Square wave from Colpitts

Thread Starter

cesare.magno

Joined May 19, 2021
18
Hello everybody!! can I use a colpitts oscillator to generate a square wave? i thought connecting a smith trigger (74hc14) out to the Colpitts oscillator is this a good idea?
 

Ian0

Joined Aug 7, 2020
9,669
Yes.
Provided that the maximum amplitude of your Colpitts oscillator is higher than the high threshold of the 74HC14 and the minimum amplitude is lower than the low threshold of the 74HC14.Screenshot at 2021-05-31 21-12-57.png
Alternatively, you could amplify it several times with several sections of 74HCU04 wired as amplifiers (large resistor between input and output). You could then couple capacitvely so that amplitude and threshold voltages wouldn't be a problem.

You could even use a section of the 74HCU04 as the oscillator.
 

atferrari

Joined Jan 6, 2004
4,764
Are you sure you need to build a Colpitts to later need something else to finally get a square wave as output? Difficile da credere.

Is it homework?
 

Ian0

Joined Aug 7, 2020
9,669
Are you sure you need to build a Colpitts to later need something else to finally get a square wave as output?
I agree - using a logic gate to turn a sine wave into a squarewave would mean that there would be spare logic gates, and it only takes one inverter to make Pierce oscillator.
 

sparky 1

Joined Nov 3, 2018
756
The Italian seems to work, why might I need diodes ?
La geometria di un circuito di accensione e spegnimento è molto più semplice perché ci sono solo due soglie di tensione.
english:
The geometry of an on and off circuit is much easier because there are only two voltage thresholds.

Italian:
Il ruolo del filtro e anche il ruolo dell'impedenza nel partitore di tensione che è la forma matematica capacitiva Z.

english:
The role of filter and also the role of impedance in voltage divider that is math form capacitive Z.

Italian:
Il filtro ha anche una costante di tempo nel percorso di feedback di un amplificatore in Colpitts è un graduale cambiamento di tensione.
L'approccio funzionale può essere mostrato in matematica.
Ecco perché la visuale potrebbe non essere intuitiva.

english:
The filter also has time constant in the feedback path of an amplifier in Colpitts is gradual voltage change.
The functional approach can be shown in mathematics.
That is why visual might not be intuitive.

The derivation of the capacitive voltage divider in terms of Z was answered after class.
Most of the electrical engineering students multitasked a tall beer and a game before diving into the next chapter.
 
Last edited:

sparky 1

Joined Nov 3, 2018
756
The equivalent crystal model can be used in many different oscillator types.
A Colpitts crystal oscillator shown figure 7 has rough output because the startup requirements.
https://www.qsl.net/zl1an/Crystal_Oscillator.pdf

I like the op amp because performance and gain set but the same concept Zc
shown video
The quality in a crystal injected oscillator will require careful attention to many details.
The distorsion analysis of a triggered square wave will reveal avoidance
but the loading effect will be sort of ok for some applications.
 
Last edited:

LowQCab

Joined Nov 6, 2012
4,023
The 7555 also puts out a Square-Wave on Pin-3, and
has an Open-Collector Ground Output on Pin-7.
Either Circuit will create very stable Square-Waves, and with substantial drive Current.
.
.
.
PWM Comparitor .PNG7555 Triangle Oscillator .png
 

Ian0

Joined Aug 7, 2020
9,669
i implemented square wave, but it has ripple, is there any way to take it out?
Is it real? Or is it due to the position of your scope earth? Is your scope probe perfectly calibrated?
Do you have enough decoupling capacitance across the power supplies next to the IC?
Try a resistor between the output of the inverter and your capacitive load.
 

Papabravo

Joined Feb 24, 2006
21,159
How does that work then? Is a single inverter really faster than a group of six?
No, but the LVC technology has smaller features and lower propagation delays as well as inputs that are tolerant of voltages above Vcc which can go down to 1.65V
 

Ian0

Joined Aug 7, 2020
9,669
You can get 74LVC14 with six in a 14 pin package, but I'll let you off!
I also note about the LVC is that the hysteresis and threshold levels are more accurate than the ordinary HC. You could make a relaxation oscillator out of one and have a fair chance of being able the predict the frequency. I bet it would run at 4MHz!
 

Papabravo

Joined Feb 24, 2006
21,159
You can get 74LVC14 with six in a 14 pin package, but I'll let you off!
I also note about the LVC is that the hysteresis and threshold levels are more accurate than the ordinary HC. You could make a relaxation oscillator out of one and have a fair chance of being able the predict the frequency. I bet it would run at 4MHz!
I did specify the "1G" variant in my original post. Ding rejected.
Actually better for that purpose would be the 74LVC1GU04. The unbuffered parts make dandy oscillators.
 

Thread Starter

cesare.magno

Joined May 19, 2021
18
the circuit is mounted on a breadboard with a metal base. Maybe I need to connect the metal base (ground) to the negative voltage of the power supply? the waveform is real and I didn't check if the probe was well set, you say it could be a probe problem? could decrease the value of the divider resistances be an idea? (unfortunately the circuit is in the laboratory and I have to wait for tomorrow to see). Perhaps, the values of the resistors are too small, I am talking about resistors to halve the supply voltage
 
Last edited:

Ian0

Joined Aug 7, 2020
9,669
I did specify the "1G" variant in my original post. Ding rejected.
Actually better for that purpose would be the 74LVC1GU04. The unbuffered parts make dandy oscillators.
I might perhaps have mentioned that back in post #2. I can’t see a good reason for making a Colpitts oscillator with a transistor then using logic gates to make a square-wave when I could make a Pierce oscillator with one of the logic gates,
 
Top