Project: "Taurus" hi-end audio amplifier (45W + 45W)

Taurus is an high-end amplifier capable of delivering up to 45W per channel to 4Ohm loads. The project is pretty straightforward: a simple and yet powerful power supply, and both channels built around the LM3875 IC. Although easy to build, this project is far from being recommended to beginners, since it uses the mains. The project eventually can be adapted to work with 115V~: all you need to do is to use a 250VA transformer suitable for 115V~, and a proper primary fuse (the secondary fuses won't change).

List of components:
C1/2/7/8 – 10mF electrolytic capacitor (50V);
C3/5/9/11 – 100µF electrolytic capacitor (50V);
C4/6/10/12 – 100nF polyester capacitor (63V);
C13/14 – 22µF non-polarized electrolytic capacitor (6,3V);
D1-4 – 6A05 rectifier diode;
F1 – 2A fast fuse;
F2/3 - 6,3A slow-blow fuse;
HS1/2 – 1,2°C/W passive heatsink;
IC1/2 – LM3875 audio power amplifier (LM3875TF);
J1 – Earthed AC power plug;
J2/3 – RCA socket;
JW – 26AWG stranded wire;
LP – 230V~ resistored neon bulb;
R1 – 10KR double gang potenciometer (10% tol., 1/8W);
R2/6 – 4,7KR carbon resistor (5% tol., 1/8W);
R3/5/7/9 – 1KR carbon resistor (5% tol., 1/8W);
R4/8 – 33KR carbon resistor (5% tol., 1/8W);
S – DPST switch;
T – Toroidal transformer 230V~ pri. 2x18V~ sec. 250VA.

Electrical parameters:
- Va min = 207V RMS
- Va max = 253V RMS
- Ia (Va = 230V RMS, Z = 4Ohm) = 991.9mA RMS
- Ia (Va = 230V RMS, Z = 8Ohm) = 686.7mA RMS
- P (Va = 230V RMS, Z = 4Ohm) = 228.1W
- P (Va = 230V RMS, Z = 8Ohm) = 157.9W

Amplification characteristics (each channel):
- Input impedance: 3,197-10KOhm
- Input sensitivity: 500mV RMS
- Gain: 30,63dB (34V/V)
- Frequency response (-3dB): 7.234-120000Hz
- Output power (Z = 4Ohm): 45W
- Output power (Z = 8Ohm): 29W

Just a thought:

I was planning to use CL-30 NTC thermistors to avoid a big inrush current through the capacitors, along with a BU1506 rectifier bridge (calculated to withstand the prolonged inrush and the nominal current of 3.12A). But I decided to take the thermistors out, which allowed me to use a discrete rectifier bridge with four 6A05 rectifier diodes instead. The inrush current would heat them up, but keeping them hot would be the problem. That would cause sags on the supply when driving the amplifier after being on and idle for some time (allowing the thermistors to cool down again). For this reason, it is a very bad idea to use thermistors on a amplifier. Also, adding thermistors could cause more harm than good for a couple or reasons besides the one raised by driving the amplifier after being idle for some time:
- The inrush current would be greatly reduced to a couple of amps, but capacitor charging would take much longer (maybe long enough to blow the secondary fuses or perhaps impair the transformer in the absence of these - yes, providing that someone can be dumb enough to the point of using the wrong type of fuses after blowing some fuses blows of the right type). Transformers are quite prepared to take enormous inrush currents of very short duration (and inherent winding resistance will reduce the current to much less than the 400A specified in the 6A05 datasheet), but are definitely not prepared to take relatively small inrush currents above the nominal rating even if for a few seconds. The same happens with slow-blow fuses.
- Thermistors can fail quite dangerously. Turning the power off and then on without allowing them to cool would cause them to fail.

Hello,

Perhaps it is also an idea to translate the PDF into english.
The idea is good overall.
Is there also a PCB for the amplifier planned?

For the ones who are interested in the used IC I attached the datasheet.

Kind regards,
Bertus

bertus said:

Hello,

Perhaps it is also an idea to translate the PDF into english.
The idea is good overall.
Is there also a PCB for the amplifier planned?

For the ones who are interested in the used IC I attached the datasheet.

Kind regards,
Bertus

Click to expand...

Thanks for the suggestion! I'll post the translated PDF tomorrow, or as soon as I can. The PCB is yet to be made, but I'll post it anytime soon.

Just an important notice - This project asks for the LM3875TF version of this chip.

The PDF is now translated and posted, as suggested by Bertus.

F2 and F3 are generally not a good idea; if only one of them blows out, it essentially turns your FWB into a half-wave bridge that continues to power a potentially bad circuit.

An old trick is to place a very low value resistor, hi-wattage, between the xfmr center-tap and ground. That way, during a fault condition, the resistor takes the brundt of the power instead of burning up the more expensive xfmr (until the primary fuse blows).

The resistor between the center tap and the ground will cause the ground to float a bit relative to the input voltage. Even if the resistor is very small, it will cause ground instability and therefore the amps will start motorboating. Also, a small resistor will not protect the transformer.

If, for instance, F2 happens to blow, the increased current will cause F3 to blow too. It F2 happens to blow or to be missing, the circuit will work the same as before, except that the current that the transformer can supply is smaller, and passed that threshold F3 will blow too (precisely because all the needed current is not shared between F2 and F3). As you said, the circuit will work as an half bridge rectifier. This transformer requires 8000mA fuses, and I've specified 6300mA ones, giving even more margin to the transformer.

I though using fuses after the brigde, but a missing or blown fuse there could blow the amplifiers, due to lack of one rail in the presence of the other. Both rails must be present at the same time. Using the fuses before the rectifier, on the AC lines, is the most appropriate way to protect a transformer while keeping the circuit safe. The manufacturer of the toroidal I'm going to be using (HR Diemen) recommends using secondary protection fuses right after the secondary and before the bridge.

In other words, the transformer won't get toasted, even considering that the form factor in case of an half bridge rectifier is higher than the form factor of a full bridge rectifier. Don't forget that if one fuse blows, the current supplied is limited by the other fuse. Secondaries only see current through it, as well as fuses, and that is all that matter.

You could use a momentary switch to apply power, and have a 2 ohm 5W resistor in the line. Cap charging and power supply start up would be 'soft'. The switch also sends unreg. voltage to a latching relay that closes and takes the resistor out of the unreg. DC line.

I'd suggest a few more beefy caps to help the power supply hold through heavy bass transients.

I don't think full regulation would be a huge change, but a little extra punch is always welcome.

The resistor between the center tap and the ground will cause the ground to float a bit relative to the input voltage. Even if the resistor is very small, it will cause ground instability and therefore the amps will start motorboating. Also, a small resistor will not protect the transformer.

Click to expand...

If it's good enough for Bob Heil, it's good enough for me. In fact, that resistor has saved me beau coup bucks and transformers. ...take it, or leave it.

nomurphy said:

If it's good enough for Bob Heil, it's good enough for me. In fact, that resistor has saved me beau coup bucks and transformers. ...take it, or leave it.

Click to expand...

In what circumstances? Will a resistor like that actually limit the current and save the transformer? I don't think so, and all the designs I saw don't use one. The problem you are addressing to is just fictional and it would take a resistor with an appreciable resistance (like 1Ohm) in order to "save" the transformer. 1Ohm in the return path is just too bad. Why wasting money and why bother using chips such as the LM3875 if you are spoiling the design with a resistor?

Plus, such resistor might help to condemn the transformer under sustained fault conditions by protecting the fuses (or the remaining fuse), by working along with the secondary resistance of the transformer.

Really, I rather leave it because fuses, as explained, are more than sufficient for protecting the transformer. If one blows, the other will blow more easily. And don't thing that using only one secondary will cut the VA's to half. The core is the element that actually generates the heat. But even if it does, as said, the secondaries only see current, and they don't bother if they are working alone or not (it will affect the circuit because of the form factor, but not the secondary per se in terms of RMS current that it can supply). The fuse sees the same current, and will blow if it is exceeded.

I think the confusion here is because some people confuse RMS current with average current (or even DC current). Just because an half bridge design with the same transformer VA supplies less DC current than a full brigde design, if doesn't mean that the secondary protection fuses are different. No, they are to be exactly the same for both designs. People, average current is only for rating diodes, because these are non-linear devices. RMS current should be used for rating resistive devices, and that is how fuses are rated (and secondaries too for that matter).

Plus, HD Diemen recommends it!

The PCB artwork is now available! I provided a very solid ground (perhaps a bit overkill for this design) in order to guarantee stability.