i recently made an amplifier like this and it sound suprisingly good.
I couldnt find any posts on unbiased class b amplifiers so here is one.
feel free to give me any suggestions and stuff.
feel free to give me any suggestions and stuff.
But the BASE of Q1 & Q2 are joined ... How can you have + & - bias 😕
I would love to see the distortion performance.
I would love to see the distortion performance.
The output opamp needs to be pretty beefy - MJE2955 has Hfe about 30 @5A so U2 will need to sink 170mA or so to reach clipping into the 4R load.
I recall an 'antient circuit' without R3 using a single LM741 to drive GERANIUM output transistors AD161 & AD162.
The explanation of "no bias" was that those old transistors had sufficient collector to base leakage to suffice given a .3V junction.
The explanation of "no bias" was that those old transistors had sufficient collector to base leakage to suffice given a .3V junction.
There is a similar circuit published on a german site:
https://www.elektronik-kompendium.de/public/schaerer/endstu_p.htm
Scrolling down to "Technische Daten" provides some technical data, also distorsion, which curiously decreases with increasing frequency. The author explains it with ripple influence.
https://www.elektronik-kompendium.de/public/schaerer/endstu_p.htm
Scrolling down to "Technische Daten" provides some technical data, also distorsion, which curiously decreases with increasing frequency. The author explains it with ripple influence.
the tl072 i used didnt explode or anything but i also used a 220 ohm resistor for r3
It isn’t necessarily a BAD exercise to go do, but you will quickly find the circuit’s limitations. And when you solve them one at a time, you will end up with something more complex.
I'd expect the TL072 to succumb to over-voltage (50V total supply) before it got a chance to drive any output.
I bet the output signal was rather small - the finite gain of those transistors might give you half an amp of peak current - IYL. If you were able to drive to the rails (or even close) the. 072 would not survive. A 5532 might. While playing with this, just lower the supply to +/-15V. A 3 or so amp current limited supply might be a good idea too.
I used to love playing around with circuits like these years ago 🙂 so keep working on it 👍
The NE5532 might not be very happy running on 50 volts 😱 and as others have said, the low current gain of the outputs is a real issue. The opamp (a real one) can not supply enough base current.
Now the lack of base current (opamp output current) becomes apparent.
1watt into 4 ohms:
Distortion:
Here is a real (commercial) Class erm.... well Class C in my book...
This uses Darlington's and so there is a huge crossover 'dead zone' for it to traverse.
The NE5532 might not be very happy running on 50 volts 😱 and as others have said, the low current gain of the outputs is a real issue. The opamp (a real one) can not supply enough base current.
Just pick some representative ones from the standard library such as the LT1056 which is like a TL071/2
Now the lack of base current (opamp output current) becomes apparent.
1watt into 4 ohms:
Distortion:
Here is a real (commercial) Class erm.... well Class C in my book...
This uses Darlington's and so there is a huge crossover 'dead zone' for it to traverse.
I’m kind of surprised to see 7 volt peak driving a pair of power trannies directly with an op amp. A pair of CFPs will drive a lot more. The crossover region remains, though, and might even get worse. Fixing the crossover region of simple EF buffers inside the feedback loop is what the diamond buffer was made for. And you get the gain that’s needed.
What do you think those resistors to the opamp power pins do?
They will not reduce the supply voltage because the opamp draws very little current. You would need much higher values and also add a Zener from each supply pin to ground to regulate the supply.
Like this. It allows the opamp to run on the -/+25 volt main supply but it still has the same limited drive problems and also seems to run into stability issues at close to max output.
Well not by very much 🙂 Even if the opamp could supply 30ma base current it would only pull the rail down by 0.6v or so.
Add a 1 ohm, or smaller, resistor to the output of U2 and then it will be easy to plot the current output of U2. When I had a bunch of identical op amps in my collection, I would just stack several DIP package op amps in parallel to get higher output current. As others have noted, the output voltage swing of the U2 operating as a follower will limit the output voltage of the amplifier. Running two of these in bridge mode would enhance the output power capability.