I made variable optical attenuator using silica PLC with mach-zehnder structure.
microheater(Ti/Au) was deposited on interferometer arms, and deep trench was formed near the microheater for thermal isolation.
microheater width is 7um, and the width of the silica ridge, containing the heater, formed by deep etch is 25um.
When current flows along the heater in DC mode, some heaters get open when the current reaches ~40mA, which is less than for sufficient attenuation.
Does anyone have similar experience?