135KHz adjustable Oscillator

[engineer1000]

I want to design a variable 135khz oscilator say from 130khz to 138khz. At the moment Im using a 555 timer with a pot.Fed through a D type to get the 50% duty cycle. The problem being if I turn off the oscillator and then turn it back on the frequency has shifted slightly so I have to re-adjust. So what I want is an oscillator that once set will stay at that frequency with little or no drift . Any ideas welcome

 

[chuckey]

In principle there is no reason why you get this change of frequency. There are two causes of frequency drift :- 1, thermal, does the frequency drift back to where it is was?, does it change with temperature?. The other, 2, is sensitivity to power supply voltage, the 555 is very susceptible to this. An on board regulator could cure this as it might be that your bench PSU has a varying voltage output after switch on. Astable oscillators might suffer the same effects. You could try a wien bridge oscillator, which depends to a less extent on the power supply variations. A proper LC tuned circuit or a crystal oscillator are the ultimate for stability, providing they are designed and built with stable components.
Frank

 

[Syncopator]

There are two causes of frequency drift :- 1, thermal.
The other, 2, is sensitivity to power supply voltage, the 555 is very susceptible to this.

That's news to me Frank. One of its well known attributes is that in the astable mode it is indifferent to changes in supply voltage.

 

[rohitkhanna]

It seems you are looking for ~<1% stability (short-term, long-term) & repeatability. Use a stabilised V+ for the 555. Any decent 3-pin regulator should be good. Also use NP0/C0G (1%) capacitor(s) and precision (1%) resistor(s) in the timing part of the 555. If you want to trim the frequency, make sure any trimmer resistor is MUCH smaller in value than your precision resistor, and then seal it once adjusted.

However I notice that you want to make it variable, so you HAVE to use a pot/ preset. just make sure the trimmer is good quality, and of the smallest value required to do the job.

That should fix your troubles.

 

[chuckey]

Your right, Syncopator, worst case seems to be 1% change in frequency for a volts change in Vcc. As the temperature change due to chip is about 500 ppm, the sensitivity to Vcc is still high.
Frank