We are working on a design using the LMX2485E as the local oscillator in a VHF radio receiver. We have come across an issue where the LMX2485E appears to get stuck in a quazi-stable state where it indicates that the pll is out of lock when it is in fact in lock. We are currently working to identify a mechanism to reliably get the LMX into this state but have not yet identified one.
The synthesizer can generally be instructed to change frequency many thousands of times and operates normally, sporadically it will lock up normally but the out of lock line out of the IC will remain in the out of lock state. any subsequent frequency change in the current VCO sub-band will leave the IC reporting it is out of lock though the pll has locked successfully. A clue is if the synthesizer is then instructed to change to a frequency in another VCO sub-band which will interrupt the RF signal to the LMX momentarily the out of lock condition disappears and the LMX2485E appears to have broken out of the Quasi-stable it was stuck in and the synthesiser can now be set to frequencies in the susceptible VCO sub-band and works normally.
The layout and decoupling for the LMX2485E recommended by TI has been generally followed apart from one VCC 100pf decoupler which has not been place as closely to its IC pin. This has been cut and strapped and has not fixed the issue. The IC is also mounted on a multilayer pcb over a good ground plane. We have access to several different date of manufacture of ICs and the issue occurs on all so we do not believe it is batch related.
We would be grateful for any advice, comment or assistance to help us resolve this issue.
Regards
Robert Miller
