Subwoofer amplifier with 2SA747 and 2SC1116

[ObliVion]

Hi!

I had an old DUX Reciever I think it was a Sound Project 1200A. Well it broke and I couldn't figure out what was wrong so I tore it appart. I am building a subwoofer for my workroom and I was hoping somebody would help me with some examples of a mono block for the sub, using the parts I took from the amplifier.

The output transistors are two pairs of 2SA747 and 2SC1116 from SanKen (no longer in production ). I also ripped the powersuppply witch is 35V 6A.

I've searched the net and this group but as I ain't that skilled I don't now if the output transistors will fit any of the designs.

Thank you
ObliVion

[ObliVion]

I was wondering if I could replace the output transistors in something like this and replace the opamp with a discrete one. I want no chips.

/ObliVion

[Stocker]

You are where we all have been: a handful of probably-good parts and a goal.

Wait.

Read.

Read more.

You will be very happy with yourself if you study until you know what is happening inside your circuit. The idea you posted can work, but to me it is more satisfying to design your own stuff from scratch than to just pooge somebody else's idea to work with what you have.

[ObliVion]

Stocker:

I agree very much. It would be much more of an achievement surely.

Then any could links to theory of push-pull output stages, differential amplifiers, and such?

Thank you
ObliVion

[ObliVion]

I have been reading all sorts of stuff from the net. Looked at 50 different schematics and come up with the one attached. It was a lot of work as the BD an MJE transistors were not in the database of my simulation program. I think I got the SPICE models imported the right way. This circuit works in the simulator but there are so many things I don't quite understand. i know the basics of the push pull state and differential state as I have a very limited description of them in a danish book of mine. The only thing I know how to calculate the values for are the current mirror. This boils down to the fact that I have no solid mathematic proof that this will not blow up in reality! Could anyone be so nice as to comment on this circuit and direct my gaze at some sources for information on how to calculate the values of the varies parts invovled.

Thank You
ObliVion

[1audio]

At first glance it looks like Q12 is upside down. The emitter should be toward the load. if your simulator ran then I would look very carefully at the currents and voltages. Also the 1 Ohm resistors in the collectors of the drivers will cook something.

The design it simple and should work for a sub. However its a lot of work to go through not to use an IC.

To calculate the values work back from the peak voltage across the load. this will dictate the peak current (approx) and you can then work out the drive required back to the input.

Optimize the input pair for noise and make sure the gain will work from the beginning to the end.

Overall gain should be between 20-30 dB typically. (THX specifies approx 30 dB for consistancy.)

I would look carefully at using a pair of LM3886's or LM3875's for a sub. Its a lot quicker, but you won't learn as much.

Good luck.

-Demian

[ObliVion]

Well I turned the transistor Ran the simulation and the current on the speaker is 106 amps!!!!!

Somethings surely wrong

/ObliVion

[ObliVion]

Well I have given up on trying to design this myself as I can't do/can't understand/don't know the math. I have been looking at the ESP site. I have decided to build project 03 (ESP Project 03). I was hoping that someone could help me figure out whether or not i can replace the MJE3055/MJE2955 pair with my own 2SA747/2SC1116 pair. As mentioned earlier the 2S? (SanKen) transistors that I have are no longer in production, the only data that I have been able to gather are as follows:

2SA747:
p-n-p
Vmax 120
IcMaxAmp 10.000
PMax W 100.000
hFE 60
fT MHz 15

and

2SC1116:
n-p-n
Vmax 180
IcMaxAmp 10.000
PMax W 100.000
hFE 60
fT Mhz 10

Also I mentioned that I have two pairs and I only need to build a mono block. Would I benifit from using both pairs in parallel. I figure the load will be divided amongst them. But is this the only thing that will happen. Also I am not sure about the voltage of the power supply (Haven't had the time to measure) But it is somewhere between 35-45V. The author state that it can be driven at 45+/- can someone please help me verify this?

Thank you
ObliVion

[Eva]

Your transistors are much better for audio than MJE2955/MJE3055 so you may probably use them instead without trouble

[ObliVion]

I have made this board in EAGLE. Any comments would be very appriciated. The output transistors are mounted off board as you can see.

Thanks

--
Picture removed by request of original poster
Moderator.
--