This experiment will include a patch panel change halfway through. It is understood that this will consume roughly 30 minutes or more, and that will cost the overall session 2-3 shots. This is judged to be acceptable in view that DIII-D previously made effective use of 2 hour control development sessions, with patch panel changes included in the 2 hours. 1) ELM suppression control development This experiment will be performed in a well-known ITER similar shape ELM suppression target (discharge 158103) to avoid scenario development as much as possible and isolate challenges to obtaining the optimal diagnostic coverage. If ELM suppression cannot be sustained reliably during the rotation of the n=2 RMP [Wilcox MP 2017-36-04, Wilcox SET 2017], we will (unlike the previous experiments with strike point constraints) have the freedom to optimize the suppression utilizing our improved understanding of the shape and rotation dependence [Paz-Soldan NF 2019] and other parametric dependencies [Lunia APS 2021]. The controller will also help assist us in maintaining suppression to the best of our abilities throughout the rotation. Finally, we will be prepared to use edge localized ECCD to reduce the RMP suppression threshold if necessary to maintain suppression during the RMP rotation [Logan NF Lett. 2023] and/or inject boron powder injection to reduce the RMP threshold [Effenberg APS 2023]. If time allows, we will intentionally scan the edge ECCD, which modeling predicts will have a large impact on the RMP island size at the top of the pedestal (even if aiming is limited further in than the island itself) [Hu NF Submitted]. 2) XPR controller development The reference discharge will be the baseline shot of MP2025-14-01. The XPR height estimate, relative to the selected X-point height, is available as a control variable using each controller based on bolometer measurement or diverter Thomson scattering Te measurement (DTS). Only one controller will be used for feedback control after getting responses from the feedforward system identification. Feedforward Ar seeding is planned to pick which controller response seems more likely to succeed in getting control dynamics. Here, a decision point is placed to check whether a suitable Ar response can be obtained for control. If we can’t get a good response from Ar seeding for a few shots, Ar will be changed to N, and the feedforward seeding for system identification will be repeated to get a suitable N response for control, which will suggest the main experiment will not use Ar. Once a suitable response from impurity seeding is obtained, we will turn to the closed-loop feedback control test with the initial gains calculated based on the response from the system identification. The gain will be optimized from the observation of the initial feedback control test. If time allows, the controller will be tested with further target waveform variations and various scenario parameters according to any scans or changes planned in the main experiment to confirm controller viability during such scans. If performance leaves something to be desired, we can try to refine it during the main experiment.