diff --git a/firmware/console/binary/tunerstudio_configuration.h b/firmware/console/binary/tunerstudio_configuration.h
index 9ecb914ed5..1ad46c2793 100644
--- a/firmware/console/binary/tunerstudio_configuration.h
+++ b/firmware/console/binary/tunerstudio_configuration.h
@@ -8,54 +8,12 @@
  * @author Andrey Belomutskiy, (c) 2012-2020
  */
 
-#ifndef TUNERSTUDIO_CONFIGURATION_H_
-#define TUNERSTUDIO_CONFIGURATION_H_
+#pragma once
 
 #include "rusefi_types.h"
-#include "rusefi_generated.h"
-
+#include "scaled_channel.h"
 #include "tunerstudio_debug_struct.h"
 
-// This class lets us transparently store something at a ratio inside an integer type
-// Just use it like a float - you can read and write to it, like this:
-// scaled_channel<uint8_t, 10> myVar;
-// myVar = 2.4f;	// converts to an int, stores 24
-// float x = myVar; // converts back to float, returns 2.4f
-template <typename T, int mult = 1>
-class scaled_channel {
-public:
-	scaled_channel() : m_value(static_cast<T>(0)) { }
-	scaled_channel(float val)
-		: m_value(val * mult)
-	{
-	}
-
-	// Allow reading back out as a float (note: this may be lossy!)
-	operator float() const {
-		return m_value / (float)mult;
-	}
-
-private:
-	T m_value;
-};
-
-// We need to guarantee that scaled values containing some type are the same size
-// as that underlying type.  We rely on the class only having a single member for
-// this trick to work.
-static_assert(sizeof(scaled_channel<uint8_t>) == 1);
-static_assert(sizeof(scaled_channel<uint16_t>) == 2);
-static_assert(sizeof(scaled_channel<uint32_t>) == 4);
-static_assert(sizeof(scaled_channel<float>) == 4);
-
-// Common scaling options - use these if you can!
-using scaled_temperature = scaled_channel<int16_t, PACK_MULT_TEMPERATURE>;	// +-327 deg C at 0.01 deg resolution
-using scaled_ms = scaled_channel<int16_t, PACK_MULT_MS>;				// +- 100ms at 0.003ms precision
-using scaled_percent = scaled_channel<int16_t, PACK_MULT_PERCENT>;		// +-327% at 0.01% resolution
-using scaled_pressure = scaled_channel<uint16_t, PACK_MULT_PRESSURE>;		// 0-2000kPa (~300psi) at 0.03kPa resolution
-using scaled_angle = scaled_channel<int16_t, PACK_MULT_ANGLE>;			// +-655 degrees at 0.02 degree resolution
-using scaled_voltage = scaled_channel<uint16_t, PACK_MULT_VOLTAGE>;		// 0-65v at 1mV resolution
-using scaled_afr = scaled_channel<uint16_t, PACK_MULT_AFR>;			// 0-65afr at 0.001 resolution
-
 typedef struct {
 	uint16_t values[EGT_CHANNEL_COUNT];
 } egt_values_s;
@@ -230,5 +188,3 @@ typedef struct {
 } TunerStudioOutputChannels;
 
 extern TunerStudioOutputChannels tsOutputChannels;
-
-#endif /* TUNERSTUDIO_CONFIGURATION_H_ */
diff --git a/firmware/util/scaled_channel.h b/firmware/util/scaled_channel.h
new file mode 100644
index 0000000000..d01f354d52
--- /dev/null
+++ b/firmware/util/scaled_channel.h
@@ -0,0 +1,56 @@
+/*
+ * @file	scaled_channel.h
+ * @brief	Scaled channel storage for binary formats
+ *
+ * Storage of values (floating point, usually) scaled in integer storage, for transmission over the wire.
+ * This includes Tunerstudio serial/USB, and CAN.
+ *
+ * @date Feb 24, 2020
+ * @author Matthew Kennedy, (c) 2020
+ */
+
+#pragma once
+
+#include <cstdint>
+
+#include "rusefi_generated.h"
+
+// This class lets us transparently store something at a ratio inside an integer type
+// Just use it like a float - you can read and write to it, like this:
+// scaled_channel<uint8_t, 10> myVar;
+// myVar = 2.4f;	// converts to an int, stores 24
+// float x = myVar; // converts back to float, returns 2.4f
+template <typename T, int mult = 1>
+class scaled_channel {
+public:
+	scaled_channel() : m_value(static_cast<T>(0)) { }
+	scaled_channel(float val)
+		: m_value(val * mult)
+	{
+	}
+
+	// Allow reading back out as a float (note: this may be lossy!)
+	operator float() const {
+		return m_value / (float)mult;
+	}
+
+private:
+	T m_value;
+};
+
+// We need to guarantee that scaled values containing some type are the same size
+// as that underlying type.  We rely on the class only having a single member for
+// this trick to work.
+static_assert(sizeof(scaled_channel<uint8_t>) == 1);
+static_assert(sizeof(scaled_channel<uint16_t>) == 2);
+static_assert(sizeof(scaled_channel<uint32_t>) == 4);
+static_assert(sizeof(scaled_channel<float>) == 4);
+
+// Common scaling options - use these if you can!
+using scaled_temperature = scaled_channel<int16_t, PACK_MULT_TEMPERATURE>;	// +-327 deg C at 0.01 deg resolution
+using scaled_ms = scaled_channel<int16_t, PACK_MULT_MS>;				// +- 100ms at 0.003ms precision
+using scaled_percent = scaled_channel<int16_t, PACK_MULT_PERCENT>;		// +-327% at 0.01% resolution
+using scaled_pressure = scaled_channel<uint16_t, PACK_MULT_PRESSURE>;		// 0-2000kPa (~300psi) at 0.03kPa resolution
+using scaled_angle = scaled_channel<int16_t, PACK_MULT_ANGLE>;			// +-655 degrees at 0.02 degree resolution
+using scaled_voltage = scaled_channel<uint16_t, PACK_MULT_VOLTAGE>;		// 0-65v at 1mV resolution
+using scaled_afr = scaled_channel<uint16_t, PACK_MULT_AFR>;			// 0-65afr at 0.001 resolution