To reduce the impact of thermal noises, KAGRA gravitational wave detector is operated at cryogenic temperature. This makes the use of crystalline substrates mandatory since non-crystalline materials have worst performances. Within the KAGRA experiment, the crystalline material of choice is sapphire. Next generation of gravitational wave detectors will also use crystalline substrates, possibly sapphire or silicon. All these materials are birefringent which can spoil both the sensitivity and duty-cycle of the detectors and therefore substrates with lowest possible birefringence are mandatory. Within the KAGRA collaboration, we have two experiments able to measure the birefringence of the 22kg sapphire substrates with a duration of weeks. It is planned to increase the mass of the test-masses to the hundred-kg scale making the current birefringence characterization measurements not realistic to use. Here, we propose to use a pair of identical liquid crystals to measure and compensate birefringence of substrates with arbitrary size. We are now developing such experiment which will decrease the characterization duration by about ten times and possibly down to the second scale for any substrates size. This experiment will also demonstrate the possibility to compensate birefringence in gravitational waves detectors.