[NLRS] First tests with DVB-T dongle

[Dr. Gerald N. Johnson]

A very simple and effective up converter needs 3 parts. A low pass 
filter, a Drake TVI filter for HF is quite fine. An oscillator, a $2 
CMOS sine wave out put modula from Mouser or DigiKey is plenty good 
enough, 48, 50, or 64 MHz should be fine. A mixer, and something like a 
MiniCircuits SBL-1 or TUF-1 works well. Dan's Small Parts and Kits 
usually has a selection of perfectly fine MiniCircuits mixers at good 
prices. Feed the HF antenna to the mixer IF port through the low pass 
filter. Connect the LO to the LO port and the vhf receiver to the mixer 
RF port.

There are many ways to complicate this up converter. An MSA0685 or MAR-6 
makes a more than adequate RF stage. You can build a low pass filter 
from several in the ARRL handbook where you have to wind coils on 
toroids. The mixer will handle stronger signals if two of the three 
ports are terminated from DC to light, 3 dB attenuators are a good 
start. There could be an broad band IF stage to do the broad band 
termination while keeping the NF lower, with the same MSA0685 or the 
high power 11 of the family. If you have a crystal on hand, you could 
use a NE612A for an active mixer, but there you have to add a tuned 
circuit to the oscillator to make it go on an overtone for a 48 Mhz 
crystal (which can be had for 59 cents or something like from Mouser) 
and it has balanced outputs that need matching down to 50 ohms.

Many an upconverting HF receiver design tosses in extra attenuation for 
540 to 1700 kHz because of strong signals there even on a short antenna 
that would cause intermod if that band wasn't attenuated. One way is a 
1.75 MHz high pass to go along with the 34 MHz low pass.

A double gate MOSFET would mix well too with LO on the second gate or 
source and signal on the first gate. The output Z is a lot higher than 
the antenna port on the VHF receiver so it takes matching also.

I KNOW that first scheme works because I built one 10 or 15 years ago 
with the 0685 RF stage and a 144 MHz 7th or 9th overtone crystal 
oscillator. It works fine with the 2m IF over 4 MHz, but works 
adequately listening at 432 to 450 giving a wider HF tuning range. The 
mixer insertion loss is greater at the higher IF frequency but not so 
much as to make it unworkable. I have a box of assorted filters that 
I've built over the more than half century of building stuff along with 
some commercial filters and I used the Drake I think.

Anyway, there's no point in complicating the up converter, the simple 
one works.

At CSVHF a couple years ago someone proposed using DVB-T for ham digital 
TV since the one used for commercial TV has a stiff licensing fee and 
very restricted documentation and some say performs poorly when the 
receiver or transmitter is in motion, a typical ham operating mode.

73, Jerry, K0CQ

On 7/16/2012 8:45 PM, Doug Reed wrote:
>
>
> DVB-T TV is not used in the USA but there are a few US vendors selling
> the dongles. But the easiest place to get them is on Ebay, assuming you
> use Ebay. Just search for "dvb-t rtl2832 e4000" and you should find
> quite a few sellers. Don't bother buying a unit with a remote control,
> since the SDR software doesn't support them and you will not be using
> the original software....
>
> <http://www.ebay.com/itm/White-DVB-DVB-T-HDTV-Realtek-RTL2832U-Elonics-E4000-64-1700-MHz-SDR-GNU-P335-/370607013140?pt=US_Video_Capture_TV_Tuner_Cards&hash=item5649e63d14>
> This picture looks like one of mine.
>
> Some recent emails mentioned experiments to feed the antenna RF directly
> to the baseband input of the RTL2832 chip, bypassing the tuner chip. The
> idea is to directly sample everything from zero to 3MHz. There is also a
> question if the RTL2832 can support sampling in multiple "IF bands"
> under 30MHz.... There is still a LOT of development going on and not a
> lot of info available on the parts because of non-disclosure agreements.
>
> And there are people building simple up-converters from HF to VHF so the
> FunCube or DVB-T dongle can be used with HF signals. Keep in mind the
> frequency stability and accuracy of the dongle is nothing to write home
> about.....
>
> Probably the biggest problem is overload of the "DC-to-daylight"
> wideband front-end. The E4000 chip has adjustable gain but people are
> still trying to figure out how to set it correctly.... Things are
> getting better as a few people turn up with NDA's and time to write
> code. It is very much a "work in progress."
>
> 73, Doug Reed, N0NAS.
>
> S. Earl Jarosh wrote:
>> Check out
>> http://wiki.spench.net/wiki/RTL2832
>> For some source code
>> BTW where can I get a couple to play with?
>> S. Earl Jarosh, N0HZ
> ______________________________________________________________
>