there have been few questions here lately asking capacitor values use on each side of linear regulators.
the obvious answer follow recommendations given in data sheet particular regulator led me wonder values makes them oscillate. not having linear regulators hand had using ltspice. schematic used analysis attached. capacitors have realistic values built in esr , inductance.
of course, drawn behaves nice smooth start-up curve 50 mv overshoot before settling , no sign of oscillation load resistor being varied 5 k 5 ohm. started throw things @ it.
i removed both decoupling capacitors , there no change!
added 220 uf capacitor across load , resulted in overshoot on start increasing 150 mv few cycles of ringing. low frequency effect , wasn't affected decoupling capacitors.
so, make unstable? answer seems addition of small amount of stray capacitance (i tried 2 pf) between input , output of regulator. gives rise rf oscillations @ 7 mhz capacitive load above suppressed decoupling capacitors.
so, in addition having right value decoupling capacitors perhaps worth trying minimise stray capacitance across regulator in layout.
russell
the obvious answer follow recommendations given in data sheet particular regulator led me wonder values makes them oscillate. not having linear regulators hand had using ltspice. schematic used analysis attached. capacitors have realistic values built in esr , inductance.
of course, drawn behaves nice smooth start-up curve 50 mv overshoot before settling , no sign of oscillation load resistor being varied 5 k 5 ohm. started throw things @ it.
i removed both decoupling capacitors , there no change!
added 220 uf capacitor across load , resulted in overshoot on start increasing 150 mv few cycles of ringing. low frequency effect , wasn't affected decoupling capacitors.
so, make unstable? answer seems addition of small amount of stray capacitance (i tried 2 pf) between input , output of regulator. gives rise rf oscillations @ 7 mhz capacitive load above suppressed decoupling capacitors.
so, in addition having right value decoupling capacitors perhaps worth trying minimise stray capacitance across regulator in layout.
russell
ultra-low drop-out regulators don't use standard emitter-follower
pass-transistor ones oscillate, since designs
are inherently non-linear , gain round loop not constant depends on load
current , voltage difference stability criterion compromise between
many operating regions.
often suggest both lower , upper limit in capacitor esr these devices
the trick of providing high-frequency feedback internal node rather
than output pin harder pull off when there no voltage overhead
to spare.
pass-transistor ones oscillate, since designs
are inherently non-linear , gain round loop not constant depends on load
current , voltage difference stability criterion compromise between
many operating regions.
often suggest both lower , upper limit in capacitor esr these devices
the trick of providing high-frequency feedback internal node rather
than output pin harder pull off when there no voltage overhead
to spare.
Arduino Forum > Using Arduino > General Electronics > Regulator decoupling and stability
arduino
Comments
Post a Comment