Post Go back to editing

LT3743 current leakage from IC to load (probably form SW)

Hello,
 we are using LT3743 current regulator to drive a laser diode. The circuit is very similar to the example given on page 21 of its datasheet (www.analog.com/.../3743fe.pdf).
We measure a current of nearly 40-50uA from the IC to the circuit output (D1; since the load is a diode, the diode is biased for nearly half the Vf until the leakage of the diode is balanced, or up to a voltage of about 1.8V).

In order to understand where this current comes from (references given on schematic on page 21 of DS), we disconnected both 10R on sense resistor (float SENSE+ and SENSE-, excluded current injection from sensing pins, Vctrl*11uA/V), shorted C1, C2 and C3 capacitors (one group at a time), double checked Cboot capacitor has Vcc_int on one side and nothing is oscillating (and removed it too), disconnected M1 high side switch, and checked FB pin is not sourcing current. The only remaining source of current seems to be SW pin, but we do not see any information in datasheet.

EN/UVLO is 2.5V, PWM = CTRL_SEL = 0V.

Could you check if SW pin is expected to source any current?
The question has been raised since, when a discharged output capacitor is connected (e.g. 3x330uF on C3 on your schematic) to an output capacitor (always connected, 1uF in your schematic) trough a very low impedance path (this happens when PWM is set high with CTRL_SEL low), we see current peaks of more than 10A for a few nanoseconds, leading to inductive oscillations and small voltage inversions that might be dangerous for the laser diode. However, we just need to know the circuit is operating correctly here, and the output current is intended. Could you verify the integrated circuit internal operation?

Thank you for your help!