# Can't get higher frequencies in Colpitts oscillator

#### kamilkarp

Joined May 19, 2018
4
Hello!
I want to get around 120-150MHz in my oscillator. Unfortunatelly, I can't get frequencies above 70MHz. With that circuit attached I'm getting around 65MHz. Everytime I change coil or capacitors(tried many combinations of 60-200nH coils and 2p-68p capacitors). Can someone explain me, why I can't get much higher frequencies? What's wrong in my circuit?

Can someone explain how to calculate proper values of resistors/capacitors/coil or provide a book name, which describes it in a simple way?

In LTspice I can get frequencies 70-170MHz, but in reality, when I solder circuit I can't.

#### crutschow

Joined Mar 14, 2008
24,709
The gain-bandwidth of that transistor is only 150MHz minimum which is not enough for an oscillator at 150MHz.
You need an RF type transistor with a higher gain-bandwidth, probably several times the highest desired oscillation frequency.

#### kamilkarp

Joined May 19, 2018
4
It has around 200MHz average. I thought that it will be enough for like 120MHz.

Could you tell me why I'm getting 60-70MHz in my circuit? I doesn't match with the result of equation:
f=1/(2pi*sqrt(L*[c2*c3]/[c2+c3]))
Result of this equation is around 164MHz

#### RichardO

Joined May 4, 2013
2,271
The way the circuit is constructed is constructed has a large effect on the stray capacitance and inductance. How is your circuit built? A picture would be helpful.

The BC546B has a large amount of junction capacitance compared to the frequency setting capacitors in your circuit. This makes predicting the actual frequency of oscillation difficult.

The transistor you are using does not have a lot of gain at 150MHz -- its F_t is as low as 150MHz. You might try a J-FET such as a 2N3819 instead of the BC546B -- It does have a lot of capacitance, though.

p.s. I type too slow. @crutschow beat me to the F_t comment.

#### kamilkarp

Joined May 19, 2018
4
It's soldered on universal PCB.

#### DickCappels

Joined Aug 21, 2008
6,300
Parasitics in the transistor that are not accounted for in the formula as well as the ft of the transistor contribute to the actual frequency.

Take crutschow's advice in post #2 and things should improve.

#### Papabravo

Joined Feb 24, 2006
13,459
It is a common pitfall for experimenters to ignore things they cannot see or measure. The equipment you need to see and measure the characteristics of your active devices and their matching networks at those frequency ranges will cost you more than a medium size house. Try \$250 Kbucks just for starters. If you want simulation and design tools that will get to half a megabuck.

#### Bordodynov

Joined May 20, 2015
2,542
I added parasitic inductors and capacitance and the frequency decreased significantly.I used models of parasitic inductance and capacitance.Their symbols almost do not clutter the circuit (almost invisible).I also built a circuit on a higher frequency transistor.Compare the three frequencies.
See

#### Bordodynov

Joined May 20, 2015
2,542
There is one more thing that changes the frequency.This is the influence of the instrument (oscilloscope).

Joined Mar 10, 2018
4,057
What are the design goals ?

1) Frequency accuracy.
2) Drift with T and V.
3) Power out.
4) Jiter.

Regards, Dana.

#### RichardO

Joined May 4, 2013
2,271
Nice pictures. Because is it hard to see the connections on the circuit side of the board, it would help if you labeled the parts on the board to match the reference numbers on the schematic.

The layout is okay but not great. The two main weaknesses are the lead lengths and the ground.

I don't see the power supply bypass caps. If they are there they have leads that are way too long.

The ground area should be large and all the connections to ground should be close together. Similarly, the power trace should be short and direct to the components.

Note that the wiring to the inductor is almost as long as the inductor. The total inductance is a lot larger than you expected because of this.

#### Janis59

Joined Aug 21, 2017
1,143
RE: Kamilkarp: ""Could you tell me why I'm getting 60-70MHz in my circuit?""
Because the F(T)=200 MHz dividing to 70 is still K(U)=3 but 200/170=~1.
Solution: take a common base circuit.
For example: tank L and C is between collector and Vcc. From emitter to gnd goes 0,1...1...10 kOhm. From C to E is set small variable capacitor ca 5...30 pF for accurate generation initiation point adjusting (one end-too mild, other end- too hard regime). The base is set by one resistor to gnd and another to Vcc, but good high frequency capacitor is added from base to gnd of interval 2000pf...100 nF who creates the common base regime by AC, whilst common emitter is obvious for DC.
This is very stable circuit of classics, what works just with any combinations of nominals.

Only way around it is to take an any bjt with F(T) at least some 600 MHz or higher.