IMD
InterModulation Distortion
 

IMD is the result of mixing of modulation components, frequencies within your voice, in a non-linear component or device within your transmitting system. Most people think of power amplifiers when asked about IMD but all amplifiers can produce IMD. This includes those in the early stages of your transceiver as well as those in your receiver. Odd order products are the ones that cause us problems. Second order products generally fall way outside the passband of interest.

Most modern transceivers are designed to provide pretty good performance with relatively low IMD. Some are better than others. Power amplifiers can produce IMD if, for any reason, they are non-linear at any time during the operating cycle. IMD can also be produced by many of the modifications being performed on transceivers these days. Increasing audio pass band, increasing the bandwidth of the filters, changing the transmit carrier insertion point, boosting bass, boosting treble, and turning up the output power can all result in increased IMD.

Below are some references to information from Tom Rauch, W8JI, and Rich Measures, AG6K, that will shed a bit more light on this problem and, not only help you understand how IMD is produced, but also introduce you to a method that can be used to measure IMD yourself without the need for expensive test equipment. It will also help to dispel some of the myths that are circulating around the bands.
 

Tom Rauch, W8JI, and Rich Measures, AG6K, both agree that trying to use a two tone test for judging IMD on an SSB transmitter is not the right way to do it.

Tom suggests either using a 3 tone test with the 3rd tone being low frequency so it modulates the other tones at a syllabic rate, or using voice modulation (an adjacent channel power test).

Rich says that the best method to measure IMD is something any of us can do as long as we have a step attenuator. He says that you use your receiver to tune off either side of the signal you want to test. Modulate the amp under test with "aaaaaaaaaaah" or something similar... a constant vocal sound, not a whistle. Adjust the RF gain for a mid scale meter reading, say S9. Then tune the receiver to the operating frequency and insert as much attenuation as necessary to get the same meter reading with the same kind of modulation. The amount of attenuation you have to crank in is equivalent to the IMD.

W8JI - Transmitter splatter - This is not as simple as it sounds and can't be checked as simply as some will have you think. You can NOT use a band scope (built in or outboard) for these checks and simply switching sidebands is totally ineffective as a check for IMD. On this page Tom says: "There are two tests that are better. One test is an adjacent channel power tests, with normal voice modulation of the transmitter, another test I developed uses a three-tone signal."
http://w8ji.com/transmitter_splatter.htm

W8JI - Checking signal bandwidth using a receiver can be done but MUST be done carefully, with a good receiver, WITHOUT noise blanker circuits active, and typically not with a DSP radio. You can't do these checks on the air with less than perfect conditions or in the presance of noise or other signals. http://w8ji.com/checking_bandwidth_with_receiver.htm

W8JI - Receivers - some are not as good as others. This is especially true when using one to listen for IMD. http://w8ji.com/receivers.htm

AG6K - The adjacent channel power test that Tom refers to is also addressed by Rich Measures on the email reflector amps@contesting.com:

9/22/2005
"The main tool needed to measure total IMD is a second SSB receiver -- preferably one with a double-filter -- tuned c. 4KHz above or below the signal being measured."

"With 0db attenuation, take the S meter reading at the adjacent frequency, tune to the fundamental frequency, add attenuation as needed to give the same S-meter reading as before. The # of db of attenuation is how many db down the total IMD is."

9/23/2005
"The IMD level produced by the transmission of human speech can not be found by a 2-tone IMD test. I use a steady "ahhhhhhhhh"."

"I do not set the meter. With the RF gain at 10, the splatter moves the meter to some level, perhaps c, half-scale. I note the average reading, move to the fundamental frequency, and add whatever attn is
needed to achieve the same reading as before. The # of db on the attn is the total IMD level."

"My HP step attenuator set is somewhat better than that. From S-1 to S-5, a 1db change makes a noticeable difference on the S-meter."

AG6K - AND on his web site: http://somis.org/D-amplifiers4.html
 
Both of these guys have experience and knowledge that we all can benefit from. The time spent reading their sites is time well spent.

Additional references:
 

 

August 14, 2006