DS92001 Datasheet, PDF (9/17 Pages) National Semiconductor (TI) – 3.3V B/LVDS-BLVDS Buffer

DS92001
www.ti.com
SNLS147F – JUNE 2002 – REVISED APRIL 2013
Match electrical lengths between traces to reduce skew. Skew between the signals of a pair means a phase
difference between signals which destroys the magnetic field cancellation benefits of differential signals and EMI
will result. Do not rely solely on the auto-route function for differential traces. Carefully review dimensions to
match differential impedance and provide isolation for the differential lines. Minimize the number of vias and
other discontinuities on the line.
Avoid 90° turns (these cause impedance discontinuities). Use arcs or 45° bevels.
Within a pair of traces, the distance between the two traces should be minimized to maintain common-mode
rejection of the receivers. On the printed circuit board, this distance should remain constant to avoid
discontinuities in differential impedance. Minor violations at connection points are allowable.
TERMINATION
Use a termination resistor which best matches the differential impedance or your transmission line. The resistor
should be between 90Ω and 130Ω for point-to-point links. Multidrop (driver in the middle) or multipoint
configurations are typically terminated at both ends. The termination value may be lower than 100Ω due to
loading effects and in the 50Ω to 100Ω range. Remember that the current mode outputs need the termination
resistor to generate the differential voltage.
Surface mount 1% - 2% resistors are the best. PCB stubs, component lead, and the distance from the
termination to the receiver inputs should be minimized. The distance between the termination resistor and the
receiver should be < 10mm (12mm MAX).
PROBING LVDS TRANSMISSION LINES
Always use high impedance (> 100kΩ), low capacitance (< 2 pF) scope probes with a wide bandwidth (1 GHz)
scope. Improper probing will give deceiving results.
Copyright © 2002–2013, Texas Instruments Incorporated
Product Folder Links: DS92001
Submit Documentation Feedback
9