Hi Guys, I want to double an AC/RF 10 MHz frequency of low voltage (about 1 Volt) on a coax line.
It also wouldn't hurt to have it amplified somewhat, so I was thinking this should be easily done
with the front half of a shortwave radio in that the incoming freq is magnetically coupled, tuned with a tank circuit,
amplified, then the 20 MHz harmonic passed with a second filter
which is what a radio does when it passes the IF, but rejects the other sum or difference frequency for being redundant.

I want to do it solid state, but at least with an op amp in a real socketed DIP!
Any advice on where to look for a calculator for the coils?
I have tuning gangs, but I can't necessarily match any online LC circuit. I could measure what I have though.
Cheers, Art.

Not sure where this is going but one tends to use "clocking" using a nand gate, or multi-vibrator where you can set them to divide, or when triggered give two pulses when one is applied etc.

Marc

Hi Marcc,
Not that I've looked into a digital circuit, but I presume logic would output a square waveform where I'd prefer AC.
Where it is going is for a more modern project to clock two microcontrollers at two different speeds but keep their instructions in beat with each other, particularly to ease communication between the two.

This gets a bit off topic, but the circuit I describe, I think should be able to be tuned to input one frequency to an amp with one tank circuit, and the other tank circuit should select between a number
of harmonics present in the amplified signal. I think a signal generator does this to select a higher range than the oscillator inside it, just switch in a new filter for a new harmonic range,
and an old radio doing the same thing by passing either the sum/difference freq, and rejecting the other.

Quite correct. Logic is just 0's and 1's. The gates on a 4000-series IC (or equivalent) will go high or remain low depending on the state of the inputs. Generating a sine wave will require more elaborate circuitry, especially since you will need to go into the negative for half the cycle.

I've never done it myself, though there are a few circuits out there that will convert a square wave to a sine wave.

I have actually made a 455kHz Crystal locked oscillator using three sections of a Quad NAND (I think..still have circuit) That fed a dual gate FET as the (controlable) mixer, the other gate having the audio from a 555.

To knock the corners of of the square blocks the audio is fed through about three RC filters that are just resistors & caps.

I am picturing notch filters & perhaps Cavity Resonators, some forms of the latter, I consider attenuators as the design I see as flawed.

Most of my Physics books ARRL Handbook & various things have info as well as capacitance reactance charts & I did score a somewhat dangerous 110V Heathkit Grid Dip Meter & bonus coils recently and that is now a useful (repaired) piece of clobber.

Marc

Hi Art,

Re - "Any advice on where to look for a calculator for the coils .....??"

Not sure if I understand your question as I'm sure you are aware of "LC calculators" and "L calculators" on Google. Constructing or buying an L???

When constructing LC circuits, particularly when using an existing tuning C, you need to be aware of the Q of the circuits. If using an existing tuning C ....... and you construct an L to obtain the frequency required .....you may end up with a very low Q and poor tunability. The reverse is also true. An understanding of the different L and C value combinations for a particular frequency is required .... or trial and error.

Cheers,
Ian

Yes it's an LC calculator I'm thinking of, but I was kinda hoping for something more focussed on radio
coupling coils and tank circuits. In Dick Smith Funway 2 book there is a shortwave TRF radio with a tank circuit that must pass 10 MHz, and I'd say it's a common winding for pocket SW radios.
I would replicate that except the tuning gangs I have are of higher quality from amateur radios but of different capacitor ranges.

The Funway 2 radio does the same thing as a 30's valve radio.. filter and amplify, filter and amplify.
It also cheats and uses a 4007 as a three stage analogue amplifier even though it's meant to be a digital chip.

If I wanted to double the frequency of a 10MHz clock (for use with microprocessors, so it'd be something like TTL style signals), assuming the clock is 50% duty cycle, I would run the clock through a delay line of time delay 1/4 of a full clock cycle, then feed the delayed and the undelayed clock into a XOR gate. The gate's output should look like a 20MHz clock. The timing of one 20MHz pulse to the next pulse may be uneven over 2 cycles, but micros don't really much care about that, as long as the time deviation is less than say 10%.

Or you could start with a 20MHz clock, then feed it into a flip flop wired to divide it by two, to create the 10MHz.

With an oscillator generating 2nd harmonics something like a 20MHz version of an IFT is all that is likely required.


Marc

Hi Wa2ise, Yes the logic signal is fine for the micro, I have tried both AC and DC PWM at the same frequency with the same results.
In this case since I have access to an AC frequency I'd like to maintain a sinusoidal waveform for use with other things.
You could always buffer the AC signal if you wanted to, but it wouldn't be as easy to reproduce the AC wave.

Marcc, you're right, every superhet will create two modulated carriers at it's mixer, and one of them is unwanted and filtered out. Then the remaining (chosen IF) frequency is amplified.
A TRF radio cannot expect a particular frequency through it's amplification stages and uses tuned circuits. It should be extremely obvious that a shortwave radio front end can be tuned to 10 MHz.
I have also learned that my siggen indeed does produce frequencies higher than it's oscillator by passing an harmonic. It might even be possible to select more than one harmonic as well.

So basically a tank circuit resonant at the desired freq is all that's required.

https://www.youtube.com/watch?v=JzSBWJu24gU&list=PL70E4DEE7486FB18C.

Do be careful of what is on the web, there is quite a bit of misinformation out there.

I have generated a square wave & fed it through an attenuator to knock the corners of it.

One of the more savage filters is to use a bandpass one. I have had one in an AWA and its ability to attenuate the unwanted was fantastic.

Marc

Circuit Girl is very well known.. actually works for Valve software last time I heard,
plus she's showing you the result right on the scope as she's commentating.

In addition to this, I hear you can make a crystal ring on an harmonic with half of it's labelled frequency.
Not that I see too many reasons to do so, but it's doing the same thing.

Ok, so to prove to myself, I wound an unknown inductor a while back to tap into later.
I used that on the floor with an old radio tuning gang and even a clip lead on one side of the tuning circuit.
No need for the rest of Circuit Girl's stuff, I have an RF siggen

As you can see, both the tank circuit and the waveform are rubbish, but enough to prove the point.
Even better in my room where you can vary the value of the incoming RF and the tank circuit.
Certainly worth shortening the leads, and shielding everything.

Image Link

Image Link