diff --git a/CHANGELOG.md b/CHANGELOG.md
index ecc585d6e..96b2df80f 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -18,6 +18,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/).
 - HAL MGM3100 Handler: Use axis specific gain/scaling factors. Previously,
   only the X scaling factor was used.
   PR: https://egit.irs.uni-stuttgart.de/fsfw/fsfw/pulls/724
+- Bugfix for RM3100 MGM sensors. Z value was previously calculated
+  with bytes of the X value.
 - DHB `setNormalDatapoolEntriesInvalid`: The default implementation did not set the validity
   to false correctly because the `read` and `write` calls were missing.
 - PUS TMTC creator module: Sequence flags were set to continuation segment (0b00) instead
diff --git a/src/fsfw_hal/devicehandlers/MgmRM3100Handler.cpp b/src/fsfw_hal/devicehandlers/MgmRM3100Handler.cpp
index ce215a644..307d0a558 100644
--- a/src/fsfw_hal/devicehandlers/MgmRM3100Handler.cpp
+++ b/src/fsfw_hal/devicehandlers/MgmRM3100Handler.cpp
@@ -325,7 +325,7 @@ ReturnValue_t MgmRM3100Handler::handleDataReadout(const uint8_t *packet) {
   // trickery here to calculate the raw values first
   int32_t fieldStrengthRawX = ((packet[1] << 24) | (packet[2] << 16) | (packet[3] << 8)) >> 8;
   int32_t fieldStrengthRawY = ((packet[4] << 24) | (packet[5] << 16) | (packet[6] << 8)) >> 8;
-  int32_t fieldStrengthRawZ = ((packet[7] << 24) | (packet[8] << 16) | (packet[3] << 8)) >> 8;
+  int32_t fieldStrengthRawZ = ((packet[7] << 24) | (packet[8] << 16) | (packet[9] << 8)) >> 8;
 
   // Now scale to physical value in microtesla
   float fieldStrengthX = fieldStrengthRawX * scaleFactorX;