Very rough sketch ... 6 refers to a 6mhz clock oscillator, 24 to a 24 MHz clock oscillator. The two clicks could lock and produce an error signal which could modulate the 24 MHz clock
****************************************
Hi & thank you for your response....
....looking very quickly at it, I do not see how this setup can reduce close-in PN.
The PN characteristics of this schematics will be defined by the 24 MHz oscillator in your picture, just as in a Locked Loop.
If your goal is to reduce close-in PN, or at least its effects, there are other things you can do:
1. In the Clapps schematics, take the signal current from the xtal.
2. Run the xtal close to max DL.
3. Use multiple coupled oscillators (2 or more)
4. Use local feedback
5. Be carefull with your choice of Post amps.
Follwing these steps, you will be able to massively reduce PN of you 24 MHz clock system.
Cheers...
*****************************
Interesting article, though very different from jbordens sketch...
Difficult to get right, easier to screw up, expensive components...
My recommendations:
Stick with 1 to 5 above, as they are cheap and very easily implemented.
Good luck
Last edited:
Hi Jesper,
confirmed delivery date is May 17th, so I believe 3-4 days to arrive in Italy.
Andrea,
a question about the TWTMC... with regard to the power supply requirements.
How much current does the clock need (5Volts operating voltage) And to which extend do I need to suppress PSU noise and ripple?
BTW, I looked for them in the thread but was unable to find them.....
a question about the TWTMC... with regard to the power supply requirements.
How much current does the clock need (5Volts operating voltage) And to which extend do I need to suppress PSU noise and ripple?
BTW, I looked for them in the thread but was unable to find them.....
Last edited:
About the power supply:
- the TWTMC-C (Colpitts-Clapp oscillator) needs 2 rails, +15V around 20mA for the oscillator and +5V (or +3V3) 10-15mA for the squarer
- the TWTMC-D (Driscoll oscillator) needs 2 rails, +6V 10 to 25mA (depending on the Crystal ESR) for the oscillator and +5V (or +3V3) 10-15mA for the squarer
In any case you should use a clean power supply, since PSU affetcs the close in phase noise performance.
14 to 15.5 should be fine. About the noise, the better you can find. Demian's PSU I have embedded in the daughter board TWTMC-D&D has very low noise. You can use also a very clean shunt regulator, since the current is almost constant.
Hi again ... I have a question possibly for Herbert ... I have been reading a bit in your "Reproducible Low Noise Oscillators" and notice your section on magnetic materials in relation to inductor selection for the oscillator. To this end I've been looking for a 2.7 uH air core inductor and I've found this one from Digikey:
Electronic Components and Parts Search | DigiKey Electronics
Do you think it can be used for the oscillator (it has a 3.2 ohm resistance)?
Thanks for any feedback & cheers,
Jesper
Electronic Components and Parts Search | DigiKey Electronics
Do you think it can be used for the oscillator (it has a 3.2 ohm resistance)?
Thanks for any feedback & cheers,
Jesper
Hi Jesper,
data provided are the individual measurement of each crystal made by Laptech, as I requested when I placed the order.
You can think a crystal as a series R-L-C filter with a capacitance in parallel. As you can see in the attached data, eg 6.144MHz, RR is the ESR of the crystal (R of the RLC filter), C1 is the motional capacitance (C of the RLC filter) and C0 is the capacitance of the crystal's terminals (in parrallel to the RLC filter).
Q is the Quality factor of the modelled RLC filter, and characterizes the resonator's bandwidth relative to its center frequency. The higher the Q the narrower the bandwidth, the better the crystal. Higher Q means a lower rate of energy loss relative to the stored energy of the resonator.
Typically, the higher the Q of the resonator, the lower the phase noise of the oscillator. The goal is to get highest loaded Q as possible, starting from an high Q resonator. The Driscoll and the Butler circuits provide high loaded Q, since the resonator sees a very low impedance (crystal in the emitter circuit).
I will start a new Group Buy soon, including TWTMC-D Driscoll oscillator board and TWTMC-OVN oven board.
The frequency depends on the oscillator used. The Driscoll oscillator works with SC-Cut crystals from 5 MHz up to 100 MHz.
I have not yet defined the price of the oven PCB, I will start a new GB soon, crystals and boards.
- Status
- Not open for further replies.