*3.3. Hybrid Compensation Method*

Third, we tested the OAP using the hybrid compensation method. In the same way, the CGH basement (silica) and the fold sphere mirror were chosen according to our laboratory, and the related geometry parameters are shown in Figure 9. Ultimately, the design result of the customized CGH for the hybrid compensation system is illustrated in Figure 10.

**Figure 9.** The geometry parameters of measuring this OAP via the hybrid compensation method.

**Figure 10.** (**a**) Simulated residual wavefront of the hybrid compensation method (test CGH) in Zemax (PV = 0.0001λ); (**b**) simulation pattern of the customized CGH for this hybrid compensation system; (**c**) CGH photograph; (**d**) error of the fold sphere mirror (PV = 0.054λ).

Another multiple combined CGH was designed and produced. The accuracy of the CGH basement was the same as that of the basement in Section 3.2. As shown in Figure 10, the multiple combined CGH was involved in four parts such as the test CGH, alignment CGH and so on. Three parts were transmission CGH, except the part aligning CGH with the interferometer. The test CGH and alignment CGH with the fold sphere mirror both adopted the first order diffraction with the same grating groove depth of 692.2 nm. However, other parts adopted the third order diffraction with the corresponding groove depth of 158.2 nm. The residual wavefront of the test CGH was only 0.0001λ (PV), as shown in Figure 10a. At the same time, the error of the fold sphere mirror, which was measured via interferometry, was about 0.05λ (PV), far less than the error of this OAP.

During the measuring process, the customized CGH was placed close to the fold sphere mirror by a customized fixture, with which the relative position between the CGH and fold sphere could be guaranteed, as presented in Figure 11. At the same time, the customized fixture was mounted on a five-dimensional adjuster so that they would be regarded as the whole during the adjusting process. After repeatedly adjusting, the position error of each element can be reduced to a negligible value. Meanwhile, the errors of the fold sphere mirror and CGH were far less than the error of OAP, so the testing result can be regarded as the shape error distribution of the OAP.

**Figure 11.** Photo of measuring this OAP via hybrid compensation (with customized fixture).

### *3.4. Experimental Results with These Three Methods*

This OAP was measured using three methods, respectively. After the measured data were simply processed in MetroPro software (V9.1.1) [21] to remove piston, tilt, and power, the experimental results were obtained and are shown in Figure 12.

(**a**) Test result with the auto-collimation method

(**b**) Test result with the single CGH method

(**c**) Test result with the hybrid compensation method

**Figure 12.** The test result of Φ135 mm OAP: (**a**) auto-collimation; (**b**) single CGH; (**c**) hybrid compensation.

The resulting value of auto-collimation method as PV = 0.583λ and RMS = 0.092λ, the resulting value of the single CGH method as PV = 0.572λ and RMS = 0.089λ, and the resulting value of the hybrid compensation method was PV = 0.615λ and RMS = 0.096λ. The testing results with di fferent methods were nearly the same, although there was a slight di fference on the shape distribution, which was mainly caused by the mapping distortion of CGH. Furthermore, there were a few small residual adjustment errors, some of which were particularly induced by the angle adjustment in the hybrid compensation method, as it is a delicate process. Hence, a more accurate alignment technique is still being investigated.
