Continue to Site

# Very wide band oscillator frequency limits.

#### Lightium

##### Active Member
I am playing with the following circuit.

I can get it to osc. from 340Hz to 60MHz, but I can't go higher or lower. Has anyone seen this circuit before and know what would an equation be to express this circuit? It seems that the frequency is controlled C1, L1. f=1/2*pi*sqr(C*L) and has well as R2, but I don't know how to include R2 into the formula.

Last edited:
I'd guess that it is not R2 directly, but changing the biasing changes the voltages on the transistor junctions and causing a "varicap" type effect.
(It's a seriously weird circuit!)

The GBW verses Ic seems a tad low, eg. running out of G at Fhigh :

Seems like biasing has ~ 2.3V / 1K = ~ 2.3 mA in emitter Q1.

What is source for circuit, eg. URL, doc, whatever ?

Regards, Dana.

I can get it to osc. from 340Hz to 60MHz,
You did it on a simulator only or in real too?

The 2N2222 is running about as fast as it can. Its "Ft" = 250mhz. Look for a faster transistor.

For an even more bizarre hypothetical outcome, remove R1, C1 and raise R2 which is loading the base impedance R/hFE from 1k to 10k and observe the limiting action of Q2 Vbc going forward biased. For an ideal transistor with any hFE (Beta) say 10 which is less sensitive, as this relies on the action of Ic vs Vbe, you get a dual transistor Astable Multivibrator in the GHz range without any real load capacitance based on the L/Re alternating sawtooth ramps. Thus increased Re increases the frequency with one diode drop output swing.

After option> Other options > 0.5 f (seconds sample interval) I get 1.22 E12 Hz with Re = 1 Mohm
This stresses the need to include the Miller capacitance of any transistor you want to emulate added to the circuit and any other parasitics of ESR/ESL/C to Falstad to enhance the realism. Of course one can customize the esoteric parameters of the Transistor Model as well or use the option for the Spice Model as well for both. That raised the f to 1.85 E12 Hz ( lol) https://tinyurl.com/2mu4leng since there were no parasitic C's added.

Last edited:
This might be more interesting to use a Lithium 3V or a CR23 cell with Re=50 Ohm and draw 43 mA @ 3V to create a squarewave at 150 MHz using Beta=10.

But then adding a 10 pF probe load changes this result in particular Ic2 and rising f also with lower Vcc voltage.

Last edited:
You did it on a simulator only or in real too?
I was hoping to use an incandescent light bulb from a pen light for the inductor. What do you think?

I was hoping to use an incandescent light bulb from a pen light for the inductor. What do you think?
It's a bulb not an inductor - and what's difficult about winding a few turns of wire?.

I doubt that would work well.

Low value inductors are very easy to make - this gives some examples:

I was hoping to use an incandescent light bulb from a pen light for the inductor. What do you think?
As a start, I would measure its inductance and cold resistance if I have one.

Thinking out of the box used to be my favorite since after graduation. What I usually have as components and devices, in every period of time, is very limited in comparison to what most engineers around the world are allowed to have.

Thinking on, do torch bulbs even have windings?, from what I can vaguely remember they were usually just a straight piece of wire - and pretty poor connections (length of connections and screw fitting etc.) to mess up what inductance there might be?.

Thinking on, do torch bulbs even have windings?, from what I can vaguely remember they were usually just a straight piece of wire - and pretty poor connections (length of connections and screw fitting etc.) to mess up what inductance there might be?.
I've looked under a microscope some are coils and some are compounded coils.

I doubt that would work well.

Low value inductors are very easy to make - this gives some examples:
I am planning on making a coil winding jig.

I am planning on making a coil winding jig.
For VHF coils such as those, that's typically an appropriate size drill bit or screwdriver shaft!

For such small coils it would take longer to set up a custom winding jig than just wind by hand.

For VHF coils such as those, that's typically an appropriate size drill bit or screwdriver shaft!

For such small coils it would take longer to set up a custom winding jig than just wind by hand.
I am sorry, the jig is for other coils that are linked to other threads here on ETO.

I am sorry, the jig is for other coils that are linked to other threads here on ETO.
OK, just a bit of confusion there!

- coils need to be shielded as done in UHF tuners otherwise there is too much stray interference that will affect resonant frequency..
- the pitch between windings is done to reduce the self-capacitance and raise the SRF.
- rather than playing with them, learn how to choose the resonant frequency, and impedance of L, C and then the Q gain factor.
You could wind on a coarse plastic toilet seat bolt. Rugged UHF coils use a heavier gauge as changes in each winding diameter and spacing alters the C value more than the L.
There are many coil calculators that don't do justice without computing SRF with capacitance.

L increases by square of n turns so use only 2 or 3 turns max, and reduce C between 1 and 10 pF

But don't use your original design. That is basically an astable with filtering, which I already showed. Instead, define what you want to see in measurable terms of f, V and distortion etc.

Last edited:
- coils need to be shielded as done in UHF tuners otherwise there is too much stray interference that will affect resonant frequency..
- the pitch between windings is done to reduce the self-capacitance and raise the SRF.
- rather than playing with them, learn how to choose the resonant frequency, and impedance of L, C and then the Q gain factor.
You could wind on a coarse plastic toilet seat bolt. Rugged UHF coils use a heavier gauge as changes in each winding diameter and spacing alters the C value more than the L.
There are many coil calculators that don't do justice without computing SRF with capacitance.

L increases by square of n turns so use only 2 or 3 turns max, and reduce C between 1 and 10 pF

But don't use your original design. That is basically an astable with filtering, which I already showed. Instead, define what you want to see in measurable terms of f, V and distortion etc.
I just hadn't considered the Q. I guess I'll have too, I am trying to make some very small tank circuits.

I just hadn't considered the Q. I guess I'll have too, I am trying to make some very small tank circuits.
Then why try and use a bulb?, which is far larger than winding a few turns of wire?.

Replies
18
Views
2K
Replies
7
Views
2K
Replies
6
Views
1K
Replies
10
Views
3K
Replies
33
Views
5K