TTL RC Oscillator

[makorihi]

Hello all,

I've been trying to make a TTL RC Oscillator as shown ( **broken link removed** ), and I've been able to do so with questionable success. It seems that for a given capacitance, the oscillator works for a given R +- a small fraction of R. I'm thinking that this might have something to do with the internal resistance of the inverters (I'm using LS04) or possibly the reactance of the capacitor that I'm not factoring in. One combination of C and R that worked for me was C=.01uF and R=5kΩ.

Thanks in advance

 

[MikeMl]

Replace the TTL with CMOS, and you can make it work over a much wider range of resistances. Another trick, use the hex inverter that has the Schmitt-Trigger hysteresis build in, and you can make an RC osc with just one section.

 

[makorihi]

Replace the TTL with CMOS, and you can make it work over a much wider range of resistances. Another trick, use the hex inverter that has the Schmitt-Trigger hysteresis build in, and you can make an RC osc with just one section.

Thanks a lot.

Can you explain to me in detail exactly why TTL doesn't work in this case and why it doesn't work over a wide range of resistances?

 

[MikeMl]

Because the resistor has to do two things simultaneously; one it to provide the RC delay, the other is to bias the gate into its "linear" range where it exhibits voltage gain. TTL requires you to sink a lot of current out of its input to bias it to the center of its switching range. Look at a datasheet for a TTL gate and read IiH, IiL, ViH, and ViL specs. Now do the same for a CMOS gate. Its mA for TTL, while it is only uA for CMOS, so R can be up to several megOhms in CMOS, allowing delays (or oscillator periods) of 10s or seconds.