Thursday, 28 November 2013

Satellite TV, RX and positioning

Satellite TV, RX and positioning

Playing with satellites TV or radio is not too hard or expensive. I did a fair bit on it a few years ago (C and K band, ~4 GHz and ~12 GHz) and still have some of the gear. I could see my house in Brisbane on Google maps from the big white C band dish, ex Sky. It was a heavy fixed dish, jerry-rigged with a linear-actuator for position control. 

A full new C (and usable on K) band system is around $400 on eBay.   Sometimes they are available for free, migrants and expats used them to get overseas TV. I had one, disassembled, but tossed it when we moved to Gold Coast. Should/may have kept pole, as big steel is expensive.

All satellite TV use a LNB (low noise block) to receive and down convert to a standard IF range of about 1 to 2 GHz for a TV set-top box. The coax is used to carry control signals as tones (diseqc) or DC. The coax is very high performance, but cheap because of the massive use of it.

The big dishes are controlled by linear actuators and a positioner box. I have one set. I last used it to with a remote control to raise and lower a TV, (on big drawer slides) that was in front of a window (acker Heath Robinson(UK), Rube Goldberg(USA), Bruce Petty(AU)) . It worked but wasn’t pretty. Disassembled now. I will try to find photo. Linear actuators are cheap ~$75 and powerful; could be used for lots of weird purposes.


A big dish can be used with any frequency just by changing the receiver/antenna at the focal point. Dishes are only useful for higher frequencies as the receiving antennas get too big.


For lower frequencies, Yagi or similar types, often multiple,  are used.
 

Positioning dishes is fairly easy. One dimension is plug and play, with many receivers having built-in positioner control, or using a separate box. I have a Superjack DG-120+l and control box. I used the box for it with the TV lift.
 
Two dimensions is not much harder, just use two positioners bolted together at 90 degrees, as per the Yaesu G-5500. Any positioner or rotators can be used (see ARRL Antenna Handbook re satellites). With two rotators or linear actuators, quite big dishes or arrays can be controlled. One advantage of satellite antenna is that they can be on the ground, putting them higher may be necessary to clear obstacles.  
 

However, the control is manual with two controllers, again per Yaesu G-5500. Low orbit satellites (like Funcube or weather) are visible for about 15 minutes. It may not be too hard to track them just using signal strength and knowing their trajectory. Finding them as they come over the horizon is probably the difficult bit, but again, their position is known and the positioners can be calibrated.
 

It is "possible" to track satellites automatically, I suppose, but I have not seen a reasonably priced system. There may be some for the Yaesu G-5000, given that it is a standard piece of amateur hardware. A Raspberry-Pi or similar could do it, the actuators all give position feedback (reed-switch pulses as minimum) but the program would be fairly equipment dependent.
 
I may be able to track satellites on 2m or 70 cm. I am currently building a tilting mast (literally: yacht) and a 2 m vertical and a 70 cm crossed pair antenna. The antenna will be on a horizontal fibreglass pipe I had made ~$80, as a "T", with a rotator below. I could put a second pole-mount rotator on the fibreglass pipe to give vertical rotation. With the two rotators, I could point an antenna anywhere in the sky. It would be easy to make the 70 cm circular polarization with phasing. Then I could TX and RX! Could do same with 2m if make crossed too.


Some of the receivers (and SDR satellite programs, such as BlindScan and CrazyScan), have positioner control via diseqc, but one dimension.
 

Changing a TV channel with the remote can include moving the dish automatically; pretty neat really. Very common in Europe and Asia.
 

Given that most amateur TV is DVB-S, I might check some of this anyway, as I want to be able to receive the picture (via box or card), as well as the signal (using BladeRF) from Brisbane. I have got an old Foxtel K band dish on the roof and a 1200 mm dish in the back shed.
 

 
Awesome spook card, it has a pass-through for rtl-sdr or BladeRF SDR. Using Crazyscan/Blindscan and BladeRF with SDR-Console (or what-ever for spectrum analysis) just sounded too cool…  Little Pentium system might get used for something other than print server and scanner.
 

With a dish, at best, I can only use the 1200 dish with a positioner on existing Foxtel mount. I don't have space for a C band dish again. Not that there is much on C band now, most is on K.
 
Might drag out old Dream 800 set top box and see what still works. Dream box has pass through, so I might be able to get satellite spectrum; a first IF tap again! Bizarre if it works.
 
To be continued...

 

Saturday, 9 November 2013

BladeRF on Haswell i5 running Windows 8 with SDR-Console at 935 MHz 20 MHz bandwidth

BladeRF on Haswell i5 running Windows 8 with SDR-Console at 935 MHz 20 MHz bandwidth

Screen shot of BladeRF running on Windows 8 with i5 Haswell processor.



The BladeRF windows installer uses an unsigned driver. With the extra security of Windows 8 it will not normally even give you the option of installing unsigned drivers, Windows 7 does. However Windows 8 has a special restart where the unsigned driver block can be disabled. I think I described it in earlier blogs.

Using the earlier beta of SDR-console V2 per earlier blogs, the bandwidth is 20 MHz rather than 30. I think the narrower bandwidth is more appropriate re Nyquist, about half the 38.5 MHz bandwidth of the BladeRF. At 20 MHz, the CPU is barely busy at about 4 % versus 10 times that of screen shots at 30 MHz in earlier blogs. Simon Brown, the author of SDR-Console said in his Yahoo forum that the FFT runs at 30 MHz too. Dropping the bandwidth to 20 MHz seems to calm everything down.

   Screen shot of BladeRF running on Windows 8 with i5 Haswell CPU showing CPU load



My computer uses an ITX motherboard but is capable of running a (cheap) 27" 4K monitor (same LCD panel as Apple, Dell etc.) through its Displayport using Intel integrated graphics, while it is doing all of this. It gets a bit warm, but is in a tiny case. A HDMI camera input PCIe uses the one expansion slot. 4K monitors are very good for a number of reasons, but I will discuss that in another post.

Screen shot of BladeRF running on Windows 8 with i5 Haswell CPU showing system components and temperatures.


Screen shot of BladeRF running on Windows 8 with i5 Haswell CPU showing detailed core parameters


Summary: BladeRF runs on Windows 8 using SDR-Console at 20 MHz bandwidth with no issues.

The whole point: BladeRF receiving DVB-T test transmission from VK4ZXI 1mW at 2 m from UT-100C DVB-T usb TX

BladeRF receiving DVB-T test transmission from VK4ZXI 1mW at 2 m from UT-100C DVB-T usb TX

Well, it works. UT-100C USB DVB-T TX transmission from my laptop being received by BladeRF on 70 cm channel.

Not a bad signal with few spurious. Amplifiers will have low pass filter.




Now to do the amplifiers and antenna.

Friday, 8 November 2013

BladeRF with SDR-Console- Sceenshots (draft)

BladeRF with SDR-Console- Sceenshots (draft)

Just a quick post of some screen shots of the brilliant BladeRF running on the equally impressive SDR-Console by Simon Brown to show cababilities of both and the noy-surprising heavy load on an i5 2500K processor.

Free to air TV channel, 7 MHz wide with BladeRF running 30 MHz bandwidth




Same signal but one edge with 150 kHz bandwidth to show detail



 
 
Machine performance with 30 MHz bandwidth, CPU @ 67 C and fan whizzing. Ran like this for 20 hours, so all quite stable.

 

 
 
CPU load with some interesting signals. Reported stuttering is probably not a fast enough CPU. 30 MHz at 12 bit resolution would push anything.


 
 
Plan to try it on Windows 8 machine. Reported problems may be due to Windows  8 not accepting (or telling you) unsigned drivers. Such can be loaded in special restart mode.

Thursday, 7 November 2013

It lives! BladeRF SDR on Windows using SDR Console V2: 30 MHz bandwidth, 300-3.8GHz



Summary of BladeRF SDR TRX

BladeRF is a high performance SDR transceiver made by a small start-up company, Nuand http://www.nuand.com/bladeRF.

Currently only mainboard is available for US$420, with a HF/VHF transverter due late November to give coverage down to 10 MHz. For receive only, an up-converter for RTL-SDR dongles could be used to go lower.

Technical Specifications:

•Fully bus-powered USB 3.0 SuperSpeed Software Defined Radio

•Portable, handheld form factor: 5" by 3.5"

•Extensible gold plated RF SMA connectors

•300MHz - 3.8GHz RF frequency range

•Independent RX/TX 12-bit 40MSPS quadrature sampling:  LMS6002D is a field programmable RF  transceiver http://www.limemicro.com/products/LMS6002D.php
 
 

•Capable of achieving full-duplex 28MHz channels

•16-bit DAC factory calibrated 38.4MHz +/-1ppm VCTCXO

•On-board 200MHz ARM9 SOC with 512KB embedded SRAM (JTAG port available)

•On-board 15KLE or 115KLE Altera Cyclone 4 E FPGA (JTAG port available)

•2x2 MIMO configurable with SMB cable, expandable up to 4x4

•Modular expansion board design for adding GPIO, Ethernet, and 1PPS sync signal and expanding frequency range, and power limits

•DC power jack for running headless

•Highly efficient, low noise power architecture

•Stable Linux, Windows, Mac and GNURadio software support

•Hardware capable of operating as a spectrum analyser, vector signal analyser, and vector signal generator

Being a start-up, much of the software is still being developed, but some is available to make an awesome SDR receiver cum spectrum analyser.

Why?

While I was interested in them as an SDR, my immediate need was for a spectrum analyser for my DVB-T project covered in an earlier post.

I am trying to amplify the 1 mW DTV-T output on the 70 cm channel from the UT-100C modulator dongle. I have three amplifiers, trying to get about 5 W or so, to transmit.

Progress

Bottom line of BladeRF: it works as specified.
 

 




A commercial TV signal at ~635 MHz with 30 MHz bandwidth on USB3 Windows 7.

My Sandy Bridge i5 CPU is running at 37% and its fan is wizzing. Bloody amazing!
Imagine if it could be demodulated on the computer as well. I think there are 5 TV channels 7 MHz band.

I had to use the TV antenna with its masthead amplifier. My discone was a waste, too low and we are in a shadow area for free to air TV.

How

I initially tried to get it working on a Windows 8, but ended up getting it going on a clean Windows 7 i5 Sandy Bridge machine, first using USB2, then with no drama on USB3.

Per the Nuand blog entry, http://www.nuand.com/blog/platform-support-update/, the stand-alone Windows installer (http://nuand.com/downloads/bladerf_win_installer.exe) works. It puts everything into the Program Files (x86) directory. The installer includes firmware and the FPGA image (.rbf files). The main program to manage the BladeRF is “bladeRF” (in start list).

Run BladeRF from Start menu to get command window. Check you can talk to device. Load FGGA image and LEDs on board should start flickering. That’s all! Then to SDR software.

 Updated .rbf files can be substituted as they become available. The FPGA image must be loaded each time the device is turned on or reset, not sure why, but no big deal at this stage. 

As such, there is no need to go through the rather complex process of compiling everything to do the install as detailed in the wiki: https://github.com/Nuand/bladeRF/wiki/Getting-Started%3A-Windows. The description of the process is useful if you wanted to change any of the source code.

Windows SDR software: SDR-console V2 beta

The only Windows software I can find that supports bladeRF is SDR-console V2 beta. The current beta doesn’t but I managed to get an earlier version of SDR-console that was designed to work with bladeRF: Build 2.1 Beta 1545, see http://www.ham-radio.ch/kits/sdr-radio.com/2.1/ for all versions. Setup SDR as per any other.

Bandwidth list “only” goes to 5 MHz, but maximum gives 30 MHz (maximum can be set in radio definition). V2 “only” goes to 1 GHz, not sure if setting or otherwise. At the moment, not a problem, but it would be good to go to 23 cm band as UT-1000C can go that far.

Windows 8

As for Windows 8, not sure what the problem is. Main one is non-signed driver, which can be overcome with special start-up mode for Windows 8 (not a criticism, I can see why they do it) It may be a Windows 8 issue, as that is one of the listed issues for the device. It may also be a change in the firmware that is incompatible with the current V2 .dlls (the latest beta lists bladeRF but can’t find it). Will try again in next few days.

I am happy

While everything is still a bit hairy, I am happy.

I have a spectrum analyser for my DVB-T project. Now I can set up the amplifiers without clipping/splatter etc.


Everything else is a bonus. I can only encourage others to buy one of these devices and start experimenting. The company is a start-up and needs all the support it can get for a fantastic device.
 


SDRs at the first IF of IC-7410 TRX as a panadator: More detail

Note: I take no responsibility for any attempts at doing what is described here. I am not an expert with these radios and am following my own interpretation of how this is done. It is shared on the basis of the philosophy of amateur radio.

The main advantage of a 1st IF tap is avoiding the problem of sharing one antenna with a RTX and SDR; not simple.
As mentioned in an earlier post, I have installed a tap into the 1st IF of my IC-7410. The tap is made into the TRX’s first IF (64,455 kHz), above the roofing filters (the key to a good RX). The ICOM even have a socketed test point to do it (some hesitation playing inside a new $2000 TRX! Much reading of circuit diagrams). With the plug (hard to get but only $1.50 and are used on all main brands of RTX), isolation amplifier- http://www.cliftonlaboratories.com/z10000_buffer_amp.htm (gives details of how it is done), preferably through a bandpass filter (obtained but not installed) then to Funcube. I have it running, but not permanently installed.
 
 
A SDR at the first IF is really neat. It can be used as a panadator for the RX, although the SDRs display gets a bit woozy as I tune. However, you can see a large section of the band with the SDR and all the little signals that are swamped just using audio. The SDR can be used indepentently as a rx and tune within the first IF; basically using all the good front end of the RTX that the SDR doesn’t have.
The address for the 1st IF isolation amplifier I used is http://www.cliftonlaboratories.com/z10000_buffer_amp.htm. It gives details for fitting the amplifier for other TRX including FT-1000MP. When purchasing, You need to specify the gain of the amplifier, usually unity, as the signal should be fairly strong once it gets to the first IF. Delivery from the USA to Australia can take 3 weeks, ask how much extra for faster delivery.
Another way is at http://www.w1ghz.org/small_proj/small_proj.htm, about halfway down. It includes a filter to remove the local oscillator, and a further article that uses an opto-isolator to turn the SDR off during transmit. I had been looking for this site as I wanted to add both the filter and switch, having bought the parts to do it. The Mini-Circuits parts are available on special order from http://www.minikits.com.au/, as are many other interesting bits; he makes up patch cables to order too. There is another article on 1st IF taps that uses an in-line Mini-Circuits filter, which is easier to do, but I haven’t been able to find it.
The cheapest SDR to cover the 1st IF is just a RTLSDR dongle from ebay. You can get them on eBay for less than $20. Try to get one from an Australian store, otherwise you will be waiting weeks for delivery. I could only find one this morning: http://www.ebay.com.au/itm/USB-Digital-TV-Radio-for-PC-DVB-T-DAB-FM-SDR-Receiver-Dongle-RTL2832U-/331059172700?pt=AU_Components&hash=item4d14aa155c. Just search eBay for “RTL2832”.
These dongles go down to about 60 MHz, just enough for the 1st IF.
The other one you can use is a Funcube Pro+, but they are from the UK and more expensive, ~AU$200 although are plug and play. I use one of them.
For software there is SDR Console, HDSDR, SDR#, and CuteSDR among others. There is a site devoted to RTLSDRs that cover all of these and more: http://www.rtl-sdr.com/.
For a panadapter, SDR# is probably enough. See http://www.rtl-sdr.com/rtl-sdr-quick-start-guide/. Use the latest version of zadig: http://zadig.akeo.ie/. If using Windows 8, as I do, there is a special start-up mode for installing non-signed drivers. Windows 7 just gives a warning.
It can be tricky getting the RTL dongles driver and associated files installed.

Added 9 Jan 2014

I have snipped parts of the block and circuit diagrams to make it clearer where the 1st IF tap is and how it seems a reasonably safe place to put it.

However, a word of caution to others wanting to modify different TRX for a panadapter. The IC-7410 is a very different design to other radios as much of the circuits for Rx and Tx are not shared. The best place for a panadapter is before the roofing filters if they are used. Unlike older superhet designs, the 7410 generates SSB modulation in the DSP, not through a sharp skirt filter (typically 9 MHz or 455 kHz). As such, taking an IF tap near the main SSB filter of a conventional TRX is very different as the RX and TX paths are shared, unlike what is done in the 7410.

 The tapping point is most clear in the circuit diagram of the 1st Rx mixer and the last Tx mixer, as they are clearly separate, plus there is a test connector to plug into. There is possibly no need to switch the isolating amplifier out when in Tx, but the bandpass filter could still help remove the local oscillator, but it should be minimal as a balanced mixer is used. It may be possible to just run coax out of the TRX and have the isolating amplifier outside.


Similarly, the tapping point can be seen approximately in the block diagram that clearly shows the different Rx and Tx paths. The green path is Rx.


It has been useful to re-visit the 1st IF tap on the IC-7410 to reassure myself that it seems a safe place to do it.

Further, re-reading the Rx and Tx design descriptions and circuits makes it clearer that the IC-7410 is a hybrid software-defined radio (SDR). All the Rx and Tx is done in the firmware programmable digital signal processor (DSP). The analogue parts are mixers, amplifiers and band-pass filters etc. The main signal filtering is done in the DSP as is the modulation/demodulation. Most SDRs could do with the bandpass filtering of the 7410.

Coming back to one of the points of a 1st IF tap is the difficulty of using a SDR and a TRX sharing a single antenna. Reading the circuit diagram, it should be possible to find a tap point where the Rx is separated from the Tx. With that, the SDR and TRX could safely share the same antenna with a splitter and 3 db loss of signal. A job for another day... Comments are welcomed.