diff --git a/lib/ramble/ramble/test/workflow_manager_functionality/slurm_workflow_manager.py b/lib/ramble/ramble/test/workflow_manager_functionality/slurm_workflow_manager.py
index a2aae7605..2f907ab55 100644
--- a/lib/ramble/ramble/test/workflow_manager_functionality/slurm_workflow_manager.py
+++ b/lib/ramble/ramble/test/workflow_manager_functionality/slurm_workflow_manager.py
@@ -33,6 +33,7 @@ def test_slurm_workflow():
     batch_submit: echo {wm_name}
     mpi_command: mpirun -n {n_ranks} -hostfile hostfile
     processes_per_node: 1
+    n_nodes: 1
     wm_name: ['None', 'slurm']
   applications:
     hostname:
@@ -41,8 +42,12 @@ def test_slurm_workflow():
           experiments:
             test_{wm_name}:
               variables:
-                n_nodes: 1
-                extra_sbatch_headers: "#SBATCH --gpus-per-task={n_threads}"
+                extra_sbatch_headers: |
+                  #SBATCH --gpus-per-task={n_threads}
+                  #SBATCH --time={time_limit_not_exist}
+            test_{wm_name}_2:
+              variables:
+                slurm_partition: h3
 """
     with ramble.workspace.create(workspace_name) as ws:
         ws.write()
@@ -83,9 +88,17 @@ def test_slurm_workflow():
             content = f.read()
             assert "scontrol show hostnames" in content
             assert "#SBATCH --gpus-per-task=1" in content
+            assert "#SBATCH -p" not in content
+            assert "#SBATCH --time" not in content
         with open(os.path.join(path, "batch_query")) as f:
             content = f.read()
             assert "squeue" in content
         with open(os.path.join(path, "batch_cancel")) as f:
             content = f.read()
             assert "scancel" in content
+
+        # Assert on the experiment with non-empty partition variable given
+        path = os.path.join(ws.experiment_dir, "hostname", "local", "test_slurm_2")
+        with open(os.path.join(path, "slurm_execute_experiment")) as f:
+            content = f.read()
+            assert "#SBATCH -p h3" in content
diff --git a/var/ramble/repos/builtin/workflow_managers/slurm/workflow_manager.py b/var/ramble/repos/builtin/workflow_managers/slurm/workflow_manager.py
index 105a1c855..9477befc0 100644
--- a/var/ramble/repos/builtin/workflow_managers/slurm/workflow_manager.py
+++ b/var/ramble/repos/builtin/workflow_managers/slurm/workflow_manager.py
@@ -9,7 +9,6 @@
 import os
 
 from ramble.wmkit import *
-from ramble.expander import ExpanderError
 from ramble.application import experiment_status
 
 from spack.util.executable import ProcessError
@@ -78,6 +77,12 @@ def __init__(self, file_path):
         description="mpirun prefix, mostly served as an overridable default",
     )
 
+    workflow_manager_variable(
+        name="slurm_partition",
+        default="",
+        description="partition to submit job to, if unspecified, it uses the default partition",
+    )
+
     register_template(
         name="batch_submit",
         src_name="batch_submit.tpl",
@@ -109,21 +114,18 @@ def _execute_vars(self):
         # Adding pre-defined and custom headers
         pragmas = [
             ("#SBATCH -N {n_nodes}"),
-            ("#SBATCH -p {partition}"),
             ("#SBATCH --ntasks-per-node {processes_per_node}"),
             ("#SBATCH -J {job_name}"),
             ("#SBATCH -o {experiment_run_dir}/slurm-%j.out"),
             ("#SBATCH -e {experiment_run_dir}/slurm-%j.err"),
             ("#SBATCH --gpus-per-node {gpus_per_node}"),
         ]
-        try:
-            extra_sbatch_headers_raw = expander.expand_var_name(
-                "extra_sbatch_headers", allow_passthrough=False
-            )
-            extra_sbatch_headers = extra_sbatch_headers_raw.strip().split("\n")
-            pragmas = pragmas + extra_sbatch_headers
-        except ExpanderError:
-            pass
+        if expander.expand_var_name("slurm_partition"):
+            pragmas.append("#SBATCH -p {slurm_partition}")
+        extra_headers = (
+            self.app_inst.variables["extra_sbatch_headers"].strip().split("\n")
+        )
+        pragmas = pragmas + extra_headers
         header_str = "\n".join(self.conditional_expand(pragmas))
         return {"sbatch_headers_str": header_str}