I have a circuit which includes a chip with a 24MHz ceramic resonator oscillator built in. There is about 86mV p-p there. I want to steal the signal from there and make it match CMOS logic levels (0.3VDD to 0.7VDD) using a 5V supply. Clearly it must be low input capacitance and high input resistance to avoid upsetting the existing circuit.

I have tried simulating (LTspice) simple transistor amplifiers without much luck. Then I thought about using 74HC04 for the amplifier. I found a 74HC.lib but I can't get LTspice to use it.

Has anyone got any suggestions for circuits I can use for this.

 

I have a circuit which includes a chip with a 24MHz ceramic resonator oscillator built in. There is about 86mV p-p there. I want to steal the signal from there and make it match CMOS logic levels (0.3VDD to 0.7VDD) using a 5V supply. Clearly it must be low input capacitance and high input resistance to avoid upsetting the existing circuit.

I have tried simulating (LTspice) simple transistor amplifiers without much luck. Then I thought about using 74HC04 for the amplifier. I found a 74HC.lib but I can't get LTspice to use it.

Has anyone got any suggestions for circuits I can use for this.

A JFET source follower is a minimal load to put on it, but you'll need another stage for voltage gain.

 

Good thought!
More playing got me to a circuit which might work. At least the simulation looks OK.

 

See

 

Yes, I did consider removing the existing resonator and making a separate oscillator with it and then feeding that signal to the original circuit and the new circuit but I don't know whether the original circuit can cope with that. The datasheet only covers using the internal oscillator. Assuming I can extract a signal from the existing oscillator without disrupting it (and I think connecting only the gate of a JFET has that covered) then amplify that signal up to logic levels then everything will be happy. It seems to take a lot more circuitry than I expected though.

 

my library, symbol and test

 

Thank you Bordodynov. I have managed to add your library to LTspice and so I can now simulate the 74HC04U version of the stealer which works very nicely too. I added some power supply impedance and decoupling too to see its effects.

 

Yes, I did consider removing the existing resonator and making a separate oscillator with it and then feeding that signal to the original circuit and the new circuit but I don't know whether the original circuit can cope with that. The datasheet only covers using the internal oscillator. Assuming I can extract a signal from the existing oscillator without disrupting it (and I think connecting only the gate of a JFET has that covered) then amplify that signal up to logic levels then everything will be happy. It seems to take a lot more circuitry than I expected though.

On a large number of such chips - the internal oscillator is nothing more than a CMOS inverter, you add an external feedback/bias resistor and crystal or resonator. Quite often the pin to the input of that gate can be used as a logic level sync input - it would be handy to get hold of the data sheet and see if that applies here. If the oscillator is just an inverter - try to identify which pin goes to its output, that should be capable of driving an external CMOS gate. If frequency is critical, you may have to trim the oscillator - often there's a pre set variable trimmer cap.

 

I have just had another look through the datasheet and I missed this bit. It does give details for an external clock so that may be the way to go.