Signal Amplification

Thread Starter

Pooperman

Joined Feb 23, 2011
30
I am working on a project that uses a 32.768kHz crystal to generate a signal. I need to keep it separated from another part of the overall circuit due to other oscillators being present (they tend to drive one another). In order to separate them, I have an optocouple/optoisolator between the two circuits which works fine at about the output of the oscillator in question. However, due to sensitivities later in the circuit, I need to clamp down on the voltage with a zener diode before the jump across the optocouple. To do this, i need to amplify my ~1.25VPP signal (0 to ~1.25V) to about 3VPP or so because I can't use a zener otherwise.

I've tried several amplification methods (inverting opamp, transistor, non-inverting opamp) but none of them are working well. The transistor amplifies to about 1.5VPP but no more. The inverting opamp boosts it to 3.5VPP but inverts it (not good for driving the LED in the optocouple). I couldn't get the non-inverting to work at all. I've tried two inverting opamps back to back also.

Is there a simple way to amplify the signal out of the oscillator to a level where i can clamp it and still drive the LED?

Diagram of the oscillator can be found here: http://www.eleccircuit.com/32768-khz-oscillator-using-a-watch-crystal/ (the one on the left side is the one I'm using. The inverter is fed with about 1.0-1.2VDC, and both caps are 22pf)

The optocouple is just an LED from the transmitting side.

Image of an optocouple: http://en.wikipedia.org/wiki/File:Optocoupler_Circuit.svg (the shaded box labeled D1 is the optocouple)
 

Adjuster

Joined Dec 26, 2010
2,148
Obtaining 2 times voltage gain at this frequency should really not be difficult. Please post schematics of the amplifier circuits which have not worked - perhaps someone will be able to see what has gone wrong.

It would also help to know what you mean by "clamping down on" the voltage. Do you mean to limit the input to the LED end of the coupler - presumably with a resistor on each side of a Zener? Post a circuit for this too.
 

Ron H

Joined Apr 14, 2005
7,063
Are you using a 5V supply on the oscillator?
Are you actually using a 4069, or some other inverter?
Is the output of the 4069 5 volts p-p?
 

Thread Starter

Pooperman

Joined Feb 23, 2011
30
About the zener, you're right in saying that it is there to limit the voltage into the circuit from the incoming signal as it could potentially be attached to a signal generator with higher voltage (which would throw off the other side significantly). As for circuit diagrams, I can only post what I find on the internet (no scanner).

http://www.tpub.com/neets/book7/25c.htm (the NPN variety)

http://www.elexp.com/t_gain.htm (both with correct resistors to triple the value (according to the gain equations), as well as inverting into inverting unity gain)

Again about the zener, I want to make the amplifier work before I clamp (less variables to mess with). Seeing as I'm about 2 months into this project and nearly done, this problem has been quite perplexing.

As I asked before, is there a simple way of doing this where all these components don't royally screw everything up?
 

Adjuster

Joined Dec 26, 2010
2,148
You amplifier requirement is fairly easy, provided that it is approached correctly. For instance, 33kHz is a high enough frequency for you to have to be careful with op-amp slew rates and bandwidth. You probably wouldn't get far with a 741, for example - what device type did you use?

You also need to consider the need for perhaps 10mA drive capability to supply the coupler LED, but at the same time providing a high enough input impedance so as not to load your oscillator too much. This may have been an issue with a single-stage transistor amplifier, so perhaps the op-amp route would be better provided that you get a suitable device.

By the way, this amplifier idea may be missing the point anyhow. As the last poster has just pointed out, what is the supply on the 4069? Shouldn't you get more than 1.25Vp-p from the oscillator in the first place??
 

Thread Starter

Pooperman

Joined Feb 23, 2011
30
As mentioned in the first post, the inverter is powered by 1.0-1.2V because otherwise it overloads the crystal and thus it does not output the correct frequency (believe me, I tried a lot to get a higher voltage). If memory serves, I'm using an LM324 as the opamp in the other part of the circuit and it works. It is also what I tried to use for the opamp amplifier. The inverter is a HC though I cannot remember the number. I'd be able to give that info when I look at it again tomorrow.
 

Thread Starter

Pooperman

Joined Feb 23, 2011
30
As I said, I don't have the circuit right in front of me, but I don't think it's a 4069. The opamp is powered at 5V (or slightly higher). The other part of the circuit past the optocouple is working as intended. It is getting the signal high enough to clamp and then getting it through the optocouple that is the difficult part.
 

Thread Starter

Pooperman

Joined Feb 23, 2011
30
This isn't about the opamp. I'm saying the oscillator is working (i'm getting the correct output). When i used the inverting lm324 it was working (i was getting about 3.7vpp square wave as expected). It how to amplify it so that it can still power the optocouple and so that i can clamp it.
 

Thread Starter

Pooperman

Joined Feb 23, 2011
30
It doesn't really matter what is beyond the optocouple since it doesn't affect the stuff before (that is the point of an optocouple in the first place). I'm trying to control the voltage into the optocouple by limiting it to 3V (or thereabouts) and get the 32.768kHz though so that it may interact with the other circuit without the second circuit interfering with the one in question. Everything else works as I want it to except amplifying the oscillator so I can clamp it with a zener.
 

Thread Starter

Pooperman

Joined Feb 23, 2011
30
It is, yes. All I can really say is that it needs the oscillator in question to drive a second oscillator at the same frequency. However, this operation is very sensitive to voltage, which i can control with a potentiometer. However, the voltage could possibly be changes if the oscillator is replaced with a waveform generator. Then the specific setup is compromised and the oscillator can be driven very easily. This probably gives away the idea, but not how it works precisely.
 
Top