Small Signal Amplifier

[11flint]

Hello everybody!

I built a 1Hz oscillator using the time circuit from an old quartz wall clock. The oscillator is working wonderfully, I have a very nice, precise 1Hz oscillation, but the problem is, my output signal is not strong enough to drive my 74LS162 decade counter.

The output peak of my oscillator is about 1.5 volts.

I also built a 1hz oscillator using a 555IC, and that circuit's output is strong enough to drive the decade counter (peak is around 3.5 volts) but for reasons of which I am unaware, the decade counter increments on both positive AND negative half cycles. I compensated for this by using a 500 ohm pull-up resistor on the clock input, and the counter worked correctly thereafter.

So basically I am trying to achieve a similar Vout from my quartz oscillator as I am getting from the 555.

Major problem, I don't know the input inpedance of the decade counter (couldn't discern what it was from the datasheet). I wanted to build an impedance matching network, but I don't know what impedances I need to match!

Can I use just a very simple class a amp?

Many, many thanks for your help!

 

[(*steve*)]

You can use something a lot simpler than an A class amplifier.

You require less than 0.8V for a logic 0, and above 2V for logic 1.

The easiest way is to use your oscillator to drive the base of a small signal NPN transistor via a 10K resistor. The emitter should be grounded and the collector should be connected to a 4.7K resistor to +5V. The junction of the transistor's collector and the resistor is the output which will drive an LSTTL input.

 

[11flint]

Thanks Steve! Will do just that!

 

[(*steve*)]

Note that this will invert the signal, but I presume that is of no consequence.

 

[11flint]

No, no consequence. Thanks!

 

[11flint]

I think the signal is too high now. Output is almost 5v peak, and the counter is not liking it. Anyway to reduce the gain?

 

[(*steve*)]

The 4.7K as the collector load should be fine to pull up an LSTTL input I think.

What are the symptoms?

 

[11flint]

The count is irregular, very irregular, which could be due to the fact the circuit is on a breadboard so maybe stray currents are affecting the counter, but it counts relatively well when I'm using the 555 oscillator whose output signal is around 3.5 volts. Now my oscillator Vout is almost 5.

 

[11flint]

Now it's almost counting correctly. What affects the proper count sequence is where I set the 103 pot in the oscillator circuit. Here is an image of the circuit:

The pot between op-amp pins 1 and 5 are what's affecting the correctness of the count, but it only gets close, not exact, which is to count to 59. It's not the timer circuit, because it counts correctly using the 555 oscillator.

 

[Laplace]

It seems your problem would be the emitter follower that is used to drive the clock input of the 74LS162. The clock low-level input current is specified at -3.2 mA @ 0.4 V. The emitter follower resistor of 1K is not able to pull down the clock input. Drive the clock input with the open collector/4.7K resistor arrangement.

 

[11flint]

Thanks Laplace. You mean replace the 1k resistor with a 4.7?

 

[11flint]

In any event, the problem is more serious, because if I attach the output to another pair of counters, everything goes nuts.

 

[(*steve*)]

What does the waveform look like. You draw it as a square wave, but have you looked at it? I would imagine it looks more like positive and negative going spikes with a decaying trailing edge.

I suspect that it has quite slow rise and fall times and a lot of time spent in intermediate values. Logic won't like that at all, *especially* on clock signals.

You should probably change your op-amp circuit to be a comparator with hysteresis rather than an amplifier.

 

[Laplace]

What I mean is that if you look at the 74LS162 datasheet you will see that for the interface to the counter clock input you need to pull down 3.2 mA and keep the low level input voltage within a few tenths of a volt above ground. An emitter follower is not going to do that.

 

[11flint]

Uhh, I'm really new to this. You guys might as well be speaking Cantonese. Laplace: I don't know what an emitter follower is. I'm guessing its an amp with open collector and a resistor between emittor and ground? If that's no good, what sort of configuration should I use (I mean in relation to the 1Hz schematic I posted, in other words, how should i configure the transistors recieving the input from the op-amp?)

Steve: I don't have an oscilliscope. I have no idea what it looks like. How do I configure the op-amp to be a comparator? Would i then have to amplify that signal?

Many thanks.

 

[11flint]

OK, I got the configuration for a comparator. Could I then amplify that signal using the design you specified before? (collector to 5v through 4.7, emitter to ground, etc)

 

[(*steve*)]

With a comparator the output should swing from rail to rail anyway with no need for the transistor.

 

[11flint]

OMG, the problem in the end was that I was using a 47K resistor on the collector instead of a 4.7. I must be going blind.

 

[(*steve*)]

Don't worry, some of us *are* going blind.