Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Need an accurate oscillator on 15.186667MHz

Status
Not open for further replies.

MikeMl

Well-Known Member
Most Helpful Member
What is the easiest way of generating a low-phase noise RF carrier at about 1V into 1K Ohms?
It used to be I would call up International Crystal Manufacturing (RIP) and order a custom crystal.

Who does that nowadays?

Would any of the "DDS" modules do it?

Other methods?

I can re-purpose some GE ICOM channel elements (TXCO modules) or build an Colpitts oscillator from scratch if I have to...
 
I don't know how accurate is accurate or how low is low, but the Digikey will program an SiT8208 (a 20ppm MEMS oscillator, not a crystal) and you can specify the frequency to 6 decimal places.

https://www.digikey.ca/products/en/...a9&quantity=&ColumnSort=0&page=1&pageSize=250

1553722956908.png


EDIT: Actually maybe not. You said RF carrier so that means you need a sinusoid right? This is a square wave device.
 
Last edited:
How accurate, how stable does this need to be?

Would an LC oscillator be good enough. Or an LC oscillator under control
of a micro which counted it against a crystal and adjusted it
every 1 mS to 1 S?

15.186667MHz

15186667 * 3 is 45560001, is that 1 Hz important?

Looking at the frequencies (and standard crystals)

45 Mhz xctl osc + 16.8 Mhz xctl (availble)

=> 45e6 + (16.8e6/3.)/10 -> 45560000

=> (45e6 + (16.8e6/3.)/10)/3. -> 15186666.666666666

The (16.8e6/3.)/10 term 45.560000 might be hard to filter
from the 45 Mhz term etc..
 
It is a replacement for receive channel crystal in an old GE MASTR II VHF repeater.
In years past, for Ham radio use, I would order a crystal, take apart the ICOM channel element module, install it in the ICOM, tweak to the center of the channel, and we were good to go...

In application, this oscillator is multiplied by 9 and used for low-side injection into a mixer in a receiver with a 11.2MHz first IF. The channel center is 147.880 MHz.
 
Last edited:
It used to be I would call up International Crystal Manufacturing (RIP) and order a custom crystal.

Who does that nowadays?
I cannot advise on crystal suppliers in the USA, but there must surely be one who will take an order for a 1 off crystal (for a price).

Would any of the "DDS" modules do it?
A nice cheap AD9850 DDS module would make the frequency you want, no problem.
But, and it is a big but, after multiplication there will be significant phase noise and low level spurious frequencies.

For a simple solution, just suck up the cost of a one-off crystal.

JimB
 



The part chosen by dknguyen is better in performance , but is a 3.3V programmable XO. ( internal fractional N PLL)


MEMS OSC have higher Q's than typical AT cut XTALs. They are superior mechanical, non-piezo effects, better thermal, EMI, Phase noise. Field Programmable.


Consult with Digikey Tech support about getting either the XO or VCXO for your frequency.

There may be a learning curve otherwise. https://www.digikey.com/en/supplier-centers/s/sitime



Since you have a 9th order harmonic oscillator, , the XO needs to have harmonics ( e.g square wave)
It has far superior phase noise, more than adequate tolerance and stability.


Just order the frequency you need for 3.3V and interface with an LDO and series cap or R as needed to inject 9th harmonic.
https://www.digikey.ca/en/ptm/s/sitime/8-reasons-to-replace-crystals-with-mems-oscillators/tutorial

https://www.digikey.ca/products/en/...a9&quantity=&ColumnSort=0&page=1&pageSize=250
 
Last edited:
If the freq you want is not an integer division or multiplication of the xtal in the Dds then you'll have some phase noise.
One way I can think of is to go ahead and use a Dds, then construct a Pll with a reasonably stable Vco running at your required freq, and lock it to the Dds with a slow Pll loop filter, the Pll will reduce the noise.
 
These are the folk I have used, decades ago!
**broken link removed**

The web suggests they may still be in business - no guarantees..
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top