... Mek's creations and more

Up converter for receiving of shortwave stations on PC

A cheap USB DVB-T, DAB+ and FM receiver is available on the interner, which many enthuiasts use to listen to other frequencies. This receiver, based on RTL2832 or R820T2 chipset, has many Chinese clones, but all of them have similar frequency range - from 50 MHz to 2 GHz. There is no FM or AM tuner in this USB dongle, demodulation is done solely in software - either in software made by its manufacturer, or in a universal program, such as the popular SDR Sharp. It shows a nice frequency spectrum of around 1 MHz so that you can see what is "in the air" and which frequencies to focus on while tuning.

Hardware restrictions of these cheap USB dongles do not allow receiving of shortwave frequencies, which are typically a few MHz. It can be circumventer by using an up converter which shifts these low frequencies into higher frequencies that can be received by the USB dongle.

This up converter adds 100 MHz to frequencies in the 0 - 30 MHz range, so that shortwave stations can be tuned on frequencies 100 - 130 MHz. Because no crystal oscillator oscillates precisely at 100 MHz - it is a bit less or more - it is wise to find out the exact frequency by a high "peak" around 100 MHz in SDR Sharp and write this frequency in the Shift field. Then you can happily tune on the "real" low frequencies and the software computes the frequency given by up converter. I have written value -99 992 500 into the Shift field, as can be seen in the video below.

In my case, there is unexplainable interference below 2 MHz, that changes its frequency depending on how I change frequency while tuning, and it changes frequency faster than I tune - I can't explain in better. But I wasn't able to get rid of this annoyance. Probably it depends also on PCB - one man from internet built this converter on a professional-looking PCB with SMD components and encountered no interference at all. When I build this converter again, using his PCB layout, the situation got much better. I was able to receive stations down to 1 MHz, but the annoying interference was still there... I dunno.

The schematics is taken from Tomasz Watorowski's page. However, there are a few important things not described there that I had to find out myself.

Inductors are air ones, everyone can make his own depending on which kind of Cu wire and diameter of pencil or a pen to wind on is at hand. I have used this Czech page to calculate inductors of required inductance. All my inductors are from laminated 0.5 mm-diameter Cu wire that can be easily obtained from a defunct TV, transformer, etc.

  • 150 nH: 6 turns, 7.4 mm coil diameter, 11 mm coil length
  • 270 nH: 9 turns, 5.5 mm coil diameter, 8 mm coil length
  • 330 nH: 9 turns, 7.4 mm coil diameter, 12 mm coil length
  • 470 nH: 12 turns, 7.4 mm coil diameter, 15 mm coil length

Input coil of 56 μH inductance is used from a dismantled TV, I didn't make it myself.

I only needed to buy two things to build this up converter::

Unfortunately, both are only available in SMD package.

Input voltage of 5V is taken from USB.

Update 10th June 2017: I managed to filter out some interference around 4 MHz and I got some more SW stations that were not there before. You can do the same by adding a 47 pF capacitor in series with antenna input. Unfortunately, there is still much interference below 3,5 MHz.

Last important thing is to set up RF Gain to 14-21 dB in SDR Sharp, otherwise you don't get any signal. This fact is also not mentioned anywhere (as it if was self-explaining), which has left me to wonder why the up converter is not working. Of course, you need to switch to AM mode.

I designed the PCB myself. Its dimensions are 8 cm x 8.5 cm and it is designed to fit inside Z76 box. There is only one wire connection - from Vcc to pin 8 of SA612.

Shortwave radios are best received after sunset and at night, when signal propagates best. I wish you good luck with building this up converter and happy tuning!

Comments (12)

:} lol :D =) :o) :B 8-D :P :-)) :-< ;) :-/ :( :.( O.o ;o) :-* 8-| :-| 8)
Cowboy   11th February 2023 4:18 pm
Many helps to me. Good job.
Mahdi   26th November 2019 8:27 pm
with the sma input cable in place, I connected a 1 meter long wire to the input of the circuit, but it didn't help! those multiple high peaks around 100 mhz were still there. maybe I should desolder that completely, and give it another chance. one interesting thing is that when I move the battery around the circuit at various positions, the waveforms change! I will try to assemble a clean copy of it on a breadboard at some future time when I am less busy.
thanks again
Mek   26th November 2019 4:40 pm
The SMA connector is not a problem. The antenna should be just a simple wire - no coaxial, no shielded cable, nothing. Just a simple copper wire. Solder it on the input where the SMA connector is (you can leave the SMA connector soldered as well). Length of the cable should be a few meters, the more length the better, and should be placed horizontally, in highest place possible.
I'm pretty sure this will help :)
Mahdi   26th November 2019 8:52 am
you mean I should replace the antenna? in the figure below, the purple coaxial is input, on one end it has female sma, the other end is soldered to the board. I just bought it as a connector, I don't know the impedance of it, whether it is 50 ohm or 75 ohm. a pdf file of it is uploaded in:
the shield is soldered to the pcb gnd, and the inner connector is soldered to the lead of the input capacitor, C4.
for connecting any other antenna, I used the same cable, my sdr dongle is shipped with an antenna with a magnetic base and a male sma connector; I used it. I connected the male sma coming out of the magnetic base to input female sma of my pcb, and with some alligator clip, I attached some wire in place of antenna at the magnetic base. It made almost 1 m length of coaxial cable.
Do you think that is the problem? should I disconnect the female sma connector?
Mek   25th November 2019 6:05 pm
OK, that may mean the problem actually is the antenna. Is it just a wire, or a standard 50 or 75-ohm coaxial cable? I have seen people using coaxial, no wonder they didn't get anything out of it, so just asking :)
I know the "hack" of replacing a dll in sdrsharp folder, it didn't work well for me either. Anyway, you can get best reception in the 6 MHz and 9 MHz bands, the higher the frequency, the least you can get (because not many stations broadcast there). Also the higher bands are better propagated during day, not night. If you replace a DLL then maybe you can get down to 16 MHz or so, but it's still not enough for shortwave.
It's normal that you get FM stations around 20 MHz, I am getting them as well. I consider it a bug somewhere in the RTL-SDR design :)
Mahdi   24th November 2019 11:34 pm
now I can say that the problem is not the antenna! because even without any antenna, just sma connector is left unconnected to anything, the same peaks and noises are present everywhere.
In the future days I will try to make a new copy, as I said in my previous comment, to see whether it has the same problem.
by the way, there is one experimental method by Oliver Jowett for rtl sdr HF reception, it is a single dll file to replace in the sdr sharp folder. even with that, I am still receiving FM stations! with that software modification I can tune to above 15 Mhz, but there are just some FM stations there! a single station appeared on 16 Mhz, and at the same time, on 90 Mhz! I dont know why.
any way, thanks for your support and help
If I come up with any solution, I will let you know.
Mek   24th November 2019 10:19 pm
Try just a long wire and connect it directly in place of the "SMA" input as on the schematic. The stock antenna that came with the dongle will not work, it's just not meant for short wave.
About the overloading - I only have read it somewhere, that if there is a strong signal somewhere, mirrors of it can appear in various places around the spectrum. I also saw it when my brother used his PMR to broadcast, the transmission was there in various frequencies, not just the only one. May be the same situation here. But I don't think it's important.
You can connect your shielding to GND, it's OK. I just didn't use a traditional SMA connector so there was no shielding to connect, that's all :)
I don't know about the other circuit you posted link to, it may work as well. I have used NE612 also in other projects, and it worked without any filters, but it is commonly known that the NE612 / SA612 is easily overloaded by strong signals and then nothing usable comes out of it, so I ended up using some simple filter as well in my other projects, but some other, older projects are without filters and perform just fine. I guess the filters in the up converter are there only to suppress regular FM stations (but they are still there, only weaker). Try the other circuit and see if that one works. Otherwise I don't know why this one doesn't work for you :(
Mahdi   24th November 2019 10:00 pm
yes, the two TO92 packages are 78l05 for sa602 and lf33 for oscillator
currently in my country it's night, and with conventional radios I can receive short wave now.
for the antenna I have tested possible configurations, the same telescopic antenna shipped with my sdr dongle, with various lengths of it extended. I have also tried connecting a long wire inside room to that antenna when it is in the shortest length. I also connected some wire to an aluminium window pane, and from that again to the antenna. they all make the same result.
why my sdr is overloaded?? what are the possible causes?
in your uploaded file above, you mentioned that do not connected shield of input wire to the circuit gnd, but I have it connected to gnd. can it be the cause?
I have another sa602 and 100 Mhz oscillator, and the pdf below has made an up-converter without any input or output filters, or impedance matching. maybe I should give it a try with my another sa602. It has just 3 capacitors to the inputs of ne602, and a voltage divider at the output of oscillator.
Mek   24th November 2019 9:27 pm
If you have many peaks, that may mean the SDR is getting overloaded, and those peaks are only some false peaks, not really present... otherwise I don't know how to explain them.
Your power source is a 9V battery, but the circuit should be powered by 5V. I can see you are having a TO92 case in your circuit, is this a 78L05?
Try tuning around 6 MHz and around 9 MHz, those are most busy on short waves. Also, what is your antenna?
Another important note - try tuning in the evening and at night, when short waves propagation is best.
Mahdi   24th November 2019 9:17 pm
Many thanks, I really appreciate your kind reception in accepting my request and answering my questions
I've played with the gain, it has some effects, but still no reception
I think there is something wrong with my circuit, there are actually many sharp peaks around 100 Mhz! not one. Some screenshots are given:
I don't know whats wrong with it. my circuit is assembled on a perfboard, all components except sa602 are through hole, capacitors are multi-layer, inductors are fixed type inductors like resistors, and it's powered by a 9 V battery. for L4, which is 470 nh, I used two 100 nh and one 270 nh in series. some pictures of it are shown in the links below. the black wire sma is output, the purple sma is input.
one interesintg thing about it is that when I move the 9 V battery around, signals change! not much, but there are some changes happening
somewhere else I read that L4 and C14 are there to suppress crystall oscillations at 100 Mhz, but in my circuit there are many noises around 100 Mhz. I dont know what to do else with it. do you have any suggestions for me?
Mek   24th November 2019 3:24 pm
Hi, Mahdi,
make sure you have RF Gain in your SdrSharp or any other program you may be using, set to some higher level, from 20 to 40 dB, otherwise you get nothing.
It's good there is a sharp peak at 100 Mhz, that means the oscillator is working. The peak should also disappear when you disconnect the upconverter from power.
Otherwise, I would prefer communicating using these comments under the article, because other people may find them useful as well. Good luck troubleshooting.
Mahdi   24th November 2019 2:01 pm
Hi, I am Mahdi
I just build this up-converter, but mine doesn't work at all, there are only some sharp peaks around 100 Mhz, some FM stations, and the rest is all noise, and I am getting mad with it
I was searching through the net and I arrived at your page
Is there any chance I can get in contact with you, maybe by email, and get some help on trouble shooting my converter?