# Insertion Loss vs Transmission Loss

Insertion Loss and Transmission Loss are often confused by hams. A couple of recent examples demonstrate the confusion:

• from VK2XSO speaking on loss in a connector at 600W power level (VKLOGGER 2012): [w]ith the 0.1dB insertion loss you will probably heat up the connectors with something around 5 to 10W;
• from VK4OX speaking on the same problem: [i]nsersion [sic] loss is the main problem. As the frequency goes up, so do the losses. A 0.1 dB loss in a 1000W system is 23 Watts. Any connector dissipating 23 Watts will get HOT; and
• from K2UE at (eHam 2012) Insertion loss = Heat.

# Insertion Loss

Insertion Loss is defined (FS-1037C 1996) as:

The loss resulting from the insertion of a device in a transmission line, expressed as the reciprocal of the ratio of the signal power delivered to that part of the line following the device to the signal power delivered to that same part before insertion.

The definition says nothing of the conversion of signal energy to heat.

# Transmission Loss

Transmission Loss is defined (FS-1037C 1996) as:

The decrease in power that occurs during transmission from one point to another

# Some examples

## Example 1

Let us take a theoretical example using a quarter wavelength of lossless line with Zo=75Ω inserted between source and load in a matched 50Ω system.

Since the line is lossless, it cannot convert RF energy to heat, so Transmission Loss for the line section MUST be 0dB.

The input impedance of the 75Ω lossless line section with 50Ω termination is 112.5+j0Ω, and the Mismatch Loss at that point is 0.7dB, so Insertion Loss is 0.7dB.

Insertion of the line section will reduce the power in the load by 0.7dB, but no RF energy is converted to heat in the line section..

## Example 2

Let us take a theoretical example using a half wavelength of lossless line with Zo=75Ω inserted between source and load in in a matched 50Ω system.

Since the line is lossless, it cannot convert RF energy to heat, so Transmission Loss for the line section MUST be 0dB.

The input impedance of the 75Ω lossless line section with 50Ω termination is 50+j0Ω, and the Mismatch Loss at that point is 0dB, so Insertion Loss is 0dB.

Insertion of the line section will not reduce the power in the load, and no RF energy is converted to heat in the line section.

## Example 3

Let us take a practical example using a quarter wavelength of RG-11 with Zo nominally 75Ω inserted between source and load in in a matched nominal 50Ω system at 1MHz.

Being a real line, it has loss and there MUST be transmission loss. Using TLLC, the Transmission Loss is 0.317dB, ie the power out of the line section is 97% of the power into the line section.

Again using TLLC, The input impedance of the 75Ω line section with 50Ω termination is 109.46-j4.53Ω, and using ATLLC, the Mismatch Loss at that point is 0.654dB, so Insertion Loss is the sum of Transmission Loss and Mismatch Loss at line input giving  0.967dB.

Insertion of the line section will reduce the power in the load by 0.971dB, and RF energy is converted to heat in the line section due to the 0.317dB of Transmission Loss.

## Example 4

Let us take a practical example using a half wavelength of RG-11 with Zo nominally 75Ω inserted between source and load in in a matched nominal 50Ω system at 1MHz.

Being a real line, it has loss and there MUST be transmission loss. Using TLLC, the Transmission Loss is 0.595dB, ie the power out of the line section is 86% of the power into the line section.

Again using TLLC, The input impedance of the 75Ω line section with 50Ω termination is 52.73-j0.15Ω, and using ATLLC, the Mismatch Loss at that point is 0.003dB, so Insertion Loss is the sum of Transmission Loss and Mismatch Loss at line input giving  0.598dB.

Insertion of the line section will reduce the power in the load by 0.598dB, and RF energy is converted to heat in the line section due to the 0.595dB of Transmission Loss.

# A cautionary note

All of this discussion has assumed that a 'matched' system is perfectly matched. The actual reduction in load power in a less than perfectly matched system may be more or less than the nominal Insertion Loss of the component. To that extent, the nominal Insertion Loss of a component is more a quality indicator than a reliable predictor of reduced power in a less than perfectly matched system.

Cascading components does not simply add their Insertion Loss. In Example 3, the insertion loss of a quarter wave of line was 0.967dB, yet when we cascaded two such sections in Example 4, the Insertion Loss is actually less at just 0.598dB.