Extremely simple amplifier (4 transistors) with low THD.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello friends,
Today I tested an extremely simple project (4 transistors) developed and produced by a friend of mine from Brazil.

His website is this: AlfaKits Eletrônica

The cool thing is that on his site is shared free of charge the electrical scheme and layout of the PCB for those who want to produce for their own use. Production for sale is prohibited.

I was impressed by the level of THD he achieved with a simple 12 to 15V power supply and only 4 transistors.

Check out the results:

I couldn't post the photos. Following is the link to the results:
RB Eletronica - UNO


I'll publish on my YT channel the details about that test, soon.
YouTube
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
Neat design, reminiscent of a singleton input cap-coupled output amps that I have seen around here on DIYA.

attachment.php
 

Attachments

  • Uno-Amp-Schematic.jpg
    Uno-Amp-Schematic.jpg
    85.1 KB · Views: 1,849
No emitter resistors, and a single diode as a base spreader as a consequence, to limit the risk of thermal runaway is a good recipe for Xover distortion.
It may give more "edge" to loud music, but for softer programs it could be somewhat unpleasant.
In addition, the speaker will see a permanent 30~40mA DC current.
Not catastrophic for a big 100W or so unit, but not ideal for a 5 or 10W one
 
I also had a ZX spectrum, and IIRC the OP stage was even cruder, something like a single, binary-controlled transistor.

The THD figure of the 4-transistor amp might be relatively low (hard data please?), but Xover distortion does generally not have a large impact on the global THD figure, it is simply extremely unpleasant, even at very low numerical levels.
An oscilloscope or an ear is a better detection tool than a THD meter in this case
 
Founder of XSA-Labs
Joined 2012
Paid Member
I ran a quick simulation using commonly available actives for the typical roles here.
attachment.php


Increased the operating voltage to 24v and got this:

For 8Vpp (2.83vrms) into 8ohms for 1w:
Code:
Harmonic	Frequency	 Fourier 	Normalized	 Phase  	Normalized
 Number 	  [Hz]   	Component	 Component	[degree]	Phase [deg]
    1   	1.000e+03	4.087e+00	1.000e+00	    2.16°	    0.00°
    2   	2.000e+03	1.525e-03	3.732e-04	  113.93°	  111.77°
    3   	3.000e+03	2.855e-04	6.987e-05	  178.73°	  176.57°
    4   	4.000e+03	2.991e-04	7.319e-05	  178.34°	  176.19°
    5   	5.000e+03	2.397e-04	5.866e-05	 -179.55°	 -181.70°
    6   	6.000e+03	1.984e-04	4.855e-05	 -179.62°	 -181.77°
    7   	7.000e+03	1.701e-04	4.162e-05	 -179.72°	 -181.88°
    8   	8.000e+03	1.488e-04	3.642e-05	 -179.75°	 -181.91°
    9   	9.000e+03	1.323e-04	3.237e-05	 -179.78°	 -181.94°
   10   	1.000e+04	1.191e-04	2.914e-05	 -179.80°	 -181.96°
   11   	1.100e+04	1.083e-04	2.649e-05	 -179.82°	 -181.98°
   12   	1.200e+04	9.923e-05	2.428e-05	 -179.84°	 -181.99°
   13   	1.300e+04	9.160e-05	2.241e-05	 -179.85°	 -182.00°
   14   	1.400e+04	8.505e-05	2.081e-05	 -179.86°	 -182.01°
   15   	1.500e+04	7.938e-05	1.942e-05	 -179.87°	 -182.02°
Total Harmonic Distortion: 0.040360%(0.292814%)

It has 2nd order dominant distortion with a monotonically decreasing profile. So should sound good. I do not see it as particularly low THD though, but that is to be expected of a 4-active singleton amp:
attachment.php
 

Attachments

  • Uno-Amp-LTspice-v1.jpg
    Uno-Amp-LTspice-v1.jpg
    201.5 KB · Views: 1,125
  • Uno-Amp-LTspice-v1-FFT.jpg
    Uno-Amp-LTspice-v1-FFT.jpg
    185.3 KB · Views: 1,146
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
It is able to make 7w into 8ohms with fairly low distortion:
Code:
Harmonic	Frequency	 Fourier 	Normalized	 Phase  	Normalized
 Number 	  [Hz]   	Component	 Component	[degree]	Phase [deg]
    1   	1.000e+03	1.075e+01	1.000e+00	    2.16°	    0.00°
    2   	2.000e+03	9.502e-03	8.841e-04	  100.52°	   98.37°
    3   	3.000e+03	2.996e-03	2.788e-04	    7.43°	    5.27°
    4   	4.000e+03	1.224e-03	1.139e-04	 -122.71°	 -124.87°
    5   	5.000e+03	2.125e-03	1.977e-04	 -173.41°	 -175.57°
    6   	6.000e+03	1.677e-03	1.560e-04	  118.78°	  116.62°
    7   	7.000e+03	9.150e-04	8.514e-05	   18.63°	   16.48°
    8   	8.000e+03	1.325e-03	1.233e-04	  -92.76°	  -94.91°
    9   	9.000e+03	1.647e-03	1.532e-04	 -167.37°	 -169.53°
   10   	1.000e+04	1.374e-03	1.278e-04	  120.45°	  118.30°
   11   	1.100e+04	9.276e-04	8.632e-05	   25.36°	   23.20°
   12   	1.200e+04	1.068e-03	9.935e-05	  -82.19°	  -84.34°
   13   	1.300e+04	1.302e-03	1.212e-04	 -161.21°	 -163.36°
   14   	1.400e+04	1.126e-03	1.048e-04	  125.30°	  123.14°
   15   	1.500e+04	7.646e-04	7.114e-05	   33.56°	   31.40°
Total Harmonic Distortion: 0.102280%(0.309343%)

I should add that this is a fairly high bias current of 640mA (sort of Class A) and dissipating about 8w per output active.
 
Amps like that were abundant in 70's, 80's. I've built shitloads of small amps with above topology back then. The ones with germ. outputs (a popular pair was AD161/162) were particularly "good" sounding (had a really good presentation of new-vawe and punk guitars - which used to be popular then). Another popular Si pair for car-radios was BD433/434.

If made with a darlington input (or a bit more heavily bootstrapped than above amp); it can be fed directly from a ceramic cartridge record player (audiophiles beware!)
 
Amps like that were abundant in 70's, 80's.
The originals were generally much better than this example: here is a large collection from various sources:
A collection of vintage, single supply, low to medium power amplifiers for Daniel

This one from a Motorola appnote probably served as a model:

741853d1552217915-collection-vintage-single-supply-low-medium-power-amplifiers-daniel-dan17-jpg


I have also described a number of evolutions based on the same theme, and they really bring novel features and good performance.

Otherwise, stick with the good old original circuits: they are still current and usable, unlike some pale copies
 
I ran a quick simulation using commonly available actives for the typical roles here.
782328d1568958186-extremely-simple-amplifier-4-transistors-low-thd-uno-amp-ltspice-v1-jpg


Increased the operating voltage to 24v and got this:

For 8Vpp (2.83vrms) into 8ohms for 1w:
Code:
Harmonic	Frequency	 Fourier 	Normalized	 Phase  	Normalized
 Number 	  [Hz]   	Component	 Component	[degree]	Phase [deg]
    1   	1.000e+03	4.087e+00	1.000e+00	    2.16°	    0.00°
    2   	2.000e+03	1.525e-03	3.732e-04	  113.93°	  111.77°
    3   	3.000e+03	2.855e-04	6.987e-05	  178.73°	  176.57°
    4   	4.000e+03	2.991e-04	7.319e-05	  178.34°	  176.19°
    5   	5.000e+03	2.397e-04	5.866e-05	 -179.55°	 -181.70°
    6   	6.000e+03	1.984e-04	4.855e-05	 -179.62°	 -181.77°
    7   	7.000e+03	1.701e-04	4.162e-05	 -179.72°	 -181.88°
    8   	8.000e+03	1.488e-04	3.642e-05	 -179.75°	 -181.91°
    9   	9.000e+03	1.323e-04	3.237e-05	 -179.78°	 -181.94°
   10   	1.000e+04	1.191e-04	2.914e-05	 -179.80°	 -181.96°
   11   	1.100e+04	1.083e-04	2.649e-05	 -179.82°	 -181.98°
   12   	1.200e+04	9.923e-05	2.428e-05	 -179.84°	 -181.99°
   13   	1.300e+04	9.160e-05	2.241e-05	 -179.85°	 -182.00°
   14   	1.400e+04	8.505e-05	2.081e-05	 -179.86°	 -182.01°
   15   	1.500e+04	7.938e-05	1.942e-05	 -179.87°	 -182.02°
Total Harmonic Distortion: 0.040360%(0.292814%)

It has 2nd order dominant distortion with a monotonically decreasing profile. So should sound good. I do not see it as particularly low THD though, but that is to be expected of a 4-active singleton amp:
782329d1568958277-extremely-simple-amplifier-4-transistors-low-thd-uno-amp-ltspice-v1-fft-jpg

What about AC analysis ??
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.