diff --git a/qlasskit/types/qfixed.py b/qlasskit/types/qfixed.py
index 8666a9e6..6968ce61 100644
--- a/qlasskit/types/qfixed.py
+++ b/qlasskit/types/qfixed.py
@@ -49,7 +49,11 @@ class QfixedImp(float, Qtype):
 
     def __init__(self, value: float):
         super().__init__()
+
         self.value = value
+        # v = str(value).split('.')
+        # self.value = int(v[0]) % self.BIT_SIZE_INTEGER + (float(f'0.{v[1]}')
+        # if len(v) == 2 else 0)
 
     @classmethod
     def from_bool(cls, v: List[bool]):
@@ -200,6 +204,40 @@ def add(cls, tleft: TExp, tright: TExp) -> TExp:
 
         return (tleft[0], QfixedImp._from_qint_repr((tleft[0], res[1])))
 
+    @classmethod
+    def sub(cls, tleft: TExp, tright: TExp) -> TExp:
+        """Subtract two Qfixed"""
+        an = cls.bitwise_not(cls.fill(tleft))  # type: ignore
+        su = cls.add(an, cls.fill(tright))  # type: ignore
+        return cls.bitwise_not(su)  # type: ignore
+
+    @classmethod
+    def mul(cls, tleft: TExp, tright: TExp) -> TExp:  # noqa: C901
+        a = len(list(filter(lambda b: b is bool, tleft[1])))
+        b = len(list(filter(lambda b: b is bool, tright[1])))
+
+        if a == 0 and issubclass(tleft[0], Qint):  # type: ignore
+            tconst = tleft
+            top = tright
+        elif b == 0 and issubclass(tright[0], Qint):  # type: ignore
+            top = tleft
+            tconst = tright
+        else:
+            raise Exception(
+                "Qfixed mul works only between a Qfixed and an integer constant"
+            )
+
+        v_const = int(bool_list_to_bin(tconst[1])[::-1], 2)
+
+        if v_const == 0:
+            return cls.const(0.0)
+
+        v = top
+        for i in range(v_const - 1):
+            v = cls.add(v, top)
+
+        return v
+
 
 class Qfixed1_2(QfixedImp):
     BIT_SIZE = 3
diff --git a/test/qlassf/test_fixed.py b/test/qlassf/test_fixed.py
index 01353cf5..f34a753f 100644
--- a/test/qlassf/test_fixed.py
+++ b/test/qlassf/test_fixed.py
@@ -16,7 +16,7 @@
 
 from parameterized import parameterized, parameterized_class
 
-from qlasskit import qlassf
+from qlasskit import Qint2, qlassf
 from qlasskit.types.qfixed import Qfixed1_3, Qfixed2_3, Qfixed2_4
 from qlasskit.types.qtype import bin_to_bool_list
 
@@ -63,8 +63,18 @@ def test_fixed_gt(self, qft, a, b, r):
             [Qfixed2_3, 3.5, 0.5],
         ]
     )
-    def test_fixed_add(self, qft, a, b):
+    def test_fixed_add_sub(self, qft, a, b):
         self.assertEqual(qft.add(qft.const(a), qft.const(b))[1], qft.const(a + b)[1])
+        self.assertEqual(qft.sub(qft.const(a), qft.const(b))[1], qft.const(a - b)[1])
+
+    @parameterized.expand(
+        [
+            [Qfixed2_3, 0.5, Qint2, 2],
+            [Qfixed2_3, 0.5, Qint2, 0],
+        ]
+    )
+    def test_fixed_mul(self, qft, a, qit, b):
+        self.assertEqual(qft.mul(qft.const(a), qit.const(b))[1], qft.const(a * b)[1])
 
 
 @parameterized_class(("compiler"), ENABLED_COMPILERS)
@@ -119,16 +129,38 @@ def test_gte(self):
         qf = qlassf(f, to_compile=COMPILATION_ENABLED, compiler=self.compiler)
         compute_and_compare_results(self, qf)
 
-    def test_sum_const(self):
+    def test_add_const(self):
         f = "def test(a: Qfixed[2,4]) -> Qfixed[2, 4]:\n\treturn Qfixed2_4(0.5) + a"
         qf = qlassf(f, to_compile=COMPILATION_ENABLED, compiler=self.compiler)
         compute_and_compare_results(self, qf)
 
-    def test_sum(self):
+    def test_add(self):
         f = "def test(a: Qfixed[1,4], b: Qfixed[1,4]) -> Qfixed[1,4]:\n\treturn a + b"
         qf = qlassf(f, to_compile=COMPILATION_ENABLED, compiler=self.compiler)
         compute_and_compare_results(self, qf)
 
+    def test_mul(self):
+        f = "def test(a: Qfixed[1,4]) -> Qfixed[1,4]:\n\treturn a * 3"
+        qf = qlassf(f, to_compile=COMPILATION_ENABLED, compiler=self.compiler)
+        compute_and_compare_results(self, qf)
+
+    def test_sub(self):
+        f = (
+            "def test(a: Qfixed[1,4], b: Qfixed[1,4]) -> Qfixed[1,4]:\n"
+            "\treturn a - b if a > b else b - a"
+        )
+        qf = qlassf(f, to_compile=COMPILATION_ENABLED, compiler=self.compiler)
+        compute_and_compare_results(self, qf)
+
+    def test_sub_const(self):
+        # TODO: allow negative overflow
+        f = (
+            "def test(a: Qfixed[2,4]) -> Qfixed[2, 4]:\n"
+            "\treturn (a - Qfixed2_4(0.5)) if a > Qfixed2_4(0.5) else 0"
+        )
+        qf = qlassf(f, to_compile=COMPILATION_ENABLED, compiler=self.compiler)
+        compute_and_compare_results(self, qf)
+
     # def test_to_int(self):
     #     f = "def test(a: Qfixed[2,4]) -> Qint2:\n\treturn int(a)"
     #     qf = qlassf(f, to_compile=COMPILATION_ENABLED, compiler=self.compiler)
diff --git a/test/utils.py b/test/utils.py
index 6eec672d..1284fa33 100644
--- a/test/utils.py
+++ b/test/utils.py
@@ -174,10 +174,10 @@ def res_to_str(res):
                 qi = qf.returns.ttype.from_bool(qc[1])
             except:
                 qi = qc[0].from_bool(qc[1])
-            return qi.to_bin()   
-        elif type(res) is float:       
+            return qi.to_bin()
+        elif type(res) is float:
             qi = qf.returns.ttype(res)
-            return qi.to_bin()   
+            return qi.to_bin()
         else:
             return res.to_bin()
 
@@ -191,7 +191,7 @@ def res_to_str(res):
 
     res_original = qf.original_f(*args)
     res_original_str = res_to_str(res_original)
-    
+
     # truth_str = "".join(
     #     map(lambda x: "1" if x else "0", truth_line[-qf.output_size :])
     # )