1590 kHz clever circuit

[meowth08]

Hi,

We are tasked to make an oscillator with frequency of 1590kHz.
Then we started to make several designs and we came up with the clapp oscillator.
I have started the thread here before and MikeMl helped me.

Then, our fickle minded teacher told us that we can do or use any components for a circuit as long as we come up with the requirement.

frequency = 1590 kHz
Voltage peak to peak minimum= 3

Any clever circuits you can suggest?

 

[JimB]

What is considered as "clever" ?

For a medium frequency RF circuit like this, I would first think of a conventional Colpitts circuit.
There is no reason why you should not use any of the other conventional oscillator circuits such as Hartley, Franklin, Vackar (Tesla), and their many variants.

You could also use a circuit such as the Wien Bridge, or the RC Phase Shift oscillator, neither of which requires an inductor.


On the other hand, you could use a VCO, such as those used in function generators.

Or if you want to be really silly, use a DDS such as an AD9850, you will also need some support device such as a PIC load the tuning words.
Simple it is not, but it could be considered as "clever".

JimB

 

[panic mode]

do a PLL, that's clever and you get stable yet selectable frequency

 

[Roff]

do a PLL, that's clever and you get stable yet selectable frequency

The catch 22 is, you have to have a reference frequency that's stable. So he still needs an oscillator.
I would use a Colpitts oscillator, as suggested by Jim. Not as the input to a PLL - just the oscillator. Make it adjustable, if necessary.

 

[meowth08]

What is considered as "clever" ?

I thought there would be other possible designs that would result to lesser components or lower costs compared to the clapp oscillator. That for me would be clever .

Thank you for your inputs, I would stick on the previous design: clapp oscillator.

I checked on a PLL on google and on a single look, it would cost more.

And yes! We'll make the frequency adjustable just to make sure we can tune to the desired freq.

Thanks again.

 

[Winterstone]

I thought there would be other possible designs that would result to lesser components or lower costs compared to the clapp oscillator. That for me would be clever .

And yes! We'll make the frequency adjustable just to make sure we can tune to the desired freq.

For a tunable linear oscillator in the lower kHz range I always would propose an opamp-based oscillator.
It is much simpler to design if compared with transistor-based circuits - in particular if frequency tuning is required.
Question: Specific tuning requirements (capacitive, resistive)? Which range? THD requirements?

 

[jpanhalt]

A Schmitt inverter plus one resistor and one capacitor will work: https://www.electro-tech-online.com/custompdfs/2013/01/AN-118.pdf

John

 

[MikeMl]

A Schmitt inverter plus one resistor and one capacitor will work:...

This works, but is not as frequency stable as a Colpitts LC oscillator, especially at 1.5MHz. What is the frequency stability spec? Particularly as a function of supply voltage?

 

[Winterstone]

For a tunable linear oscillator in the lower kHz range I always would propose an opamp-based oscillator.
It is much simpler to design if compared with transistor-based circuits - in particular if frequency tuning is required.
Question: Specific tuning requirements (capacitive, resistive)? Which range? THD requirements?

Sorry - only now I have realized that the desired frequency is not in the lower kHz range (1590 Hz) but 1590 kHz.
Nevertheless, I think that an opamp-based solution is still possible (but an opamp with a GBW of approx. 100MHz will be required).

 

[jpanhalt]

This works, but is not as frequency stable as a Colpitts LC oscillator, especially at 1.5MHz. What is the frequency stability spec? Particularly as a function of supply voltage?

I don't recall the OP giving any stability specification only that it must be adjustable, low cost, and low component count.

Of course, low cost is always difficult to define, because it is so volume sensitive. But I will bet that Schmitt oscillator will give any other design a run for its money.

John

 

[meowth08]

Hi,

As for the stability, teacher says it should be stable for not less than 2 minutes.

I'll post the circuit later. I don't have it right now.

 

[RCinFLA]

Cannablize an A.M. radio's local oscillator.

 

[meowth08]

Cannablize an A.M. radio's local oscillator.

I'll try if I can get old AM radio from my grandpa.

Here's the design.