diff --git a/GPU/GPUTracking/Definitions/GPUSettingsList.h b/GPU/GPUTracking/Definitions/GPUSettingsList.h
index 974ef6a9f0d18..b871a6fd3a8cd 100644
--- a/GPU/GPUTracking/Definitions/GPUSettingsList.h
+++ b/GPU/GPUTracking/Definitions/GPUSettingsList.h
@@ -536,6 +536,7 @@ AddOption(solenoidBzNominalGPU, float, -1e6f, "", 0, "Field strength of solenoid
 AddOption(constBz, bool, false, "", 0, "force constant Bz for tests")
 AddOption(setMaxTimeBin, int32_t, -2, "", 0, "maximum time bin of continuous data, 0 for triggered events, -1 for automatic continuous mode, -2 for automatic continuous / triggered")
 AddOption(overrideNHbfPerTF, int32_t, 0, "", 0, "Overrides the number of HBF per TF if != 0")
+AddOption(overrideTPCTimeBinCur, int_32_t, 0, "", 0, "Overrides TPC time bin cut if > 0")
 AddOption(deviceType, std::string, "CPU", "", 0, "Device type, CPU | CUDA | HIP | OCL1 | OCL2")
 AddOption(forceDeviceType, bool, true, "", 0, "force device type, otherwise allows fall-back to CPU")
 AddOption(synchronousProcessing, bool, false, "", 0, "Apply performance shortcuts for synchronous processing, disable unneeded steps")
diff --git a/GPU/Workflow/src/GPUWorkflowTPC.cxx b/GPU/Workflow/src/GPUWorkflowTPC.cxx
index b64c25b63cc54..f895587b8b020 100644
--- a/GPU/Workflow/src/GPUWorkflowTPC.cxx
+++ b/GPU/Workflow/src/GPUWorkflowTPC.cxx
@@ -306,7 +306,7 @@ bool GPURecoWorkflowSpec::fetchCalibsCCDBTPC<GPUCalibObjectsConst>(ProcessingCon
     }
 
     if (mSpecConfig.outputTracks || mSpecConfig.caClusterer) {
-      mTPCCutAtTimeBin = pc.inputs().get<o2::tpc::AltroSyncSignal*>("tpcaltrosync")->getTB2Cut(pc.services().get<o2::framework::TimingInfo>().tfCounter);
+      mTPCCutAtTimeBin = mConfParam->overrideTPCTimeBinCur > 0 ? mConfParam->overrideTPCTimeBinCur : pc.inputs().get<o2::tpc::AltroSyncSignal*>("tpcaltrosync")->getTB2Cut(pc.services().get<o2::framework::TimingInfo>().tfCounter);
     }
 
     // these calibrations are only defined for the tracking