A system performance analysis for Direction of Arrival (DOA) estimation using UAV swarms in the presence of small sensor gain, phase, and position errors has been derived analytically. Its analysis accounts for pragmatic errors and uncertainties when the UAVs fly in practice. This includes the signal model with deterministic unknown location errors, which is extended to cases when the location error is stochastic. The Cramer-Rao Bound (CRB) is derived for the joint estimation of DOAs, accounting for variation in sensor gain, sensor phase, and sensor location errors for the swarm. This work examines the results from the first experimental campaign to study the behavior of these systems in a laboratory setting. Both numerical simulations and practical experiments have been obtained to verify the theoretical results. The focus of this manuscript is on the latter.