LNA 144 MHz cavity super low noise amplifier with mgf-2430

 

The point is, on my opinion, that this is the best device and the best compromise (noise/dynamic) until now available for the 2 meters band for use either in E.M.E. and contest.

 

Silvered 70 ohm cavity input. Diplexer and elliptic (2 or 3 stage) filter.

 

Characteristics measured:


NF     <0,2 dB

G       >21 dB

OIP3  +40 dBm!! (± 2 dBm)

BW output    ±3MHz @ -3dB

                   ±5MHz @ -9dB

current 200 mA

 

 

The construction is simple but you must pay attention to some mechanical steps into the input line. In particular the coupling with related soldering of the cavity and the central pin of the input connector.

 

The schematics diagram comes from VE3KH. The input is obviously shorted to ground for the DC and statics discharges. This keep the gate of gaasfet safe.

Many informations can be found on Dubus 3/94 by I5TDJ (SK) and I8CSV (SK).

 

Pay attention to the cavity tuning capacitor during the soldering. High temperature will easily destroy the component.

The silver plating of the cavity is particulary important to lower the losses (these are directly related to the noise figure of the preamplifier) and to preserve the characteristic of the device on the time.

 

The total gain of my preamplifiers is a little bit lower of the average because the loss of the output elliptic filter is quite high. A narrow output is important to protect the receiver front-ent from the nearby utilities strong signals. Furthermore the loss is almost inifluential because, being it post gaasfet (same as the coax feed line) it adds a marginal deterioration of the NF (Friis equation). Therefore the output gain becomes relative and even a 3 dB loss feed line will increase the total NF only of 0,015 dB. On the other side a low gain preamplifier is also good to not overload the receiver front-end.

 

Despite some mechanical compromises, the cavity construction takes several hours because of the precision due.

 

The input must be correctly terminated and put as close as possible to the antenna/s to avoid the natural losses of the cable and also to avoid possible auto-oscillations. Some home-brewer reported instability problems solved in several ways. To avoid this, I placed a ferrite bead on the gaasfet drain. This reduces a bit the gain and preserve instability. Both on workbench or in the working position on the feed line, my lna is stable.

 

the cost of the realization of this preamplifier is relevant. High quality connectors, Airtronic compensator, output elliptic filter and particularly the gaasfet (I remind you that the mgf-2430 is able to product >1 watt @ 10 GHz!) will increase the cost of this kind of realizations.

On the other way you are sure to have THE DEFINITIVE LNA, not commercially available and, not comparable with the usual preamplifiers, even if good, regarding the report noise/dynamic.

 

sometimes is hard to believe when I read the values published of the some well–known manufacturers of LNA that use common devices and never could reach certain values

  

I attach some photos during construction and measures hoping that they can be of help to those who want to build this LNA


73, Stefano IZ3KGJ

 

 

 

Test instruments: 

Tektronix 495P + TG

HP 8560A

EATON 2075 (head with 30 dB attenuator and output 5 ENR calibrated)

Boonton 42B, 41-4A