amplifier choke, or can it be a high permeability small core like the 50-75-J.
 

The problem with AM is you need to control the supply voltage to control the output power, in other words the RF supply has to follow the audio level. As with audio amplifiers you have positive and negative going 'swings' in the voltage (think of it like the push and pull of a speaker cone) and to represent that as a varying RF level you have to start at half carrier level. That gives it the ability to increase toward full power or decrease toward zero in sympathy with the audio.

With a 'current sinking' modulator such as the LF choke in the supply, you have to pass enough current down the modulator path to be able to drop the supply to about half voltage, it is therefore relatively inefficient. A transformer coupled modulator can decrease and increase the supply voltage so you probably want to run it from 12V so it can potentially go from 24V peak to zero volts. The design with the MOSFETs attempts to do it by using one MOSFET as a controlled current source and the other as a bypass around it. Even if it had capacitive coupling at the input the modulation level would be quite low. It would make more sense to apply the modulation to a single MOSFET, in other words insert the audio at the wiper of the potentiometer and remove the left MOSFET completely. If you try that, bear in mind the audio level will vary with the potentiometer setting and importantly, note that the audio input is at +24V so be careful of the polarity of the coupling capacitor.

Brian.
 


No wonder why these old AM pirates almost always use a modulation transformer connected at the anode of the tube.
So on my 8W transistorized transmitter, if I use a modulation transformer, do I have to set the unmodulated output carrier power at half the power (4W), assumming 100% modulation?

Also, what would you suggest as a starting point for the transformer?
For example 8R at the audio amplifier connection side and 1K secondary in series with the collector?
 

Sorry wrong! Think of a standard transformer. Lets make its ratio 1:1, so at both ends of the non Vcc ends of the windings , you have the same audio voltage. Now if the Vccs are the same, you have choke modulation, but with two windings in parallel. There is power stored in the magnetic field of the choke, Q = 1/2 L i^2. the value of the choke must be such that at your lowest modulation frequency, its reactance is equal to the PA impedance. So for communications, 300 HZ would do, PA impedance say 28V @.5A = 56 ohms, so Xl = 56 ohms @300 HZ. Or L = 56/300 X 2 X PI, about 30 mH.
Frank
 


I am sorry to bring this topic back into life, but I was wondering that the LM317 that I am going to use as the power supply for the transmitter could be possibly used as a modulator?
I have seen the LM317 used as a modulator in gunn diodes, although the modulated output voltage is used to FM modulate the diode. **broken link removed**
But in my case it could be used to AM modulate the transmitter.
Would that be possible to be done with the LM317 instead of using a power audio amplifier and a transformer?
 
Last edited:

I've seen articles about making an LM317 into a power amplifier at audio frequencies. So your idea is an atypical role, although not totally farfetched. You'll have to test to find out how it behaves at higher frequencies.
 
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…