Vorläufig / Preliminary, work in progress
If you mouse click on a picture it will be expanded. You can go back to the web page with the back arrow of your web browser.
RF Amplifier 3W
In Chinese shops you can buy for about 12 EUR an RF-Amplifier module with an advertised output power of 3 W.
On the right is a picture of the module.
Let's see with some measurements, if it does so.
- Supply voltage:DC+12~15 V (typical 15 V 0.3-0.5 A)
- Working frequency: 2 MHZ-700 MHZ, input and output impedance: 50 ohms;
- Maximum output power: 34.8 dBm (3W)
Maximum input power: <+10 dBm
Typical power Gain: 35 dB (> 300 MHz Gain slightly decreased)
- Size: 7 x 3 cm / 2.76 x 1.18 in
- Weight: 28 g
I use the following equipment:
Adjustable Power Supply with current limiting.
Step Attenuator 0 - 31 dB
Attenuator 20 dB (SMA connectors)
RF Wattmeter RW-151D, company Kuranichi Keisokui (5, 25, 150 W fullscale)
RF Testgenerator frequency 10 MHz, output power +4 dBm at 50 Ohm, waveform: sinus.
DSO Hantek DSO5072P
- Coax cables to connect
The input power was varied between -16 dBm to -22 dBm
The gain factor was about 46.6 - 49.6 dB
At 12 V supply the maximum output power (sine) was 0.7 W = 28.6 dBm = 17.0 Vpp
At 15 V supply the maximum output power (sine) was 1.2 W = 30.7 dBm = 21.7 Vpp
The current draw was 0.54 A to 0.58 A
The power efficiency was 11% to 15%
See following the DSO screen shots from the output signal:
The left screen shot shows an overdrive at Vcc = 12 V, on the right you see the output with the same drive level but at Vcc = 15 V.
If you want to see more details of the measurement, have a look at this spread sheet (Open Office).
With the Vector Network Analyzer nanoVNA and some attenuators it was possible to measure the frequency response.
See below the setup:
See below the the frequency response measurement:
There is no gain linearity specified. When you look at the gain specification, it says 35 dB. Then it is +13 / -5 dB, what is OK. The sentence (> 300 MHz Gain slightly decreased) means in reality about 5 dB.
While testing the amplifier, I found that the heat sink gets pretty hot. So, I limited my measuring time to less than a minute. With a lower quiescence current of 0.3 A, see chapter Improvement, I measured with an infrared thermometer the package temperature and calculated a thermal resistance of the mounted heat sink to about 8.4°C/W.
Some details for calculation:
- MOSFET U2 label: K2 1Y1, found no data sheet
- Heat sink: 30 x 70 x 12 mm, area = 21 cm2, estimated thermal resistance 8.4°C/W
- U2 Package: SOT-223, thermal resistance 15°C/W
- Operating Junction Temperature, Limit 150°C
- Operating Junction Temperature, safe Limit 130°C
- Maximum safe temperature rise at room temperature of 30°C = 100°C
Thermal resistance of thermal compound: 1°C/W
Total thermal resistance: 8.4 + 1 + 15 = 24.4°C/W
Maximum safe power loss: 100 / 24.4 = 4.1 W
With the previous setups (0.5 A quiescence current) it is no possible to get a 100% duty cycle operation, without additional cooling.
For more background info see PA Cooling.
What wonders me was the high quiescence current. So, I connected a trim pot of 10 KOhm parallel to R6 and adjusted the quiescence current for best efficiency at an output sine waveform at 10 MHz. That gave a resistance of 5.6 KOhm in parallel to R6, and a quiescence current of about 300 mA. Now the values are:
- The input power was varied between -18 dBm to -20 dBm for output sine wave
The gain factor was about 48 dB
At 12 V supply the maximum output power (sine) was 0.7 W = 28.6 dBm = 17 Vpp
At 15 V supply the maximum output power (sine) was 1.1 W = 30.4 dBm = 21 Vpp
The current draw was 0.3 A
The power efficiency was 19% (12 V) to 24% (15 V)
The power loss was 2.9 W (12 V) or 3.4 W (15 V), allows 100% duty cycle (max 4.1 W)
Within the measured values this amplifier is useful, especially considering the price. You just have to consider, that it is not a broad band linear amplifier. If you do not want to mount a fan to the heat sink if you need 100% duty cycle, apply my Improvement.
Contact Email, please enter your Email address
The entered Email address will not be published, or given away.
List of pages in this category:
-- RudolfReuter 2020-05-17 17:50:40