diff --git a/src/grib_accessor_class_from_scale_factor_scaled_value.c b/src/grib_accessor_class_from_scale_factor_scaled_value.c
index bf339246d..70d7d9763 100644
--- a/src/grib_accessor_class_from_scale_factor_scaled_value.c
+++ b/src/grib_accessor_class_from_scale_factor_scaled_value.c
@@ -180,6 +180,79 @@ static double eval_value_factor(int64_t value, int64_t factor)
     return (double)value * pow(10.0, -factor);
 }
 
+/* Return 0 if no error and set the two outputs 'value' and 'factor'
+ * value cannot exceed maximum_value and factor cannot exceed maximum_factor
+ */
+static int get_scaled_value_and_scale_factor_algorithm1(
+    double input, int64_t maximum_value, int64_t maximum_factor,
+    int64_t* ret_value, int64_t* ret_factor)
+{
+    int64_t factor = 0;
+    int64_t value = 0;
+
+    factor = floor(log10(maximum_value)) - floor(log10(input < 0 ? -input : input));
+    if (factor > maximum_factor) {
+        return GRIB_INTERNAL_ERROR;
+    }
+    value = (int64_t)round(input * pow(10, factor));
+    while ((value % 10 == 0) && (factor > 0)) {
+        value /= 10;
+        factor--;
+    }
+    if (value >= maximum_value)
+        return GRIB_INTERNAL_ERROR;
+    if (factor >= maximum_factor)
+        return GRIB_INTERNAL_ERROR;
+
+    *ret_factor = factor;
+    *ret_value = value;
+    return GRIB_SUCCESS;
+}
+
+/* Return 0 if no error and set the two outputs 'value' and 'factor'
+ * value cannot exceed maximum_value and factor cannot exceed maximum_factor
+ */
+static int get_scaled_value_and_scale_factor_algorithm2(
+    double input, int64_t maximum_value, int64_t maximum_factor,
+    int64_t* ret_value, int64_t* ret_factor)
+{
+    int64_t factor = 0, prev_factor = 0;
+    int64_t value = 0, prev_value = 0;
+    double exact        = input;
+    const float epsilon = float_epsilon();
+    int is_negative = 0;
+
+    /* Loop until we find a close enough approximation. Keep the last good values */
+    if (exact < 0) {
+        is_negative = 1;
+        exact *= -1;
+    }
+    factor = prev_factor = 0;
+    value = prev_value = round(exact);
+    while (!is_approximately_equal(exact, eval_value_factor(value, factor), epsilon) &&
+           value < maximum_value &&
+           factor < maximum_factor) {
+        value = round(exact * pow(10., ++factor));
+        if (value > maximum_value || factor > maximum_factor) {
+            /* One or more maxima exceeded. So stop and use the previous values */
+            value  = prev_value;
+            factor = prev_factor;
+            break;
+        }
+        prev_factor = factor;
+        prev_value  = value;
+    }
+
+    if (is_negative) {
+        value *= -1;
+    }
+    
+    *ret_factor = factor;
+    *ret_value = value;
+    
+    return GRIB_SUCCESS;
+}
+
 static int pack_double(grib_accessor* a, const double* val, size_t* len)
 {
     /* See ECC-979 */
@@ -187,12 +260,10 @@ static int pack_double(grib_accessor* a, const double* val, size_t* len)
     grib_accessor_from_scale_factor_scaled_value* self = (grib_accessor_from_scale_factor_scaled_value*)a;
     grib_handle* hand                                  = grib_handle_of_accessor(a);
     int ret                                            = 0;
-    int64_t factor = 0, prev_factor = 0;
-    int64_t value = 0, prev_value = 0;
+    int64_t factor = 0;
+    int64_t value = 0;
     double exact        = *val; /*the input*/
-    const float epsilon = float_epsilon();
-    int is_negative = 0;
-    unsigned long maxval_value, maxval_factor; /*maximum allowable values*/
+    int64_t maxval_value, maxval_factor; /*maximum allowable values*/
     grib_accessor *accessor_factor, *accessor_value;
 
     if (exact == 0) {
@@ -220,29 +291,13 @@ static int pack_double(grib_accessor* a, const double* val, size_t* len)
     maxval_value  = (1UL << (accessor_value->length * 8)) - 2;  /* exclude missing */
     maxval_factor = (1UL << (accessor_factor->length * 8)) - 2; /* exclude missing */
 
-    /* Loop until we find a close enough approximation. Keep the last good values */
-    if (exact < 0) {
-        is_negative = 1;
-        exact *= -1;
-    }
-    factor = prev_factor = 0;
-    value = prev_value = round(exact);
-    while (!is_approximately_equal(exact, eval_value_factor(value, factor), epsilon) &&
-           value < maxval_value &&
-           factor < maxval_factor) {
-        value = round(exact * pow(10., ++factor));
-        if (value > maxval_value || factor > maxval_factor) {
-            /* One or more maxima exceeded. So stop and use the previous values */
-            value  = prev_value;
-            factor = prev_factor;
-            break;
+    ret = get_scaled_value_and_scale_factor_algorithm1(exact, maxval_value, maxval_factor, &value, &factor);
+    if (ret) {
+        ret = get_scaled_value_and_scale_factor_algorithm2(exact, maxval_value, maxval_factor, &value, &factor);
+        if (ret) {
+            grib_context_log(a->context, GRIB_LOG_ERROR, "Failed to compute %s and %s from %g", self->scaleFactor, self->scaledValue, exact);
+            return ret;
         }
-        prev_factor = factor;
-        prev_value  = value;
-    }
-
-    if (is_negative) {
-        value *= -1;
     }
 
     if ((ret = grib_set_long_internal(hand, self->scaleFactor, factor)) != GRIB_SUCCESS)
diff --git a/tests/grib_ecc-1150.sh b/tests/grib_ecc-1150.sh
index 83fc00d59..d03dcfede 100755
--- a/tests/grib_ecc-1150.sh
+++ b/tests/grib_ecc-1150.sh
@@ -88,25 +88,25 @@ ${tools_dir}/grib_set -s lowerLimit=-6.6,upperLimit=-1.02 $tempGrib $temp2
 grib_check_key_equals $temp2 scaleFactorOfLowerLimit,scaledValueOfLowerLimit,lowerLimit "1 -66 -6.6"
 grib_check_key_equals $temp2 scaleFactorOfUpperLimit,scaledValueOfUpperLimit,upperLimit "2 -102 -1.02"
 
-#                        input  factor value
-set_lowerLimit_and_check 550        0  550
-set_lowerLimit_and_check -99        0  -99
-set_lowerLimit_and_check 6.77       2  677
-set_lowerLimit_and_check 0.001      3  1
-set_lowerLimit_and_check -6.6       1  -66
-set_lowerLimit_and_check -1.02      2  -102
-set_lowerLimit_and_check 3e-05      5  3
-set_lowerLimit_and_check -3.9e-05   6  -39
-
-#set_lowerLimit_and_check 3.14e-06    6  314
-#set_lowerLimit_and_check 3.14e-07    7  314
-#set_lowerLimit_and_check 3e-08       8  3
-#set_lowerLimit_and_check 1.0e-10     10 1
-#set_lowerLimit_and_check 0.03e-06    8  3
-#set_lowerLimit_and_check 3.14e-09    11 314
-#set_lowerLimit_and_check -3.1456e-09 13 -31456
-#set_lowerLimit_and_check 0.0000123   7  123
-
+#                        input  factor  value
+set_lowerLimit_and_check 0           0  0
+set_lowerLimit_and_check 1           0  1
+set_lowerLimit_and_check 550         0  550
+set_lowerLimit_and_check -99         0  -99
+set_lowerLimit_and_check 6.77        2  677
+set_lowerLimit_and_check 0.001       3  1
+set_lowerLimit_and_check -6.6        1  -66
+set_lowerLimit_and_check -1.02       2  -102
+set_lowerLimit_and_check 3e-05       5  3
+set_lowerLimit_and_check -3.9e-05    6  -39
+set_lowerLimit_and_check 3.14e-06    8  314
+set_lowerLimit_and_check 3.14e-07    9  314
+set_lowerLimit_and_check 3e-08       8  3
+set_lowerLimit_and_check -3e-08      8  -3
+set_lowerLimit_and_check 1e-10       10 1
+set_lowerLimit_and_check 3.14e-09    11 314
+set_lowerLimit_and_check 1.23e-05    7  123
+set_lowerLimit_and_check -3.1456e-09 13 -31456
 
 # Clean up
 rm -f $tempGrib $tempFilt $temp2