diff --git a/CHANGES.rst b/CHANGES.rst
index 736c205f2b..e559a43544 100644
--- a/CHANGES.rst
+++ b/CHANGES.rst
@@ -5,6 +5,7 @@ Changelog
 2.3.0 (unreleased)
 ------------------
 
+- #2089 Fix LDL/UDL cut-off and exponential float conversion
 - #2078 Replace dynamic code execution with dynamic import in reports
 - #2083 Lookup workflow action redirect URL from request first
 - #2082 Include sample ID in form ID for lab, field and qc analyses listings
diff --git a/src/bika/lims/content/abstractanalysis.py b/src/bika/lims/content/abstractanalysis.py
index 9b4f58d88e..c9c0cf7464 100644
--- a/src/bika/lims/content/abstractanalysis.py
+++ b/src/bika/lims/content/abstractanalysis.py
@@ -388,7 +388,8 @@ def isBelowLowerDetectionLimit(self):
             return True
 
         if api.is_floatable(result):
-            return api.to_float(result) < self.getLowerDetectionLimit()
+            ldl = self.getLowerDetectionLimit()
+            return api.to_float(result) < api.to_float(ldl, 0.0)
 
         return False
 
@@ -405,10 +406,13 @@ def isAboveUpperDetectionLimit(self):
             return True
 
         if api.is_floatable(result):
-            return api.to_float(result) > self.getUpperDetectionLimit()
+            udl = self.getUpperDetectionLimit()
+            return api.to_float(result) > api.to_float(udl, 0.0)
 
         return False
 
+    # TODO: REMOVE:  nowhere used
+    @deprecated("This Method will be removed in version 2.5")
     @security.public
     def getDetectionLimits(self):
         """Returns a two-value array with the limits of detection (LDL and
@@ -416,7 +420,9 @@ def getDetectionLimits(self):
         the analysis service doesn't allow manual input of detection limits,
         returns the value set by default in the Analysis Service
         """
-        return [self.getLowerDetectionLimit(), self.getUpperDetectionLimit()]
+        ldl = self.getLowerDetectionLimit()
+        udl = self.getUpperDetectionLimit()
+        return [api.to_float(ldl, 0.0), api.to_float(udl, 0.0)]
 
     @security.public
     def isLowerDetectionLimit(self):
@@ -575,8 +581,8 @@ def calculateResult(self, override=False, cascade=False):
                     adl = dependency.isAboveUpperDetectionLimit()
                     mapping[key] = result
                     mapping['%s.%s' % (key, 'RESULT')] = result
-                    mapping['%s.%s' % (key, 'LDL')] = ldl
-                    mapping['%s.%s' % (key, 'UDL')] = udl
+                    mapping['%s.%s' % (key, 'LDL')] = api.to_float(ldl, 0.0)
+                    mapping['%s.%s' % (key, 'UDL')] = api.to_float(udl, 0.0)
                     mapping['%s.%s' % (key, 'BELOWLDL')] = int(bdl)
                     mapping['%s.%s' % (key, 'ABOVEUDL')] = int(adl)
                 except (TypeError, ValueError):
@@ -915,6 +921,7 @@ def getFormattedResult(self, specs=None, decimalmark='.', sciformat=1,
 
         # Below Lower Detection Limit (LDL)?
         ldl = self.getLowerDetectionLimit()
+        ldl = api.to_float(ldl, 0.0)
         if result < ldl:
             # LDL must not be formatted according to precision, etc.
             # Drop trailing zeros from decimal
@@ -924,6 +931,7 @@ def getFormattedResult(self, specs=None, decimalmark='.', sciformat=1,
 
         # Above Upper Detection Limit (UDL)?
         udl = self.getUpperDetectionLimit()
+        udl = api.to_float(udl, 0.0)
         if result > udl:
             # UDL must not be formatted according to precision, etc.
             # Drop trailing zeros from decimal
diff --git a/src/bika/lims/content/abstractbaseanalysis.py b/src/bika/lims/content/abstractbaseanalysis.py
index 822ac00d86..d638a25ae0 100644
--- a/src/bika/lims/content/abstractbaseanalysis.py
+++ b/src/bika/lims/content/abstractbaseanalysis.py
@@ -19,6 +19,7 @@
 # Some rights reserved, see README and LICENSE.
 
 from AccessControl import ClassSecurityInfo
+from bika.lims import api
 from bika.lims import bikaMessageFactory as _
 from bika.lims.browser.fields import DurationField
 from bika.lims.browser.fields import UIDReferenceField
@@ -136,14 +137,16 @@
     )
 )
 
+# TODO: Use a plain string field when converting to DX!
+#
 # If the value is below this limit, it means that the measurement lacks
 # accuracy and this will be shown in manage_results and also on the final
 # report.
 LowerDetectionLimit = FixedPointField(
-    'LowerDetectionLimit',
+    "LowerDetectionLimit",
     schemata="Analysis",
-    default='0.0',
-    precision=7,
+    default="0.0",
+    precision=1000,  # avoid precision cut-off done by the field
     widget=DecimalWidget(
         label=_("Lower Detection Limit (LDL)"),
         description=_(
@@ -154,14 +157,16 @@
     )
 )
 
+# TODO: Use a plain string field when converting to DX!
+#
 # If the value is above this limit, it means that the measurement lacks
 # accuracy and this will be shown in manage_results and also on the final
 # report.
 UpperDetectionLimit = FixedPointField(
-    'UpperDetectionLimit',
+    "UpperDetectionLimit",
     schemata="Analysis",
-    default='1000000000.0',
-    precision=7,
+    default="1000000000.0",
+    precision=1000,  # avoid precision cut-off done by the field
     widget=DecimalWidget(
         label=_("Upper Detection Limit (UDL)"),
         description=_(
@@ -847,23 +852,29 @@ def getAnalysisCategories(self):
 
     @security.public
     def getLowerDetectionLimit(self):
-        """Returns the Lower Detection Limit for this service as a floatable
+        """Get the lower detection limit without trailing zeros
         """
-        ldl = self.getField('LowerDetectionLimit').get(self)
-        try:
-            return float(ldl)
-        except ValueError:
-            return 0
+        field = self.getField("LowerDetectionLimit")
+        value = field.get(self)
+        # NOTE: This is a workaround to avoid the cut-off done by the field
+        #       if the value is lower than the precision
+        #       => we should use a string instead
+        # remove trailing zeros and possible trailing dot
+        value = value.rstrip("0").rstrip(".")
+        return value
 
     @security.public
     def getUpperDetectionLimit(self):
-        """Returns the Upper Detection Limit for this service as a floatable
+        """Get the upper detection limit without trailing zeros
         """
-        udl = self.getField('UpperDetectionLimit').get(self)
-        try:
-            return float(udl)
-        except ValueError:
-            return 0
+        field = self.getField("UpperDetectionLimit")
+        value = field.get(self)
+        # NOTE: This is a workaround to avoid the cut-off done by the field
+        #       if the value is lower than the precision
+        #       => we should use a string instead
+        # remove trailing zeros and possible trailing dot
+        value = value.rstrip("0").rstrip(".")
+        return value
 
     @security.public
     def isSelfVerificationEnabled(self):
diff --git a/src/senaite/core/tests/test_limitdetections.py b/src/senaite/core/tests/test_limitdetections.py
index a6d603a515..c3803977e2 100644
--- a/src/senaite/core/tests/test_limitdetections.py
+++ b/src/senaite/core/tests/test_limitdetections.py
@@ -382,8 +382,8 @@ def test_ar_manageresults_limitdetections(self):
         for a in ans:
             an = a.getObject()
             idx = asidxs[an.id]
-            self.assertEqual(an.getLowerDetectionLimit(), float(self.lds[idx]['min']))
-            self.assertEqual(an.getUpperDetectionLimit(), float(self.lds[idx]['max']))
+            self.assertEqual(an.getLowerDetectionLimit(), self.lds[idx]['min'])
+            self.assertEqual(an.getUpperDetectionLimit(), self.lds[idx]['max'])
             self.assertEqual(an.getAllowManualDetectionLimit(), self.lds[idx]['manual'])
 
             # Empty result