Mega Simple Class A Amplifier

[diegomj1973]

Hi:

This simple design has all the ingredients to satisfy even the most demanding DIYers. Output power = 4 W at 8 ohms. THD is only 0,017 % at full power and 1 KHz. Voltage gain = 13,85 dB.

Only two stages. Three transistors. Very few components. Its performance is surprisingly good, despite its total simplicity.

High bandwidth. Low distortion. Input and output in single ended configuration. PSRR: 56 dB.

I want to hear comments from those who intend to build this simple amplifier.

regards

R4 is a 50K trimpot. C3 = 3 x 4700 uF. Iq = 1 A. R2 = 2 W. R3 = 1 W. R6 = 4 W.

 

[Calvin]

Hi,

I don't intend to build it, but allow me a comment 😉
You may want to simulate/test not only with a resistive load, but with a complex impedance load too.
Your circuit is similar to a SRPP/modulated current source which may react with oscillations on complex loads.
See if it remains stable.

jauu
Calvin

 

[gumo73]

Stability

Im for the same idea of calvin .
But for 5-10W do you use a modulated current source try
the costant current generator and calculate it for 5 ohm
output load 🙂

 

[JonSnell Electronic]

Firstly use preferred values, what is 37.83k all about?
Is it stable or haven't you tried it yet? When Q2 is turned on, will there be enough current, in the gate, to avoid distortion?

 

[knutn]

I would use a trimmer for R1 and insert a capacitor in series with R3.

 

[diegomj1973]

I have recently designed. I have not implemented yet.

I read about their possible behavior with varying loads. I was unable to test this, yet.

The basic idea is to keep it as simple and small as possible, without impairing the quality characteristics. Including a constant current source instead of the modulated current source, seem to improve the parameters, but I've tried and it has not happened as expected (at least in the simulations).

The value of 37.83 K ohms is the setting that we must carry on a trimpot. The value of the trimpot is 50K ohms.

One could use a trimpot to replace R1 and adjust to the most suitable value.

A capacitor in series with R3: Why add more capacitors, deteriorating feedback, if it is impossible to reduce the asymmetry of the output with the existing modulated current source? C3 serves to preserve the speaker of this asymmetry.

Thank you all for the opinions and suggestions.

Regards

PD: It is very simple and economical... If there are any doubts, it can be built very quickly and you can test their operation.

 

[Ketje]

Still rather simple 😉
Mona

 

[knutn]

A capacitor in series with R3 is to ensure DC stability (increasing the feedback at DC).

 

[diegomj1973]

Still rather simple 😉
Mona

In a simple and smart way, you've eliminated that large output capacitor.

This allows adjustment of offset.

Still only 13 components, but now even cheaper and with great advantages in the frequency response.

It is a very promising circuit.

 

[juma]

...Still only 13 components,...

You'll need gate stoppers too.
Raise of distortion in high frequency region is due to very low current through Q1 - 0.5 mA is too low, 3-5mA will be much better
Also, IRFP240 or similar MOSFETs will work there much better than IRF610.
Sim is a a nice toy but reality is not so forgiving.

Nice thing about that design is that by varying R1 value you can change the amp's character - from single-ended (dominant second harmonic) to push-pull like (dominant third). That's why it's called SEPP...

 

[diegomj1973]

You'll need gate stoppers too.
Raise of distortion in high frequency region is due to very low current through Q1 - 0.5 mA is too low, 3-5mA will be much better
Also, IRFP240 or similar MOSFETs will work there much better than IRF610.
Sim is a a nice toy but reality is not so forgiving.

Nice thing about that design is that by varying R1 value you can change the amp's character - from single-ended (dominant second harmonic) to push-pull like (dominant third). That's why it's called SEPP...

Gate stoppers may be needed, but that we must verify in practice.

The current through Q1 is between 1.8 mA to 2.6 mA (quite far from those 0.5 mA, which you've indicated).
The input capacitance of IRFP240 is much higher than the IRF610. This greatly worsened the situation for the power to be handled.

Thanks for the comments and suggestions.

regards

 

[juma]

...The current through Q1 is between 1.8 mA to 2.6 mA (quite far from those 0.5 mA, which you've indicated)....

Right, my mistake - I didn't look very carefully since I've built that circuit few years ago with 2sk1058 which have far lower Vgs. But still, the amp will sound better with lower impedance drivnig the gates (R1 = 1k or so...)

 

[diegomj1973]

Right, my mistake - I didn't look very carefully since I've built that circuit few years ago with 2sk1058 which have far lower Vgs. But still, the amp will sound better with lower impedance drivnig the gates (R1 = 1k or so...)

The Vgs of 2SK1058 is lower than the IRF610. The 2SK1058 could be more appropriate than the IRF610, in this circuit.

With regard to the value of R1: the circuit responds much better at high frequencies with low values ​​of R1. There is usually an optimum value, depending on the type of MOSFET (input capacitance, mainly). If the value of R1 falls far below the optimum, distortion starts to grow again (the gain of the first stage falls).

regards

 

[diegomj1973]

After some tests, I decided to build the first scheme (post 1). Some changes were needed. IRFP150N instead of IRF610; C3 is 4700 uF instead of 14100 uF. R4 is 100 K trimpot instead of 50 K trimpot. I bías = 1 A.

Here a very primitive test:

https://youtu.be/NONgRPirdkQ

https://youtu.be/XjHzq29cKoY

https://youtu.be/74HGSQSAoTo

Sorry for the bad cam and sound.

Regards

 

[jerluwoo]

1st and 3rd vids are blocked due to copyright issues.

 

[diegomj1973]

1st and 3rd vids are blocked due to copyright issues.

Sorry