sensing the signal from a IC Pin

Any circuit you connect to this pin will, at he very least, add some parallel capacitance and leakage resistance. But, unless the original R is very high, and C is very small, these effects should be able to be made, insignificant, or their effects can be compensated for by altering the original R and C and it should be possible for the net effect to be just a slight change in frequency.

What PWM chip does this involve?

Ohhhh.....

This sounds familiar ! Which PWM chip is that ? There might be a very simple answer. More info please.

Graham

hi, thanks for your replies. Its actually a PFC IC FA5502P from Fuji . I read the dataheet, it mentions capacitor can range from 330pF to 1000pF and resistance can vary from 10-75Kohm. The buffer i am connecting to this pin has 1Megh input resistance and 3pF capacitance. The capacitor now connected to this pin is 500pF. Adding this circuit there keeps the net cap within the range. I am not able to understand random behaviour.

Some questions for you to answer / look at:

Are you 'adding' this circuit with wires?

Are the timing components close to the Ct pin?

Are you driving the resistor from the ref pin?

What value of resistor are you using? (The app note shows a 22k device). This is particularly important - in my experience RC oscillators can get very unstable in very low current situations (depends on the design of the oscillator, obviously, but we're dealing with a precanned device here)

Are you using the SYNC pin to lock to an external signal? If not, do you have the pin tied hard to ground (well away from any high switching currents)? - Not doing this will **definitely** give you the problem you are seeing.

The timing diagrams shown have a nice beefy signal at the oscillator (Ct) pin, so I can only assume you haven't followed some (perhaps hidden) piece of application advice. The above might get you started.

Cheers

PeteS

Hello Krish (?),

You also need to consider the gate-drain capacitance of your buffer. When the drain (or collector) swings a lot this feeds back into that sensitive node.

Regards, Joerg

Are you using an isolation transformer for your testing?

Maybe you should (WITH an isolation transformer in there AND no other grounds!!!!!- or you will almost surely damage the circuit) ground the chip ground to earth.

The stray capacitance and high resistance you are adding may not seem significant in comparison to hundreds of pF, but keep in mind that the voltage it is seeing may be hundreds of volts rather than volts, depending on your exact setup.

Best regards, Spehro Pefhany