74HCT04N Crystal Ocsillator Circuit

Here we take a look at a simple oscillator clock generator circuit using a 74HCT04N hex inverter chip. Starting on a breadboard I build the circuit and try two different frequency crystals. Viewing the waveforms on the oscilloscope we can see some issues.
Then moving on to a soldered version we clean up the waveform by making some adjustments. Finally, I demonstrate how much difference it makes to use a good probing technique.

[@stephenwissel7902]

The 1k is a drive loading resistor, you can play with this value to soften the drive and it may help with the ringing. Cascade the spare gates at the output will help clean the the output.

[@videolabguy]

The jitter on the scope is being caused by the trigger threshold being set wrong. At 9:31 for instance, the trigger indicator is on the left side of the screen and is pointing straight at the ringing part of the waveform. Use the trigger level control to move that pointer up to the center of the square wave and it will be rock solid. Set in the ringing part of the wave like that is what makes the trace jitter. That was a great little project. I am working on a similar one right now with a much much lower frequency crystal. 31.5KHz. I am having all kinds of trouble with it. That is why I was searching for inverter based XTAL oscillators and found your video.

[@SteveRaynerMakes]

Awesome and thanks for the advice. I'll have to connect it up to the scope again and have a play around. I'm new to using the oscilloscope and really appreciate any help and advice.

[@ivolol]

For a 32.768khz crystal, I found values of 10M across the inverter, and 270k in series to the crystal, worked to get it singing. So overall you have a lot less current and feedback for the low khz ones. The best practice I think is to run it at ~3V but 5V still works. If it's just for short testing I think it should be ok.

[@therealjammit]

The 1meg resistor puts the inverter into a linear mode so it acts more like a class a amplifier. I don't know what the 1k is used for. I've seen the same circuit without it. I think protecting a random crystal from overcurrent makes sense.

[@iblesbosuok]

Please put a 100nF capacitor as close as possible to pin 7 and pin 14 of 74HCT04. Thank you.

[@vobinhthu5299]

This project is very good!❤

[@SuperFredAZ]

the resistor is there to raise the impedance at the output. A much better solution is to use the 74HCU04 which has high impedance, leave off the 1 K

[@paulperano9236]

Try grounding the unused inverter inputs.

[@SeAfasia]

Ηi Steve...clever idea...don't remember how call those issues at scope....probably "jits" think so...try with some cap ceramics...ic power supply..

[@subhashkendole3005]

Any special capacitors to be used ? I am getting just 1 Mhz

[@SteveRaynerMakes]

I'm using 22 pico farad capacitors.

[@ivolol]

The specific load capacitance to use is told to you in a datasheet for your crystal, sometimes it will even be in the title of the product because it is one of the most important parameters. Generally you have two capacitors the same, lets call them C_a and C_b. They should solve this equation: C_load = (C_a * C_b) / (C_a + C_b) + C_stray where C_stray is any extra capacitance from your wires / breadboard / microcontroller etc.

[@subhashkendole3005]

Sir, 1. I am getting square wave with 1.84 Mhz but with 6 Mhz and 8 Mhz, lower side is sine wave instead of square wave. What could be reason ? 2. Results are not good with SMD crystal as compare to metallic crystals.

[@vanhetgoor]

Maybe something else works as a capacitor and as you know capacitors behave different at higher or lower frequencies. At very high frequencies two pieces of wire in the same neighbourhood can work as a capacitor while at lower frequencies they have hardly any effect.