• First and foremost I believe we need an accurate definition of the  term -----Intermodulation Distortion---  if we are to understand and solve the problem.

·    Definition  of the term Intermodulation Distortion :  A form of interference involving the generation of interfering beats between two or more carriers according to the frequency relationship f = nf1 ± mf2,

·    where n and m are whole numbers (but not zero), with appropriate expansion for additional carriers

·     For additional definitions click on  http://www.google.com/search?q=define:Intermodulation+Distortion

·    I've also Included  an article I found in QST about Intermodulation which gives working definitions and methods for calculating the inputs and products and  examples of fixing the problem

·      I've also included an email (see below) Sent to Andy 12/03/2002 giving experiments that I had  carried out and some examples of successful things I've done  in the past  to clear up intermod in  repeaters I've had the responsibility of servicing

·       Last  but certainly not least  I've included A rough update of some of the work carried out in the last few weeks

Email from k1ugm to Andy on 12/03/2002 ;   Subject Intermod on 64

Hi Andy ,

 I just want to pass along some tests and calculations that I made on and for the 64 repeater concerning our noise problem and the 147.240 repeater.

The following equation is for the intermod product that I feel is most likely giving 64 the noise issue that 64 has been experiencing over the last several weeks.

From the model or theory; The most likely intermod relationship is

• (A * Freq1) + or - (B * Freq2) = INTERMOD FREQ

• So submitting  the number in to the equation.

• (2 * 146.640mhz)   -  (1 * 147.240mhz) =  146.040 mhz

• Like the equation says both frequencies have to be on and getting into the receiver to get the intermod affect.

The Experiment;

• Using a pair of   2-meter rigs at my qth turned one rig on at 147.240 mhz and varied it from  5 to 40 watts listening to 146.64

• It did not bring up 146.640 repeater.

• I turned on another rig on the input of  146.640 with my other rig transmitting on 147.24

• The intermod racket came on with 64.

• As I turned the 147.240 on and off the intermod noise on 64 went on and off.

• I have repeated this test several times over a 17-hour period. It always repeats the same affect.

The conditions

• There has to be an input from the 64 transmitter

• The 147.24 transmitter and a non-linear circuit element to mix these frequencies.

• The nonlinear element is almost always in the receiver or a preamp.

• But not necessarily, caused by a strong signal coming in at any frequency which is usually filter out or down by the cavities. 

• More likely I've found it to be a poor receiver design. It is really  very hard to have a truly linear receiver input, and is super hard for a pre amp.  

• This receiver’s problem mainly shows up in a one antenna for receive and transmit system.

• All though a nonlinear element could  be  a bad connection out side in the neighborhood of the antenna.

• I’ve never seen it happen in all  my years doing repeaters, and a calculation would show why.

The Possible Solutions

• Keep 146.64 and 147.24 RF out of the receiver. Make sure that the cavities are giving up to 90 db or better rejection at the these frequencies.

• Check to see if the receiver and preamp are properly shielded, again up to or better than 90 db

• Good feed thru caps. for power control, input, and output etc circuits are very important.

• Tight seals around the box lids etc. bronze finger springs are often found on high grade repeater receiver equipment.

• Check the coax from the cavities to the receiver and or preamp.  It’s far more important from an intermod issue than just keeping a repeater going with out intermod.

• Without the intermod issue a little feed back from the xmitter usually just reduces the sensitivity of the receiver but does not inject much noise and the users will never notice the difference. With intermod it can be a disaster.

• Very low swr   1.2 or better this keeps the RF off the outside of the coax and cavities.

• It useful to have  proper decoupling of the antenna from the coax. Again this reduces the RF that will get on the outside of the coax and back to the cavities and possibly the receiver box etc. 

• My educated guess is that it is a simple cavity adjustment or receiver issue.

• And then again I could be all wrong! Its happened to me before and certainly will happen again.

Respectfully yours,

Jim k1ugm

Brief Status      03/15/2004 k1ugm

• Saturday before 03/15/2004 , the  two hand held experiment discussed in the email above was duplicated at the repeater site by several of the tech crew demonstrating intermod issue
• We tried a separate antenna on the receiver and transmitter with out solving the problem
• Steve HUD has found the same problem on the new or backup repeater receiver which is similar to the one at the 64 site.

Intermodulation Reviewed

QST May 1983 17

With examples for fixing the problem

Please note the red lettered parts of the article

 A QST Article by David W. Potter, W1GZD

What is Intermodulation and how do you cure it.

 Intel-modulation (IM) is defined as the undesired mixing of two or more frequencies in a nonlinear device, which produces additional sum-and-difference frequencies.  Problems with IM seem prevalent in the vhf and uhf bands, because amateur repeaters tend to cluster around hills and mountain-tops in close proximity to commercial and government vhf and uhf radio  services. The intermodulation problem has been understood for years, but a review of the subject can be helpful.

Types of Intermodulation

  The simplest kind of intermodulation is the mixing of two frequencies.

The equation for second-order IM

 is

f_IM =f_l ± f₂

(Eq. 1)

where f_IM is the frequency of the IM product,  and f_x are the mixing frequencies.   Intermodulation products are the sum and difference of the two mixing frequencies. These are similar to the familiar products deliberately generated by the mixing process used in super heterodyne receivers. Note that if f₁ and f₂ are frequencies within an amateur band f_IM must be an out-of-band signal.

  Third-order IM products can produce both in-band and out-of-band signals when f₁, f₂; and f₃ are all in-band frequencies as given by the equation

f_lM = f₁ ± f₂ ± f₃       (Eq. 2)

   In-band signals are generated by the sum of any two frequencies minus the frequency of the third signal. The out-of- band signal is the sum of all three frequencies.  A special case of Eq. 2 is

>>>>>    f_IM = f₁ + f₁ - f₂ = 2f₁ - f₂  <<<<<<<         (Eq. 2A)

Here, the second harmonic of an in-band signal can beat with a fundamental frequency  to produce another in-band signal.

   Fifth-order IM products are given by

f_lM = f₁ ± f₂ ± f₃ ± f₄ ± f₅    (Eq. 3)

with special cases:

f_lM = 3f₁ ± 2f₂                 (Eq. 3A)

f_IM = 3f₁ ± f₂ ± f₃            (Eq. 3B)

f_IM = 2f₁ ± f₂ ± f₃ ± f₄       (Eq. 3C)

f_IM = 2f₁ ± 2f₂ ± f₃           (Eq. 3D)

  Some fifth-order products are in-band and others are out-of-band. Notice that second- and third-harmonic signals may be involved. Odd  orders of in-band mixing frequencies produce some in-band products, but when out-of-band mixing frequencies  are  involved,  even-order products may fall in band! 

  The above equations are valid for steady carriers, and it is easier to understand the concept of intermodulation by using them. Most of the signals we deal with are not steady carriers, however, but modulated ones.'   The bandwidth of an IM product may be wider than the' bandwidths of the individual signals. This is because the instantaneous frequency of the product is the algebraic sum of the instantaneous frequencies of the mixing signals. For fm, it  would  be  equivalent  to  adding  three voice signals together in a wide-band fm transmitter.  Assume that the  audio amplitude is limited on each signal to produce  a deviation no greater than 5 kHz. When the three signals are added together, they could produce an IM product having much greater deviation than any of the individual signals.  

   Intermodulation may  involve  any number of frequencies, but let's concentrate  on the more common third-order Types. I will show some examples to make These abstract concepts more meaningful

Field Examples of Intermodulation

    Consider - two repeaters that are physically located close to each other. One repeater transmits on 146.70 MHz, with a 146.10-MHz input, and the second has a 145.31-MHz output and a 144.71-MHz input.  Assume that the .31 repeater output causes the .70 repeater first receiver stage to be driven into nonlinear operation (overload). This means that mixing of all frequencies seen by the first stage will occur.   

  Case 1: The .70 repeater is off, but the 31 machine is  operating  and  a  local 145.50-MHz simplex signal is present. The 31 repeater input and output signals and the simplex signal mix to produce an IM product on 146.10MHz: 145.31 - 144.71  + 145.50 = 146.10 MHz. This signal can key up the .70 machine, which will then repeat both the  .31 machine and the simplex conversation. We will disregard the other IM products that are generated.

   Case 2: The simplex station is off the air. The .31 machine is repeating, and the input to the .70 repeater drops, but the output is still up. An IM signal is produced:  146.70 - 145.31  + 144.71  = 146.10 MHz. The 0.70 machine will now repeat the signal from the .31 repeater until someone overrides the IM signal — provided that the 0.70 repeater doesn't time out beforehand. If the system gain is adequate, the repeater could feed back on itself with a resulting characteristic audio howl.  

   Case 3: The 0.31 repeater shuts down. The input to the 0.70 machine drops, but the output is up and stays up. You may hear another signal on the output, or the repeater may break into oscillation. Why? You find a strong signal at 147.30 MHz that is' overloading the repeater receiver, causing it 'to be nonlinear. The strong signal mixes with the second harmonic of the repeater:  2(146.70)  - 147.30  = 293.40 - 147.30 = 146.10 MHz. This signal falls on the .70 repeater input.

   These cases use the popular 2-meter band for illustration. Keep in mind that intermodulation can occur on any band, and can entail an endless combination of frequencies.

Receiver as the Culprit   

 Usually, the first stage of a receiver is the one most likely to overload, causing susceptibility to intermodulation. In some cases, however, later stages may be at fault.  Receiver front-end nonlinearity occurs at relatively low input voltage levels, so the signal power involved is small.   Therefore, the IM products generated there are also low in amplitude. These signals are fed to the receiving Antenna and are radiated. If you track down these weak IM signals, they will lead you back to the receiver site! 

There is a big temptation to use a preamplifier to increase the sensitivity of a receiver or the range of a repeater. The use of these devices in repeater service is strongly discouraged.

Preamplifiers may not have the dynamic range that early stages of communications receivers have, and they usually lack the front-end selectivity found on well-designed receivers. Therefore, they are red-hot candidates for intermodulation. Preamplifiers may generate signals on the input frequency of the repeater, and users have to override these signals in order to be heard. The use of a preamplifier can sometimes degrade system performance!  

 Reducing Receiver IM Susceptibility

   It is important to realize that it takes only one signal, located anywhere in the spectrum, to drive  a  circuit  into nonlinearity, which could produce IM products when one or more other signals are present. In order to minimize intermodulation, your receiver circuitry should provide great attenuation to all frequencies except the  band  of interest.  The receiver dynamic range should also be as large as possible. For this idealized case, only a huge signal in the pass band could possibly cause overload and nonlinear operation.  Any resulting IM signals falling outside of the receiver pass band would be severely attenuated. 

Transmitter as the Culprit   

Serious intermodulation problems can be generated by transmitters when other strong signals present on the antenna (and coupled to the final amplifier) mix with the fundamental and its harmonics. Here the voltage and power levels are much greater than those associated with receiver circuits, so relatively strong IM  signals may be coupled to me transmitting antenna  and be radiated. The transmitter final amplifier is essentially an rf switch, and unless it is operating Class A (which is uncommon) it will be a nonlinear stage. Class C operation is more likely to cause intermodulation than Class  AB1 because it is more nonlinear, wide-band, solid-state amplifiers with low-Q circuits tend to be more susceptible to intermodulation  than are narrow-band,  high-Q configurations.

Reducing Transmitter IM Susceptibility    Fm transmitter final amplifiers are quite nonlinear, and intermodulation can occur if other signals mix in this stage. The Q of most final stages is not high, even for tuned final amplifiers, so the resulting bandwidth is wide. Increasing circuit Q helps the IM problem, but it may be undesirable for other reasons. Eliminating intermodulation in transmitters is, therefore, more difficult than in receivers. The use of a circulator is effective because It presents a very low impedance to signals Going from the transmitter to the antenna, But the high attenuation to signals going from The antenna to the transmitter. Since the circulator must carry full transmitter power, it is an expensive cure for inter- modulation. When duplexers are used at repeater sites, a degree of attenuation is introduced for out-of-band signals, but this alone may not be sufficient if other transmitters are nearby.  

Conclusions    In-band intermodulation products will degrade or destroy your station performance, in addition to interfering with other amateur communications. Out-of- band IM products may play havoc with non-Amateur Radio services. Further- more, transmitting spurious signals, such as IM products, is illegal. Conscientious operators are knowledgeable about inter- modulation and ensure that their stations are free from it.  

Article written By David W. Potter* w2gzd QST May 1983   17 OCR-ed from a copy of the article was done by  Jim Morris K1UGM

David W. Potter, W1GZD, is a spokesman for the Long Island CAME Association (Group Against Malicious Emissions). If it sounds like a hunting club, you're right! The group is engaged in tracking down interference/or local repeater clubs. David's inspiration to write this article came about as a result of his involvement with the group. Persons interested in working with the GAME Association are invited to contact him.