Dipl. Ing. Gyula Nagy, HA8ET

Published in DUBUS 4/2010


The large number of extremely high signal levels at VHF contests causes bigger and bigger problems (spreading of multi-PAs, multi-antennas) nowadays. Low noise figure is easy to achieve with modern devices, so the primary design requirement is the strong signal performance (IMD characteristics). Transistor selection is the first and most important step in an LNA design. You can find a very interesting review about devices with high dynamic range in [1] but unfortunately the most promising one, ATF-53189 by Avago Technologies is not included. The main data for this GaAs FET can be found in Table 1 [2].

The preamplifier described here was produced to replace circuits built with ATF-54143 pHEMT FET [3]. The ATF-54143 is very sensitive to static voltages and TX power, and although the noise figure is good, the device is not unconditionally stable. This can cause many problems in the case of a nearby broadcasting or cellular base station.


Technical data

Frequency range:…..…….144-146 MHz
Input/Output impedance:... 50
Gain:…………………….. >19,5 dB
Noise Figure………..…….<0,8 dB
-1 dB Bandwith:…………. 3,65 MHz
-3 dB Bandwith:…………. 5,81 MHz
S11:……………………...-14,3 dB @ 144MHz
Supply voltage:……...…  +12…15V/140mA
IP3:……………………….> 22 dBm
+40 dBm IP3 GaAs FET (ATF-53189)
Excellent broadcast rejection
Unconditionally stable
Acceptable sensitivity to ESD
Powered through antenna feed line, or separately
Connectors:…………...... BNC or N male/N male, or N male/N female
Dimensions (box):…….... 74 x 37 x 30 mm



Avago Technologies’ ATF-53189 is a high linearity, medium power, low noise E-pHEMT FET in a low cost surface mount SOT89 package. Packaging of the ATF-53189 can be seen in Figure 1. The package has two source leads with large surface areas for efficient heat dissipation and low-inductance RF grounding. The enhancement mode technology provides superior performance while allowing a DC grounded source amplifier with a single polarity power supply to be easily designed and built. The enhancement mode PHEMT requires about +0.6V bias potential between the gate and source (VGS) for the target drain current, IDS.

Fig. 1 SOT-89





Absolute Maximum


Drain–Source Voltage




Gate –Source Voltage


-5 to 1.0


Gate Drain Voltage


-5 to 1.0


Drain Current




Gate Current




Total Power Dissipation



Pin max.

RF Input Power




Channel Temperature




Storage Temperature


-65 to 150



Table 1 ATF-53189 Absolute Maximum Ratings, Electrical specifications                          


Modern design methods require that the ATF-53189 pHEMT should be analysed with a simulator and then followed by measurements. The most favourable DC setup (UDS = 4 V and ID = 135 mA) results in a significant heat dissipation, which the SOT-89 package can provide by soldering to the groundplane. However, this means that source feedback is not applicable, even though it could provide the best matching and stability factor.

The S parameters of the chosen FET between 50 Ohm ports can be seen in Figure 2. The input and output matching is clearly not adequate, especially the input return loss (S11). Figure 3 shows the stability factor, K. When K is greater than unity, the circuit will be unconditionally stable for any combinations of source and load impedance. It can be stated that in the examined frequency range the ATF-53189 cannot be used directly between 50 Ohm ports. The source inductance of the SOT-89 package is so low that no improvement can be obtained by that method.
Voltage feedback from drain to gate (Figure 4) is another way to improve the K stability factor, and this type of feedback need not cause a significant increase in noise figure. The gain of the FET at 145 MHz is approximately 29 dB, which proves to be too high. The amplification can be reduced to 23 dB with a 6 dB attenuator at the output of the preamp, and this attenuator further increases the K stability factor. It can be changed to 10dB if needed.

Fig. 2 S parameters Fig. 3 Stability factor Fig. 4 Simulation

The detailed circuit diagram can be found in Figure 5. A band pass filter was added at the output, which improves the out-of-band selectivity. The FM broadcast rejection is better than -60 dB at 100 MHz. The insertion loss of the filter further reduces the unnecessarily high gain to 20 dB.
The simulated S parameters of the preamplifier without the band pass filter can be seen in Figure 6. Figure 7 shows the stability factor and Figure 8 shows the noise figure.

Fig. 6 S parameters without BP filter Fig. 7 K factor of the preamp Fig. 8 Noise figure

The losses of the input matching circuit are estimated values, so some differences are possible in the practical measurements. The input circuit consists of a ‘T’ match with suitable low loss microwave capacitors and spiral inductor. The K1 security relay needs to be built in if TX power is higher than 200 W. If you connect ground potential onto the PTT connector, the K1 relay will short-circuit the inputs, and this increases the isolation from the TX power by 28 dB. The circuit layout of the built prototype can be seen in Figure 9.

Fig. 9 Layout Fig. 10 Hot air soldering


The professionally made PCBs are 1.5mm, FR4, double sided with plated through holes. The boards are solder resist coated and silk screened to show component designations, i.e. R1, C1 etc. I recommend soldering the SMD components with a hot air soldering tool (Figure 10). The bottom side of the PCB must be soldered to the walls of the box at several places: IC1, the spiral inductor, the band pass filters and the two BNC sockets.

Part list 


Connect +12…15 V to the amplifier (observe DC polarity!) and adjust the potentiometer for the needed 135 mA drain current (135 mV can be measured between the TP1 and TP2 test points).Then tune the L3 L4 TOKO inductors to the maximum signal from a beacon. For minimum noise figure, the CT capacitor at the input should be set to a position 20 degrees from maximum as shown in Figure 11. The preamplifier does not need further adjustments.

Fig. 11 Pretuned completed PCB



The wide-band selectivity of the preamplifier can be seen in Figure 12, and the detailed results are summarised in Table 2. You can see the complete list of figures of the measurement diagrams below.

Fig. 12 Wide band selectivity

Fig. 13 1dB bandwidth

Fig. 14  3dB bandwidth


Frequency [MHz]

Rejection data [dB]









Table 2  Reference frequency: 145 MHz G=20,10 dB


Fig. 15  Reverse Gain Fig. 16  Isolation increase with security relay (+27,7 dB)


Fig. 17  Input Return Loss Fig. 18  Output Return Loss


Fig. 19  First prototype TOI -Id=100 mA (Tnx HA3KZ!) Fig. 20  Noise figure of first prototype, G=16 dB (Tnx HA3KZ!)


Fig. 21  Output signal with open input, without the top of the box (local signals are visible) Fig. 22 Output signal with open input (between 100 MHz and 6 GHz), box covered



Several contest stations used the recently built 20 prototypes with good results during the IARU VHF contest in 2010. I have stocked components for 100 preamplifiers for HAMs who wish to build or use this EXTRA-2 Contest Preamplifier.

The Please KIT or the completed and tested EXTRA-2 144 MHz Contest Preamplifier can be ordered.

Please contact me!

I wish great success for all users!


  1. Henning-Christoph, DK5LV: Design of 2m and 70cm Receiver front-ends with high sensitivity and high dynamic range using GSM devices. DUBUS 4/2007 pp.47
  2. ATF-53189 Datasheet AVAGO Technologies
  3. Peter Hoefsloot, PA3BIY: A very high dynamic range LNA for 144 MHz. DUBUS 1/2002 pp.6

Special thanks to GM3SEK, Ian for reviewing the English text!


Other photos


Copyright HA8ET 2010-11