Linear Amp for Raspberry Pi 2 GPIO transmitter?

Talk about other SDR products like the FunCube, HackRF, BladeRF etc.
Post Reply
spirewalk
Posts: 2
Joined: Sat Apr 29, 2017 7:48 pm

Linear Amp for Raspberry Pi 2 GPIO transmitter?

Post by spirewalk » Sat Apr 29, 2017 7:54 pm

I want to increase the transmission output to 10-20 watts, what sort of linear amp do I need, and what would be a suitable low pass filter?

snn47
Posts: 192
Joined: Tue Dec 27, 2016 11:00 pm

you need also a filter and a liscense to radiate

Post by snn47 » Wed Jul 19, 2017 3:36 pm

Just two question did you consider that you need a liscence for the signal type and frequency you want to radiate 20 W at, and
did you consider that you also need filters, in order to attenuate harmonics of your signal sufficiently and to to keep spurious radiation of the signal narrow?

If not inform yourself before you get introuble.

rtlsdrblog
Site Admin
Posts: 2410
Joined: Mon Nov 19, 2012 11:54 pm

Re: Linear Amp for Raspberry Pi 2 GPIO transmitter?

Post by rtlsdrblog » Fri Jul 21, 2017 3:12 am

Yes what frequency are you looking at TXing on? 20W is quite a lot, are you licensed to TX? You likely need a preamp, into a linear amp.

It's actually quite difficult to build a good filter for RpiTX because it needs to be very narrowband. You can't just build a low pass filter, it should be a band pass filter.

spirewalk
Posts: 2
Joined: Sat Apr 29, 2017 7:48 pm

Re: Linear Amp for Raspberry Pi 2 GPIO transmitter?

Post by spirewalk » Thu Jul 27, 2017 1:19 pm

I am a licensed ham operator. I used this 20M TX shield (https://tapr.org/kits_20M-wspr-pi.html) to filter the harmonics. It is designed for WSPR, but also has capability for CW and SSB 20M.

Okay, so I need a pre-amp. What brand or product do you recommend?

snn47
Posts: 192
Joined: Tue Dec 27, 2016 11:00 pm

Questionable if wideband Lin-Amp for RaspPi 2 GPIO can be us

Post by snn47 » Sun Jul 30, 2017 3:13 am

The main problem is that the GBIO output produces a wideband noise. Even with liscence it is not just the harmonicics that need to attunated, but also widebadn noise.

The author only provided pictures using 100 MHz wide frequency range scan, so I wonder how bad the signal really is close up. I remember an bad 2m Tx that produced a few 100 kHz wide CW-click which stopped all from using the lower band part. Amplifying a noise signal therefore shouldn't imho not be attempted. While attenuation of harmonics can be achieved using filters, it does not solve the wideband noise problem.

I wonder if someone has tried using a tunable highly selective and narrow preselector, similar to preselector older tube-based SW TRx used to have, directly after the GBIO output. This requires retuning after frequency change by more than a few kHz.

If you look at the harmonics, while the first BPF increased the attenuation to 50 dB, harmonics degraded to -46 dB after adding the amplifier and following BPF. More important the noise spectrum increased again. Getting to +43 dBm, by adding a wideband amplifier having 23 dB of gain, will increase the noise spectrum, and additional noise not seen due to the spectrum analyzer limitations in dynamic range and because the additional amplifier will generate additional noise due to IM, will appear.

Having provided my reasoning, an alternative circuit for amplification could be the circuit described originally in UKW-Berichte by DJ7VY. It was intended for use as amplifier for low power transverter e.g. those using IE500 diode ring mixers. You will of course still need the addition to the preselector and a filter capable of the power output of the amplifier after the amplifier, and sufficient attenuation between GBIO and the preselector so you will not saturate the amplifier. With this circuit you should be able to drive other PA. A copy of the article can be found here http://www.robkalmeijer.nl/techniek/ele ... index.html or obtained directly from UK-Berichte.

Post Reply