simple shematique needed

[bosse]

hi
i have a pair of output (2sd424 2sb554 ) driver(tip31 tip32) anyone have a good sonding shematic for this transistors. iam lookin for an amp in 40 50 watts range.
thank

[drago]

Here is a schematic from the Toshiba Data Book. It delivers 35 to 80W across 8 ohm depending on the supply voltage +/-31 to 45V. For the input differential pair I used the 2SA979 which is a dual pnp made by Mitsubishi. I built this some years ago and it still works fine.

[AndrewT]

Hi,
the input resistor to ground is shown as 1k. this gives a bass roll off RC = 1mSec. F-3=159Hz.
This resistor should be about 22k to 100k.

Similarly the NFB RC=6.6Hz. This should be set about half to one octave below the input filter.

The output devices could be changed to 2sa1943/c5200 for upto +-50Vdc on the supply rails. At this higher voltage one pair would only be suitable for an easy 8ohm load, but @ +-40V would drive any 8ohm speaker to about 60W.

[drago]

Thanks, AndrewT
You are right, the input resistor should be 51K and the input cap is 4.7uF. Here is the amended circuit with higher resolution.
From the data book there are another two circuits for 120W output. One is BJT and the other FET.(output devices)

[drago]

Hi, AndrewT
I think one pair of transistors working at +/-50V may not survive the high demanding passages of some music especially classical. Because the speakers impedeance can be anything from 3 to 10 ohms at different frequencies. It may not be a safe and reliable amp when using one pair of output transistors. I stuck to two pairs at least.

[AndrewT]

Hi Drago,
here's what I said.

one pair would only be suitable for an easy 8ohm load

That was my polite way of saying that the schematic would not work reliably with 1pair of 150W output transistors.

Then I went on to say that reducing the Vrail to +-40Volts would give a reliable 8ohm capable amplifier using the same single pair of 150W devices.

Your schematic did not show 2pair.

[AndrewT]

Hi Drago,
I see your later schematic shows input RC=240mS & NFB RC=24mS.

This ratio of 10:1 should be 1:1.4 or upto 1:2 for mimimum interaction between the pass signals and feedback.

The input RC will be affected by preceeding equipment and may go down to less than 50mS depending on the source components, but this is still higher than the 24mS of the NFB.

I recommend you alter the NFB to about 100mS to 200mS and then set the input to a smaller value by selecting an appropriate blocking capacitor.

Remember to connect the input end of the blocking cap to ground using a high value resistor, about 470k to 2M2 is suitable. This can be mounted on the input RCA rather than alter the PCB.

[peranders]

Originally posted by drago
Thanks, AndrewT
You are right, the input resistor should be 51K

If you take 15k like the feedback resistor you will reduce the effects of input bias current offset.

[AndrewT]

Hi,
or raise the 15k & 510r to match the 51k. (51k & 1800r with 100uF gives gain =x29.3 +29.3db and RC=180mS)

This gets closer to the NFB RC that is required.

What disadvantages (including noise) are there to raising the feedback loop impedances in a power amp?