Purpose: To evaluate the influence of peak x‐ray tube voltage and the size of the Z field of view (FOV) on body image quality of a cone‐beam C‐arm CT system with automatic exposure control. Methods: We measured dose accumulated in an elliptical‐shaped body phantom with tissue equivalent density using a small ion chamber at 23 distributed points following the AAPM TG111 approach at two tube voltage requests (109kVp, 125kVp), 4 detector dose requests (0.17, 0.36, 0.54, and 0.81μGy/frame at the detector), and 3 FOVs (small, medium, and large in Z). For dose efficiency analysis, we scanned the same phantom again after replacing the central cylindrical part with the QRM cone‐beam phantom which has 20 inserts of various diameters and contrast steps. Six experienced observers were asked to count the number of visible circles in slices reconstructed with 1 or 5mm thickness, 0.5 isotropic in‐plane pixel size, and Siemens medium smooth convolution kernel. Results: After dose normalization, fifty percent of objects with a diameter of 6.3, 4.4, 4.2, 2.2 mm at 109kVp and 6.5, 5.1, 3.8, and 3.0 mm at 125kVp having a nominal contrast of 2.0, 2.5, 3.0, and 4.5%, respectively were detectable at a diagnostic reference dose level for routine abdomen of 35mGy. Small, medium, and large FOVs at a 125kVp and 0.36 μGy/frame setting showed 46.3, 41.5, and 37.3% detectability with a mean dose of 43.5, 48.4, and 50.4 mGy, respectively. Conclusion: The detectability of the C‐arm CT images improved significantly with z‐direction collimation, and the lower kVp protocol setting of 109 kVp provided improved detectability over 125 kVp after dose normalization even for this relatively large body phantom. This work was supported by National Institutes of Health (NIH SIG S10 RR026714‐01), by Siemens Medical Solutions, AX, and by the Lucas Foundation.