Square Wave Amplifier

[armdnrdy]

Hey all,

I've been pulling my hair out trying to amplify a clock square wave about 6 volts. I've tried all of the usual suspect BJT amplifiers including common emitter, push pull / totem pole, and direct coupling.

I need to keep the clock at around 7 volts for one purpose, but then I need to raise the voltage to somewhere above 11 volts for 15 volt logic high for another purpose in the design.

The above amplifiers didn't work out too well when bread boarded. Some didn't amplify; some distorted the square wave too much.


While looking around on the net, I came across a YouTube video of someone scoping a circuit that he roughly soldered together. The scope images looked good, so I paused the video and tried to figure out what circuit it was.
It turned out that it used 2 transistors, 2 resistors, and two diodes. It looked like the BJTs were faced in the opposite direction from one another which suggested that they were NPN and PNP like a push pull. As I mentioned, the circuit also had two diodes like a push pull, but that's where the similarity ends.

I threw a couple of 1N4148s and a few 10Ks together since I couldn't make out any values, put it together on the breadboard, and it worked.

Here's the circuit: http://www.aronnelson.com/gallery/main.php/v/diyuser/Square+Wave+Amplifier.jpg.html

Does anyone know of any surefire/easy way to do this?  Can someone take a look at this circuit to see if it has any issues of impending doom?

Thanks

[PRR]

> any issues of impending doom?

It's quite lame.

First, nothing happens until +/-1.2V, to overcome junction drops.

After that, it is at best unity voltage gain.

There's potential for current gain. However the 10K resistors mean hardly any current can get out.

> around 7 volts for one purpose, but then I need to raise the voltage to somewhere above 11 volts for 15 volt logic

How fast? How clean?

[armdnrdy]

Hey PRR,

Thanks for the response,

I actually got a decent (a bit of distortion at the top right corner) square wave at 13.6 V out of this thing! Square wave input of 7.1V and 15V VCC. I'm just a bit leery because this circuit doesn't really resemble anything I've found on the net. I don't know how stable it would be running for a while. To be honest, the YouTube video was pretty blurry and I'm not sure of much with that circuit!!

To answer your questions: I'm running the clock at 10kHz. and I'm clocking four counter ICs CD40110BE with the output, so fairly square.

I've had the thought to use the 7.1 volt clock line to clock a 555. It's kind of redundant to have two clocks but at least it will solve the problem. The 555 output is 1.7 volts lower than VCC, so it will fall nicely in the high logic level and provide a clean square wave.

[~arph]

Wouldn't a simple CMOS inverter do the trick?  Pull up the input to the supply voltage with 100k. and apply the 7V clock to the input via a cap.

[brett]

Hi
+1 for Cmos inverter. Or, for 25 fewer legs than an inverter try a Cmos transistor (common source, gain of 3, try a BS170). It'll hit either rail in a microsecond.
Or am I missing something?
cheers
PS the two transistors hooked together might have been a long-tailed pair. Maybe? It's a widely used setup (though not too common in stompboxes outside of op-amps).