diff --git a/vna_system/core/processors/base_processor.py b/vna_system/core/processors/base_processor.py
index 9842cac..b751da5 100644
--- a/vna_system/core/processors/base_processor.py
+++ b/vna_system/core/processors/base_processor.py
@@ -354,11 +354,21 @@ class BaseProcessor:
                 latest["vna_config"],
             )
 
-    def _process_data(self, sweep_data: Any, calibrated_data: Any, vna_config: dict[str, Any]) -> ProcessedResult:
+    def _process_data(self, sweep_data: Any, calibrated_data: Any, vna_config: dict[str, Any]) -> ProcessedResult | None:
         """
         Internal: compute processed data, build a Plotly config, and wrap into `ProcessedResult`.
+
+        Returns None if processing resulted in an error (indicated by {"error": ...} dict).
         """
         processed = self.process_sweep(sweep_data, calibrated_data, vna_config)
+
+        # Skip result creation if processing returned an error
+        if isinstance(processed, dict) and "error" in processed:
+            logger.debug("Processing returned error; skipping result",
+                        processor_id=self.processor_id,
+                        error=processed.get("error"))
+            return None
+
         plotly_conf = self.generate_plotly_config(processed, vna_config)
         ui_params = self.get_ui_parameters()
 
diff --git a/vna_system/core/processors/implementations/bscan_processor.py b/vna_system/core/processors/implementations/bscan_processor.py
index e15e6f1..6b81e99 100644
--- a/vna_system/core/processors/implementations/bscan_processor.py
+++ b/vna_system/core/processors/implementations/bscan_processor.py
@@ -165,6 +165,10 @@ class BScanProcessor(BaseProcessor):
             Or: {"error": "..."} on failure.
         """
         try:
+            # Skip B-scan processing for S21 mode - only works with S11
+            if vna_config and vna_config.get("mode") == "s21":
+                return {"error": "B-scan only available in S11 mode"}
+
             # Choose calibrated data when provided
             data_to_process = calibrated_data or sweep_data