Eliminate Low Frequency Intermod of LNA's

by N2CEI of Down East Microwave Inc.

Because of their performance and design simplicity, both the W6PO (1) and WB5LUA (2) preamplifier designs have been the mainstay of many week signal enthusiasts receive systems on 903 and 1296 MHz. This paper will address a simple way of eliminating low frequency intermod in both . Each design has provided state of the art noise figures and gains for many hams for many years. But, this doesn't happen without some tradeoffs. The ability to produce a good noise figure, assuming the FET of choice is up to the task, requires a low loss wide band input circuit. This in turn produces a amplifier that will amplify undesired signals as well as desired signals often resulting with unwanted intermod products.

A typical preamplifier of either of these designs, (schematic of WB5LUA design below) will produce approximately .6 dB noise figure with 13 dB and up gain at the desired frequency. (3) Preamplifiers optimized for 1296 will exhibit gain from 50 to almost 2304 MHz. In most cases, Interfering signals above the desired operation frequency are normally not as strong or as plentiful as signals from 900 MHz and down. It is the lower frequency's that cause most of the problems.


Antennas do provide a filter function of sort with some low frequency signals, but with the brute power of a local FM radio stations, local TV stations, and pager networks, your receive antenna could sometimes be a hindrance by not having any directivity or having "Strange lobes" at frequencies it is not designed for.

So what's the solution? Why not add a low frequency trap to the input of the preamplifier. Every body has been doing it on the low bands, 6 meters and even 2 meters for decades! Why not on the microwaves? Circuit losses? Well circuit loss do equate to increased noise figure dB for dB. So what we need is a low loss trap! Naturally there will be some loss but how much? And what effect will it have on the receive system if noise figure and gain are sacrificed for increased spectral purity ?

What prompted me to do this experiment and document is a Cell site that is 825 feet away from my stations antenna system. (tower base to tower base) It's signal caused intermod on my 1296 receive system when my antenna was pointed in 6 different directions. By reducing most of the lower frequency mixing products (2 local FM radio and 2 TV stations) with a trap on the input of the preamplifier, I have reduced the intermod to only slight if I point the antenna directly at the cell site (lucky me, its not a popular direction!). In my situation, I don't have my preamplifier tower mounted but the insertion loss of my feed line is only .6 dB from Yagi to preamp. The added noise figure in the preamp that the trap causes (.07 dB) is not much of a factor in my systems total noise figure. I follow the preamplifier with a band pass filter to remove any other products or signals that might cause a problem when the signal enters the transverter's mixer .

Examine the modification to the input circuit. The trap is a short length of # 22 buss wire. (.600 " total length, input connector to ground) The loss of a trap that will attenuate all frequencies below 400 MHz. on a properly adjusted preamplifier is less than a tenth of a dB. (< 0.10 dB) at 1296 MHz which equates to an increase in noise figure by that much. If you need to eliminate higher frequencies or higher level signals, a shorter length trap could be used but extra loss would exist. Depending on your receive system, and situation, it may be acceptable. After installing the trap you will need to shorten the input inductor (L1) by one turn on 1296 and depending on the size of the input inductor up to 3 turns on 903. It may then need to be readjusted for noise figure on your desired frequency if you have a noise figure meter by stretching or compressing it. The gain will drop between .5 to 1.5 dB but the degradation in noise figure will be minimal.

WB5LUA Preamp Circuit Layout with Mods (4)



The standard Preamplifier specs. @ 1296 MHz. 14.14 dBG 0.59 dBNF 7.1 dB input RL

The modified Preamplifier specs. @ 1296 MHz. 13.27 dBG 0.66 dBNF 6.4 dB input RL

Both preamplifiers were optimized at 1296 MHz. for best noise figure before swept data was recorded.


Please review the standard and modified test results and the schematic changes to the generic WB5LUA preamplifier and give them a try if you have a similar problem. I have implemented these changes to W6PO designs with similar results. This improvement may not cure all of your Intermod problems but it should help. The goal is to reduce the unwanted signals enough to get the preamplifier back into a linear state. Then most of the intermod will go away. An added benefit is that the input to the preamplifier now has a DC short on the input circuit adding protection against static and lightning
Footnotes

1-W6PO, Eimac Application Notes # AS-49-36

2-WB5LUA, Al Ward, RF Design, Feb., 1989

3-WB5LUA design as offered by Down East Microwave Inc.

4-Standard 33, 23, 1420,1691 LNA circuit board layout offered by Down East Microwave Inc.