100W Class B Amplifier

772

03:24:01

https://ibb.co/xqsTFVy

Check out my latest amplifier project and the most sophisticated yet. I've gathered all knowledge I've collected on audio amplifiers over the years in this final Class B 100W amplifier. Some of its features are:

- Auto-Biasing - No need to set the output voltage offset. With the use of an op amp integrator, the amplifier will always find its output settled on 0V, without external adjustment. This is handy, because typically, zero offset is rarely achieved when adjusted manually, and the output is usually prone to temperature drift once the amp has warmed up. 

- Increased Ripple Rejection - The preamp and driver stages are more than well filtered enough to achieve ripple rejection ratios of more than -90dB.

- Power Transistors Short Circuit Protection - The output transistors are short circuit protected with an overcurrent protection. At these power levels such a protection is mandatory. A short circuit on the output will be restricted below the maximum power the transistors are rated for. WARNING ⚠ - A short circuit state is not healthy to the amplifier despite the protection. A short circuit should be broken ASAP. The protection can only protect from transistor damage for so long, before imminent failure occurs, as every part of the circuit will be under strain. 

- Improved Thermal Compensation and bias stability - The bias current range is limited to a very narrow range between 2mA for a pair to 50mA via the trimmer R20. If T9 is mounted on the same heatsink as the output transistors, thermal runaway should be brought to a minimal.

- DC Speaker Protection - DC speaker protection is employed via the relay, where a delay circuit of 5s halts the speaker enabling until the amplifier has reached a steady state upon turning it on. Also, a DC offset of more than 2.5VDC for more than 300ms will trigger the protection and disengage the speaker. At these power levels, this is another mandatory protection.

- Broad Frequency response - The amplifier has a very broad bandwidth of less than 10Hz to more than 50kHz. 

The amplifier has a lot more subtle improvements to ensure stable and fidel operation. Enjoy!

published 7 years ago

PrathikP

7 years ago

A voltage source and a resistor?

Robert_Kidd

7 years ago

Follow the link.

Karma247

7 years ago

Very nice 👍

zorgrian

7 years ago

We used to use a lamp bulb as speaker protection! 220v incandescent bulb in series with the speaker

zorgrian

7 years ago

Nice design @thebugger

thebugger

7 years ago

@Zorgrian Doesn't a light bulb create significant voltage drop? A 100W bulb would have a typical resistance of 400ish ohms. I've seen it used on primaries of transformers for protection, but not on the outputs of amps.

sircube

7 years ago

The bulb only has that resistance when is hot and bright, when it is cold, it's basically a short, so that would work as a protection: if enough current passes through, the filament heats up, increasing its resistance, basically it is a PTC

zorgrian

7 years ago

An incandescent bulb is a non-linear device. Interesting as it can also be used to make a kind of simple compressor. @sircube is totally right in the summation of the principal upon which we used this 'device'. A cheap and very useful way to protect your speaker + output transistors!

zorgrian

7 years ago

As a bass player, I used this mostly to stop speakers from being blown. I don't really like the compressed bass sound. Instead, I prefer a really punchy uncompressed bass sound that's raw.... The bulb works well as if it lights up too much, then you know your kicking it out!

thebugger

7 years ago

Yes, I am aware of using a bulb for its PTC in some circuits, but haven't thought of using it in an amplifier. Makes sense though, that it'd have a more nonlinear and ,,crunchy" effect on bass frequencies, as bulbs tend to have a low time constant, in the bass range, so the constant resistance shifting would definitely cause some unforseen effects on the speaker, especially keeping in mind that a speaker is not a purely active load. Quite nonlinear and unpredictive to my opinion, but most guitar players are always looking for that extra bit of distortion, so it makes sense actually. In Hi-Fi circuits I wouldn't recommend using this method though, as it creates a higher impedance and frequency dependent output of the amplifier, and in Hi-Fi circuits to drive a speaker as accurate as you can, you always aim for lowest source impedance and flat frequency response. By the way, that's part of the reason tube amps sound so special. The source and load impedance are interconnected in tube amps, so as the load wobbles its impedance (through reactance), the source impedance also wobbles. This creates an intrinsic relationship between speaker and amp, that makes a very peculiar sound to it. Also tubes tend to distort more in even harmonics than odd, when overdriven, so that makes the distortion more pleasant for the human ear.

sircube

7 years ago

Right, I did not think about the low frequency distortion problem

add comment in editor