# electronics equation question

Discussion in 'General Electronics Chat' started by scoundrel421, Oct 8, 2018.

1. ### scoundrel421 Thread Starter New Member

Oct 8, 2018
4
0
I know this equation (f=1/3.3RC) is an Oscillator Frequency equation; just wondering where the '3.3' came/derived from. NOTHING I have studied thus far used that figure as a Constant...any ideas?

2. ### OBW0549 Distinguished Member

Mar 2, 2015
2,536
2,133
Don't feel bad; I've been "doing electronics" since the late 1950's and I've never seen it either.

My guess: it's probably specific to a particular integrated circuit oscillator, and not applicable to anything else.

3. ### scoundrel421 Thread Starter New Member

Oct 8, 2018
4
0
Most likely, that is the case. I ran into the equation in a practical exercise; it posed no problems in solving it, but I just didn't understand where the figure used as the constant came from or was derived from. The practical exercise was building and testing an oscillator for a logic probe, and that formula was for calculating the frequency of the oscillator with particular variables, i.e., the resistance (R) and capacitance (C) values

4. ### MrChips Moderator

Oct 2, 2009
16,887
5,199
As a general rule of thumb, we know that tau = RC.
For a first order filter 2πf = 1/RC,
or f = 1/6.3RC

If you check the application notes of various timing circuits you will find variations on the xRC theme.

The period can be xRC where x can be anything from 1 to 6 depending on the switching thresholds of the oscillatory circuit.

The devil is in the details of the circuit.

5. ### BR-549 AAC Fanatic!

Sep 22, 2013
4,010
1,031
Something about a phase shift resonant frequency struck me when I saw it.

6. ### scoundrel421 Thread Starter New Member

Oct 8, 2018
4
0
I thought of Phase differences, too; but there wasn't anything explained/mentioned about what '3.3' represented, other than a constant of some sort...and I kinda guessed at that, too, with no real way of checking if I was right...all I know is that I had no problem doing the math and making the equation work, and having it make sense to me

Mar 10, 2018
1,576
324
Not sure why they got this result but -

Maybe its a composite factor derived from phase shift and Tpd of inverter
and non symmetric inverter thresholds, but then at low freq Tpd must be
irrelevant.

Regards, Dana.

Last edited: Oct 8, 2018
8. ### MisterBill2 Well-Known Member

Jan 23, 2018
1,234
212
For a number of devices there are special formulas for the time constant with that particular device. two that come to mind right away are the CD4047 one-shot IC and the 555 timer IC. In addition, there are some circuits where a specific RC time constant formula applies because of the trigger voltages in that specific circuit.
So the formula you present would only apply to a very specific instance.

9. ### crutschow Expert

Mar 14, 2008
19,805
5,543
It's a phase-shift oscillator.
The frequency of oscillation is where the phase-shift of each RC LPF is 180°/3 = 60°, (for a total feedback phase of 360° ) and that occurs at a frequency of 1/3.3RC.

Note that the bottom of the capacitors must all go to ground.

Tonyr1084 and AnalogKid like this.
10. ### noweare Member

Jun 30, 2017
60
9
Here is a great video from youtube on RC phase shift oscillator. Your impressive Crutschow !

Last edited: Oct 9, 2018
Ramussons likes this.
11. ### ebp Well-Known Member

Feb 8, 2018
1,837
641
60 degree phase shift occurs at 1/3.63RC Hz

The circuit at #7 is not a conventional buffered phase shift oscillator. It is a string of discrete time delays using (usually) CMOS inverters and can be built with any odd number of stages. The waveforms will depend on the time delay of each stage and the number of stages. An unbuffered CMOS inverter will typically switch at half of Vcc, but the gain isn't especially high and the output transitions won't be fast, so with very small RCs the outputs may spend more time slewing than sitting at a rail. Once the RC gets sufficiently great, each output will become more square and swing rail to rail. If the number of stages built that way is large enough, peak to peak voltage on each capacitor will get close to being rail to rail, which of course means the delay of the stage is longer because the cap voltage has to swing farther to cross the switching threshold. Schmitt trigger inverters can be used and the delay times will be quite different for the same RC values. A Schmitt based circuit will reliably oscillate with a single stage, a simple inverter probably won't. The RC values don't need to be the same for each stage.

Last edited: Oct 10, 2018
12. ### scoundrel421 Thread Starter New Member

Oct 8, 2018
4
0
THAT is exactly what I was thinking it may have been; I was just having some trouble explaining it to myself...it now makes Perfect Sense; now, Life Can Go On, and the world can start turning again *lol*...Thank You for the explanation