Simple class B or AB amp.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've been trying to use up some of the copper-clad board scraps, recycled parts etc. I have lying around, so I put together something I could use as a subwoofer amplifier. The schematic is attached. I don't know if anyone will find it interesting or useful.

It's an op-amp driving a normal emitter-follower output stage, except that bias is applied via two current mirrors.

The output devices were supposed to be MOSFETs (ECF20N16/ECF20P16), but when I came to assemble it one of them turned out to be dead (they've been used in multiple projects before, so I suppose it's no surprise). Luckily it just about works with BJTs too, so I used the ubiquitous 2N3055/MJ2955.

With two pots (RV1/RV2) for the bias adjustment it takes some extra care to set up. Because I didn't intend to use BJTs there is no allowance for thermal compensation and so it is a bit funny about higher bias currents, but simulation suggests that distortion is lowest with lower bias anyway.

All transistors are BC846/856. I bought a load of them a few years ago hoping to make tiny circuits, which I did, but they're a pain to solder so I don't think I'll ever get any more.

It works ok, but I feel it would have been much better if I had been able to use the MOSFETs.
 
Thanks...I now have another part # to use when ordrering. I am going to use some of these for a differential amp and current mirror. I think they would perform well. Use a bias servo.

The circuit looks like it should work but of course thermal compensation will be a must. You might want to add some emitter resistors on the drivers, and output drivers for that matter... it will make you bias more stable and the parts will act as if they are more closely matched.

BTW: I think your logic is backwards. As the OP-AMP output goes negative, the output circuit will go positive. The OP-AMP is driving a common base amplifier, no phase shift. It looks like you should connect your negative feedback to the non-inverting input of the op-amp and set up a voltage divider between output and ground, through a series Non-Polar capacitor to block DC and set your gain. Input on inverting input should work though.
 
I would have put in emitter resistors, but I had etched the PCB before I had to change to BJTs.

It may look like a common-base driver at first sight, but it's not. You can see that the bases of Q1/Q4 have no reference voltage. The two pairs Q1/Q3 and Q4/Q6 are roughly equivalent to two diodes in series, as a more common bias circuit would have. I know the feedback can't be backwards because it doesn't melt when I turn it on:D
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.