Say I have std. 2-pin crystal can. (Mine is 11.289 Mhz)
I want to make the simplest osc/clock possible ... hence, fewest parts-count I assume. The performance should be acceptible, but does not have to be audiophile-grade.
Please suggest a circuit.
Thanks!
I want to make the simplest osc/clock possible ... hence, fewest parts-count I assume. The performance should be acceptible, but does not have to be audiophile-grade.
Please suggest a circuit.
Thanks!
Last edited:
See figure 14 and table 9 of
https://assets.nexperia.com/documents/data-sheet/74HCU04.pdf
where the frequencies in table 9 are in MHz rather than kHz.
https://assets.nexperia.com/documents/data-sheet/74HCU04.pdf
where the frequencies in table 9 are in MHz rather than kHz.
Thanks ... but wouldn't a buffered hex inverter be better than the UNBUFFERED 74HCU04 one you suggested?See figure 14 and table 9 of
https://assets.nexperia.com/documents/data-sheet/74HCU04.pdf
where the frequencies in table 9 are in MHz rather than kHz.
Also, would a "high-speed" hex inverter, such as the 74VHC04 , be a better option?
PO74GU04
I assume this is one:
74U04 Series GHz TTL CMOS logic IC 14pin SOIC QTY-1 | eBay
Only concern is that requires 3.3v max. I'll have to add separate regulation for voltage that low. The lowest I can provide currently is the std. 5vdc.
I've seen the Potato chip mentioned on DIYA quite often.IMHO, unbuffered and faster inverter is better. Try PO74GU04.
I assume this is one:
74U04 Series GHz TTL CMOS logic IC 14pin SOIC QTY-1 | eBay
Only concern is that requires 3.3v max. I'll have to add separate regulation for voltage that low. The lowest I can provide currently is the std. 5vdc.
Thanks ... but wouldn't a buffered hex inverter be better than the UNBUFFERED 74HCU04 one you suggested?
Also, would a "high-speed" hex inverter, such as the 74VHC04 , be a better option?
A buffered inverter consists of three inverters in a row. With the extra gain and phase shift of the two additional stages, the start up of the oscillator becomes quite hard to predict. It might oscillate at a frequency set by the inverters rather than the crystal, or become conditionally stable and not oscillate at all. Even if it does work, the phase noise will be worse than with an unbuffered inverter due to the small size of the input stage of a buffered inverter. In fact the 74HCU04 is meant specifically for crystal oscillators and other analogue applications.
What you can and usually should do, is to connect the output of the oscillator to another inverter from the same package and use the output of that second inverter to drive whatever circuits you want to drive with it. The second inverter converts the signal coming out of the oscillator into a nice square wave.
Regarding other logic families, I did some noise measurements a long time ago and found that the Texas Instruments SN74AHCU04 outperformed the SN74HCU04 by far. The measurements concentrated on the PMOS side of the inverter, though. See SN74AHCU04 as a cheap FET differential pair for details.
Last edited:
Say I have std. 2-pin crystal can. (Mine is 11.289 Mhz)
I want to make the simplest osc/clock possible ... hence, fewest parts-count I assume. The performance should be acceptible, but does not have to be audiophile-grade.
Please suggest a circuit.
Thanks!
Ditch the 2-pin crystal for an oscillator module. 1 can + power.
Attachments
- Home
- Source & Line
- Digital Line Level
- Lowest part-count oscillator/clock