# Question about RFC choke in Colpitts Oscillator

#### silv3r.m00n

Joined Apr 15, 2010
69
I am fiddling with this simple colpitts oscillator:

It has a RFC choke in the collector.
I am trying to understand how it works. I assume that :

1. The choke provides LOW impedance path to DC hence allowing quiescent current or operating current for the transistor.
2. The choke acts as high impedance for the ac oscillations at high frequency hence provides gain by acting as high collector resistance for the AC.

Are the above assumptions correct ? or is there more to it ?

However, the RFC choke is an inductor and causes phase shift between the voltage and current.
So with the RFC choke the amplifier is not a strictly 180 degree inverting amplifier ? OR is it ?

If rfc cause phase shift the loop phase shift will not be 0, and the oscillations should not work.
But the above circuit works fine, producing oscillations around 175 Khz.

Am I missing something ?

#### Marley

Joined Apr 4, 2016
384
You are absolutely correct. L1 provides a DC path but blocks the AC signal.
I would expect its inductance to be considerably higher than the inductance of L2.
If everything works - well done!

#### Danko

Joined Nov 22, 2017
1,019
Compare them:

#### Attachments

• 1.8 KB Views: 2
• 1.5 KB Views: 1
Last edited:

#### silv3r.m00n

Joined Apr 15, 2010
69
You are absolutely correct. L1 provides a DC path but blocks the AC signal.
I would expect its inductance to be considerably higher than the inductance of L2.
If everything works - well done!
but doesn't L1, the rfc choke in collector cause 90 degrees phase shift between v and i, during amplification ?
how is the oscillator able to maintain a loop phase shift of 0/360 deg ?

#### silv3r.m00n

Joined Apr 15, 2010
69
in both oscillators, the v and i are out of phase by about 90 degrees.
so how is the oscillator able to maintain 0/360 deg loop phase ?

the tank is 180 deg diff, the amplifier is about 90, so total is 270 only.

#### LvW

Joined Jun 13, 2013
1,160
in both oscillators, the v and i are out of phase by about 90 degrees.
so how is the oscillator able to maintain 0/360 deg loop phase ?
the tank is 180 deg diff, the amplifier is about 90, so total is 270 only.
Well, the influence of the inductor L1 ist not so strong as you think because L1 is to be considered in parallel to the remaining feedback path. More than that, the transistor is a REAL current source with a finite source resistor r_out.
The frequency determining feedback path is 3rd-order lowpass - see the ladder structure r_out - C2 - L2 - C4.
There is one single frequency, where the phase shift crosses the (-180) deg line and this frequency - together with the phase inversion of the transistor stage - can fulfill the oscillation condition.

#### silv3r.m00n

Joined Apr 15, 2010
69
Well, the influence of the inductor L1 ist not so strong as you think because L1 is to be considered in parallel to the remaining feedback path. More than that, the transistor is a REAL current source with a finite source resistor r_out.
The frequency determining feedback path is 3rd-order lowpass - see the ladder structure r_out - C2 - L2 - C4.
There is one single frequency, where the phase shift crosses the (-180) deg line and this frequency - together with the phase inversion of the transistor stage - can fulfill the oscillation condition.
so it finds a frequency which satisfies the loop condition of 0/360 degree phase shift.

the RFC choke L1 (Rout) in collector is also a part of the frequency determining network
which does not look neat. calculating the frequency by hand gets difficult then.
it affects both the gain and the frequency.

also this circuit has considerable distortion and difficult to modify.
i wonder why it is mentioned in so many websites and books.

#### LvW

Joined Jun 13, 2013
1,160
so it finds a frequency which satisfies the loop condition of 0/360 degree phase shift.
the RFC choke L1 (Rout) in collector is also a part of the frequency determining network
which does not look neat. calculating the frequency by hand gets difficult then.
it affects both the gain and the frequency.
also this circuit has considerable distortion and difficult to modify.
i wonder why it is mentioned in so many websites and books.
Yes - you are right, "calculating the frequency by hand gets difficult".
More than that, in addition to L1 also the output resistance r-out as well as the input resistance at the base (both just roughly known) are part of the frequency-determining feedback path.
I only can guess why its mentioned so very often - perhaps because it is one of the classical oscillators ?
It certainly has the disadvantage that it will not be possible to tune the frequency without touching the loop gain and with it the oscillation condition.
There are other circuits which allow separate (orthogonal) frequency tuning without touching the oscillation condition.

#### silv3r.m00n

Joined Apr 15, 2010
69
In the 2nd circuit , Common Base, what is the function of C1 (battery to ground) ?

I see that removing it has no effect on the amplitude or the frequency of the oscillations.

#### Danko

Joined Nov 22, 2017
1,019
In the 2nd circuit , Common Base, what is the function of C1 (battery to ground) ?
Picture below answers your question. SC1 and SC2 are possible sub circuits.
C1 localizes HF current in oscillator area.

Last edited: