I have made a design with HMC494LP3 divider. The design works well if I apply any signal to the positive input (I am using single input only). The problem with design starts when I turn off signal - the board is self osciallating.
I try to reduce this effect by adding additional resistor (between GND and negative input). The values which I try was: 10k 8.25k4k2.45k50ohmI tried with parallel capacitance and without it.
But the result was always the same - self oscillation at ~1.875GHz. I tried also to remove this resistor completely and self osciations are still present, but at lower frequency ~0.9375GHz (two times less than with a resistor).
Can You help to find me where the problem can be?
- I am sure the problem is related with this IC since when I have it DC blocked and when I remove capacitor before HMC494 self oscilations are present, and when I remove capacitor after HMC494 the oscialtions disapears. I can post a schematic or give any additional data if needed!
Since I am using only 1 output (negative one) I tried to terminate output with the AC termination or resistive termination, but it do not remove oscillations.
Some dividers will self-oscillate in the absence of an RF input signal. For the HMC492, 493 and 494 dividers a shunt 8.25k ohm resistor placed on one of the differential inputs was the appropriate value to prevent self oscillation.
For additional information refer to application note at the link below.
Also, we recommend DC blocking the IO's on these dividers, please refer to the FAQ on our microwave frequency dividers posted on EZ.