Oscillator fails after a week

[jmvinay1]

Hi everyone,

I had designed a cross coupled oscillator on PCB with dielectric constant 3.2 using transistor NE662M04 , johnsons(company) inductors and capcitors at 300Mhz and 2GHz(two different circuits). They worked completely fine on the day of fabrication with slight shift in frequency of oscillation. After a week i tested the same circuits, found both of them not working.

Peculiar behaviors are as follows(there were not seen on the day of fabrication),
1. Circuits current jumped suddenly from 0 to 130mA for the change of voltage from 0.73V to 0.74V. They were showing hysteresis behavior upon decreasing voltage.
2. Oscillation frequencies were same in both the circuit. And were not correlated to LC tank values.

What could be the possible reason for this behavior?Is it a heat related issue during soldering?

Thank you.
Vinay

 

[vegeman]

Hi everyone,

I had designed a cross coupled oscillator on PCB with dielectric constant 3.2 using transistor NE662M04 , johnsons(company) inductors and capcitors at 300Mhz and 2GHz(two different circuits). They worked completely fine on the day of fabrication with slight shift in frequency of oscillation. After a week i tested the same circuits, found both of them not working.

Peculiar behaviors are as follows(there were not seen on the day of fabrication),
1. Circuits current jumped suddenly from 0 to 130mA for the change of voltage from 0.73V to 0.74V. They were showing hysteresis behavior upon decreasing voltage.
2. Oscillation frequencies were same in both the circuit. And were not correlated to LC tank values.

What could be the possible reason for this behavior?Is it a heat related issue during soldering?

Thank you.
Vinay

What about heat during operation? High frequency oscillators will become unstable at high temps. Extreme heat will obviously cause failure.

 

[jiripolivka]

The bias circuit for the VCO must be well designed and temperature-stabilized. From your few data it looks you run the transistors without any current-limiting resistors. In no good circuit is the DC current allowed to rise to 130 mA!
Typical DC current ranges from 10 to 50 mA. Take care to test your design over the full temp range!

- - - Updated - - -

If the device frequency was "not correlated to LC tank circuits", how can you name such thing an oscillator?!
What you had is a poor transistor begging for life by spurious oscillation!