SG-Labs 20 W PA V.3

  • DB2OS

    Changed the title of the thread from “Version 3 von SG-Labs 20 W Pa” to “SG-Labs 20 W PA V.3”.
  • If you want to drive the PA with the TR2300 transverter, V.2 is better.

    When using an Adalm Pluto an extra "pre-amp" is needed to get max. +16..+17dBm, then V.3 is probably the better one after this preamplification. But be careful not to overdrive the PA anyway. There are enough "wide-band-SSB" signals on air.

  • DD8PL , got my new SG-lab PA V3 yesterday. From the housing you can not see any differences, inside you will not find the 3 bridges for the PTT selection any longer but the new Vox and Pre-Amp. I am starting with the tests, so far it looks very promising.

    I drive the PA with the Pluto, Pre-Amp (CN0417) and 6dBm attenuator at a level of about 14dBm, input should typically be 40mW or 16dBm. I measure now a forward Voltage of 3,1V (from 3,5V at 20Watt).
    Further tests are ongoing. This new PA will reduce my effort significantly (can remove the EP AB003). ;)

    73, Oliver

  • Hi,

    I can second, what Oliver writes. I also received the V3-PA yesterday and did some tests and measurements and I am very happy with this new version. I already use the V2-Version in my portable setup, perfectly fitting to the output of the BU500 up-converter.

    The RF-Vox is irritating at first, because when you connect the PA to 28 Volt there is almost no current. The build-in ampere meter of my lab power supply only shows 2 digits after the decimal. I only saw "0.00" and thought, that the PA was not working. However, as soon as the input power is raised above a certain threshold (in my setup around -3dBm), the RF-VOX kicks in and the current rises above 0.6A @ 28V.

    You can also activate the PTT permanently and this leads to a bias current of 0.58A. With the PTT activated I was able to amplify signals of down to around -30dBm, resulting in an output power of around 0dBm, so the amplifier is also well suited for QRPP-experiments.

    All my measurments confirm the claims of SG-lab. I mounted the amplifier onto a heat sink with dimensions of 15x20x2cm and even at transmissions at 20 Watt output, the housing of the PA only got slightly warm.

    The plan is to install the amplifier outside in a waterproof enclosure close to the 1.8m offset dish and possibly also start some experiements with DATV on the wideband transponder later this year.

    DO1OHB: Oliver, I would not be too concerned about putting an attenuator between the pre-amplifier and the PA. The pluto gain itself can be adjusted pretty linearly, and if you set your Pluto drive somewhere between 30 and 50, you already should be in the usuable range for the input of the SG-Lab Pa. The maximum input of the PA is 20dBm, and even if you accidently set the Pluto-gain to 100% you will most likely not exceed this level, because SMA plugs and cable between the pre-amplifier and the PA will have more than 1dB attenuation.

    Vy 73

    Holger 'Geri', DK8KW

  • DK8KW, good point.

    I will take it in account in case I need additional power. Actually I guess the reduced power set-up will be suitable.
    And I know myself setting a Driver to 50% will be forgotten and 100% will be the standard ;) . But as you said, even in that case the PA will not be at risk, just at the border. I can also confirm that the PA is not running hot just warm, I have a 190x200x12mm massive alu-block in a Alu-Wifi-Outdoor housing. With the PA its just 31mm and fit in the upper side of the housing. So all in one and easy to carry also for portable / Mobil-home set up.

    Vy 73, Oliver

  • Hello Oliver,

    >And I know myself setting a Driver to 50% will be forgotten and 100% will be the standard

    Yes, I know what you mean. Thinks that can go wrong will go wrong Murphys Law). I have just send a feature request to Simon to limit the gain slider of SDR Console to a certain fixed vakue somewhere in the menues, so that a this value can only be exceeded if really needed and not in day-to-day operation. Hopefully he will come up with a solution.

    Vy 73

    Holger 'Geri', DK8KW

  • Hello Geri,

    if I understand your problem correct there is already a solution in SDR-Console:

    go to the "LIME-SDR setting" and then to "Drive"

    Here you can activate a "custom drive range" with a checkmark.

    You can then set the minimum and maximum drive levels.

    Here is the explanation from the help window in SDR-Console:


    The default drive range of the LimeSDR is 73dB which is far more than required, also this makes it difficult to select the desired value. A custom range makes adjustments much easier. The exact values you select will depend on the amplifiers used between the LimeSDR and your antenna.


    When running the LimeSDR at maximum output the signal may suffer from Intermodulation distortion (IMD), so set the maximum value to -10dB or -15dB.


    Set the minimum value somewhere between 20dB and 30dB lower than the maximum.

    I have set here my Lime-USB to min -40dB and max -10dB. For me a range of 30dB for setting the TX output power is fully sufficient and with the max of -10dB I cannot overdrive my PA and the output signal from the Lime is nice and clean.

    Kind regards


  • Hi Matthias,

    Thanks for the hint to Sigis recommendation. It worked!

    DO1OHB: Oliver, Sigi wrote in the beta tester forum for SDR-Console:

    "that is just easy cause you have TWO DRIVE sliders

    one in the tx options and one in the main tx dsp panel

    turn main at 100%

    then set the slider in tx options (search under "audio/gain") to your max allowed level (say 80% or so or if you have a high gain setup maybe only 50 or 60?!?)

    then you can not overdrie it anymore ... just do not set that slider higher as 80 ... in main tx box (dsp panel) you can then only turn it further down ... as an example ... i have set my drie in tx options to 85% and in dsp panel to 90 ... i can go a bit hgher if needed (crank up to 100 in dsp panel) but then i begin to have a bit of intermod (other call that splatter) ... so i set my level to be maximum where it is still clean in normal use ... and have some db left for voice peaks (or to break a pile up hi hi) ... and do not have the danger of overdriving the amp".

    Might be useful for your setup, too.

    That saves me at least an attenuator. :)

    Vy 73

    Holger 'Geri', DK8KW

  • DK8KW , DD1US, team that is really helpful and I will change my set-up accordingly. That make me confident to drive the PA within the limits and to save an attenuator ;) .

    This attenuator I will use for the GPSDO-Leo Bodnar to the Pluto. Yesterday I had a run and de-solder the original Pluto TCXO and exchange it with an external SMA jacked. I got also the very good hint to solder C124, but after search and find the place I was surprised. Even with glasses and 2,5 Led-lens I was just able to see the place for C124. At least for me there is no chance to solder a capacitor that is below 1mm. I wonder who could do it and how ;) . Bottom line the GPSDO is now attached via a 3dBm attenuator to the Pluto and working fine.

    Thank you, 73, Oliver

  • Hello Holger,

    yes I also had my doubts but it with a simple 100W big old soldering iron (all ESD grounded) it was easier then expected. But the TCXO is relative large and had space around and the TCXO would then be defect I guess.

    Did you use also the 8mA set-up from Leo Bodnars GPSDO and 3dBm?
    No really sure what that 8mA stand for, as the voltage should always be 3,3Vpp on a 50Ohm output...?

    Vy 73, Oliver

  • Hi Oliver,

    I have a different GPDXO, made in Germany by a company named Cartain. It was a bit cheaper than the Bodnar, and it was abvailable at the time I bought it and Bodnar had a longer delivery times.

    What I did was to check what peope said about the Bodnar-Voltage at 8mA and I basically use the same voltage with the Cartain. I only need a single GPS controlled frequency of 26MHz. I split the output through a Mini Circuits splitter (so at least losing 3dB) and then put another 3dB attenuator in the line before I feed the signal directly into the Pluto.

    The other line goes via a 20m coax cable to the LNB and stabilizes the reception. I am very happy with this entire setup.

    Vy 73

    Holger 'Geri', DK8KW

  • DK8KW , Geri, since Wednesday my QO set up is preliminary final ;) . And you was right, I started the test with a 6dBm attenuator and end up just with a direct connection from the pre-amp to the SG-lab Pa V3.
    I measured with the RF8000, a cheap Chinese power-meter, the input on the Pa and was surprised that I reached 40mW / 16dBm at 86% gain level (SDR Console) without any prior reduction. First QSO run very well.

    Interesting wise I used just 33x33cm Patch-Antenna 9cm high which is very easy to handle.

    So next step is the integration of the RX/TX antenna in one solution.(Patch-antenna and LNB)

    If anyone is interested in a nearly new SG-lab PA V2 please let me know.

    Vy 73 Oliver, DO1OHB