GOES-11 and EMWIN L-Band signal

[swingbozo]

I need an LNA/LNB that can pick up the new EMWIN signal at the current 1690.725 MHz and also the future 1692.7 MHz, then magically "downconvert" this signal to something I can decode using my sound card. There appears to be a lot of hand waving on the EMWIN government site on how this is possible but basically ZERO real practical information on how.

I've got the dish, I need the LNA/LNB and/or downconverter so I can receive EMWIN's L-Band signal. Uhm, where should I start looking?

I've seen the quorum downcoverter at like $2000, and that ain't happening.

swingbozo

[satchick]

Perhaps someone on a ham radio or scanner site would know a little about this sort of thing?


www.radioreference.com has a satellite section and that might be a place to check......

[Terryl]

The 1690.725 MHz and 1692.7 MHz "L" band signals can be received on a police scanner connected to the LNB second "F" connector, (after you find the satellite) however the EMWIN Offset QPSK signals cant be demodulated by the scanner (unless you can afford the AOR) you will need this stuff to get the info.


http://www.wernerlabsinc.com/receiver.htm

[swingbozo]

I'm familiar with the wernerlabs "send us $2000 and we'll set you right up" solution.

I'm not sure how exactly I "connect into the second LNB connector." What exactly does that mean?

Pretend I have a dish and it's pointed correctly, and a quad invacom. Now what exactly do I do?

What I'm trying to do is get the signal into the audible range so I can feed it into a computers sound card and then "the magic" happens. The governments emwin site uses gnu radio (a software defined radio) to do all the magical decoding. That part I get. The part I don't get is from the thing on the dish to my sound card.

Werner labs has a "thing" on the dish. What is that? It's fed into a downconverter that does what exactly? Then fed into their box where "the magic" happens. I need to know more about the dish lna/lnb and downconverter parts.

I've already talked to a ton of EMWIN ham radio geeks. 100% of their responses are "buy the werner labs solution, then everything works." I don't think it's realistic to force emergency managers to cough up $2000 for something available from only one manufacturer.

This can't possibly be that difficult.

I want a sub $200 EMWIN solution. The bulk of the cost of this is the dish, LNA/LNB and downconverter. Everything else can be done in software. I've got a nice dish. I've got a computer with a sound card. How do I get the magical space beams into my computer?

[swingbozo]

http://www.goes-r.gov/hrit_emwin/index.html

This includes source code and block diagrams for "A Flexible Software Based HRIT/EMWIN Prototype Solution for the GOES-R Transition"

The problem is they completely glaze over the first two parts of this. The LNB/LNA and downconverter. Then they toss in a secondary prototype board from avnet that doesn't make this stuff anymore to get the magic from the bird to the PC.

The software bits I totally understand. The magic from the dish to the sound card I do not.

[Terryl]

I'm familiar with the wernerlabs "send us $2000 and we'll set you right up" solution.

I'm not sure how exactly I "connect into the second LNB connector." What exactly does that mean?<<<<<<<<<<< #1

Pretend I have a dish and it's pointed correctly, and a quad invacom. Now what exactly do I do?

What I'm trying to do is get the signal into the audible range so I can feed it into a computers sound card and then "the magic" happens. The governments emwin site uses gnu radio (a software defined radio) to do all the magical decoding. That part I get. The part I don't get is from the thing on the dish to my sound card.

Werner labs has a "thing" on the dish. What is that? It's fed into a downconverter that does what exactly? Then fed into their box where "the magic" happens. I need to know more about the dish lna/lnb and downconverter parts.

I've already talked to a ton of EMWIN ham radio geeks. 100% of their responses are "buy the werner labs solution, then everything works." I don't think it's realistic to force emergency managers to cough up $2000 for something available from only one manufacturer.

This can't possibly be that difficult.

I want a sub $200 EMWIN solution. The bulk of the cost of this is the dish, LNA/LNB and downconverter. Everything else can be done in software. I've got a nice dish. I've got a computer with a sound card. How do I get the magical space beams into my computer?


>>>#1...The 1690 and 1692 MHz RF carriers are on the "L" band down converted signal from the satellite LNB, you can connect a police scanner that can receive that range to the LNB's second linear "F" connector and it will receive the EMWIN signal.......But there are very few (if any) scanners that will decode the "Offset" QPSK signal to do any good.

This is where the audio from the scanners output could be run into the PC's sound card for the software to do its magic.

The big problem is trying to find a scanner that will go to that high of frequency for under your $200 budget, you can look on flea bay or Craigs list to see if you can find a used one.

You can build your own frequency down converter if your handy with electronics.

All you need is an RF amplifier that will work at at 1600 to 1700 MHz, a band pass filter in that range (to get rid of the other transponders) and another amp after that.

Then a good PLL* RF oscillator to use as a LO*, this signal is fed into a frequency mixer along with the signal from the amp, the two signals will mix and produce a third this IF* signal can be any frequency you want.

Then if you have a scanner that will only receive to 900 MHz you can use an LO frequency of lets say 1000 MHz mixed with the signal from the LNB at 1690.725 MHz, this will give you an output IF frequency of 690.725 MHz, this can be received by most scanners that don't cost over $200.


The scanners audio output could then be fed into your sound card.

* PLL = Phase locked loop, a very stable RF oscillator.

* LO = Local Ocillator, used to mix with a second signal to down or up convert it to a secondary signal.

* IF = Intermediate Frequency, the frequency from a mix of two other signals, this signal can be higher or lower then the original input signals.


The only possible problem I see is the fact that they use a QPSK modulation, I don't know if you can find a scanner (or other receiver) that will demodulate this for under $200.

[swingbozo]

thanks, but I still have NO idea how to connect anything to anything. I have a nice dish pointed correctly, and I need what again after this? Pointed to what again?

If I can get signal to my pc soundcard I believe i can swizzle some information out of it.

[kutter]

couldn't you just use a DVBS tuner something like the old Twinhan 102G PCI card ... you can pick them up pretty cheap :)

some of the specifications

Tuner

Input terminal: F-type 75 Ohm
Receiving frequency: 950~2150 MHz tuning range
Input level: -65~-25dBm
Support MCPC and SCPC


QPSK & FEC

Symbol rate: 2~45Msps
QPSK filter: Root-raised cosine filter with roll-off 0.35
FEC: 1/2,2/3,3/4,5/6,7/8 and Auto

[jvvh5897]

Sounds like kutter has the right idea. The standard LNB has no place in the RX of an L band signal directly, as Terryl seems to suggest. The front end of the standard box takes the L band (aprox 1 GHz) from the LNB and converts it directly to "zero IF" baseband--that is basically audio freq if you put in a narrow band filter. There used to be a spectrum analyser project for sat boxes that took the zero IF from the frontend (there is both an I and Q signal and either could be used), stepping the demod frequency, and put in a narrow band low-pass filter (just a resistor and capacitor) then digitized the output with an LED level meter (very very cheap chip--about $.50 US) and gave that input to PC to display signal level off of sat.

That basic idea might work for a project like this. Now, there might be some problems too--a look at
http://www.goes-r.gov/hrit_emwin/Hardware%20Schematic%20Description.pdf shows something that is taking L band and converting to zero IF in two steps and that may be lots more stable than the standard STB frontend. The narrow band filter might not be an easy thing to get working right. The STB tends to set a level that is good for the QPSK demod that follows (and the STB chip ONLY does the DVB form of QPSK) and might not be what you need for your setup. Oh, and the sound card can only deal with frequency up to 20KHZ so the 128KHz bandwidth that seems to be in the specs for the service would not be digitized by sound card correctly. You could look for an old TV card, but you might have to build a driver for your needs. You might be able to digitize the output of the front end some other way though--have to think about that.

[swingbozo]

There does appear to be some overlap with the ham sat guys on this. I'm seeing quite a bit of s-band and l-band wrt pulling sat signals.

Everyone seems to have rather odd feed horn lnb/lna combo's to pull this stuff in. I'm still kinda scratching my head a bit on these things. I just can't believe the ability to pull sat L band requires so much overpriced hardware.

<scratching head>

[jvvh5897]

Shouldn't require lots of $.

For feedhorn and scalers you can read the really good work of Paul Wade N1BWT--he had one project to mod an 18 inch dish for wifi (approx 2.4GHz) so he built something better than cantenna to mount on dish.

Still for 1.7GHz the feedhorn will be about 6 inch in diameter--be a pain to mount--but would only be a tin can shape with an F connector on it and simple probe inside. You might want a cheap LNA to isolate the feedhorn as much as amplify the signal for cable loss.

For something like a spectrum analyser you could use an ATMEL microcontroller with A/D convertor built in, but for high speed data reduction of QPSK folks usually use purpose built devices like in our STBs. If the NOAA folks did not use a standard format there will not be chips for it (QPSK is used for phones and modems too so there are a number of 'standard' formats).

This could be a project I would like to see if could be done inside an STB--say sti5518 processor one since we have good source code for that (see coolsat4000 source--it is modded to support pansat2500 and digiwave7000 so lots of possible boxes availible). would really need a mod to do the QPSK but that might be the only hard part of hardware. Could display using the box connection to TV or dump data out serial port to PC.

[jvvh5897]

BTW, a good search term for use of sound card with weather satellite decoding with sound card is WXSAT.

After reading the info linked above about the "software solution", I see that the easiest part of the signal to decode is the 9600 baud rs232-like one. To do that with a standard sat box, you would use L-band feedhorn on the dish, tap the I or Q signal off of the front end and run that to sound card (or attach a PLL FSK demodulator chip like the old LM565 or that EXAR chip--XR2211--are these chips even availible anymore, hard to say).

Oh, and most of our boxes are only intended to have 1MHz freq setting--but the front end can do far better than that, the coolsat4000 code suggests that 125KHz control is possible. And it might be easier to tap I and Q lines on box like the 301-13 or the DRD430.

[swingbozo]

I snagged that code when it was flying around. I was thinking that also, but more likely I'll code up something for computer use instead of messing with an STB.

I have a line on an older werner labs setup that I'm going to play with. If I can figure out the feed horn LNA/LNB part then I'll blow the dust off of one of my old FTA dishes and play with this feed.

[jvvh5897]

The Paul Wade stuff says that a simple cylindrical feedhorn usually works OK. The general rule of thumb is that the diameter be between .6 and .95 of the wavelength (about .65 seems to be typical for Ku band LNBs and everything pretty much scales with wavelength) and that the length is not very critical but 1 wavelength is usually OK, copper, brass or AL or silver are good materials to use, but a tin can works if that is what you have. For 1.7GHz wavelength is right about 7 inches.

Better results can be had from feed if you use scaler ring(s), but for the receive side odds are you will not have to worry that much.

There are cheap in-line amps for L band that would likely be all you would need. LNBs are to block-convert receive frequency to a lower frequency so that you can use a reciever that you have for the IF--that is why there are LNBs for L band to 130MHz range--ham guys had 130 MHz gear for the polar orbit weather sats.

The sound card is likely only going to be good to receive the APT signal--there is a FAX signal just like you would send over phone line that carries the low resolution photos. From what the info says I'm not all that sure if they are planing on keeping that signal going with the new sats.

[Terryl]

Here is another link to an EMWIN equipment manufacture.


http://www.weathergraphics.com/cyclone/index.htm

I was looking at some info on this system and it looks like there is a TP on 97W that is supposed to carry the ERWIN signal, it is at 12185 (dont see a polarity listed) your standard linear LNB should be able to receive this TP, looking at the frequency involved 12185 - 10750 = 1435 MHz, this may be within range of some higher end scanners.

I would also check to see if you can get a hold of one of the LNA's from the manufacture I mentioned above.

And in my posts #3 and 6 I am in error, those are the TP frequencies not the down converted frequencies from a standard linear LNB, sorry my fault...old age sneaking in...

So it looks like you will need an "L" band LNA, and the reason behind the big cost is the fact that that band has an awful lot of of junk on it and the filtering can cost quit a bit.

[swingbozo]

The ku EMWIN feed is going to be history, if it hasn't been shut off already. Thanks for all the pointers!

[jvvh5897]

Good info guys!

Lets see about prototyping an L band feed. I went looking for a good sized "can" and did not find much made of metal, but I did find a paper oatmeal can, 5 inch in diameter and about 9-1/2 inch long. With a little creative work with aluminum foil and DURO spray adhesive that could be a good place to start. We want something that can be played with a lot and is easy to work with. If your lid is HDPE or LDPE, then they will pass microwaves. Put foil on both inside and out, try to make it as smooth inside as you can, outside can be rough. Use F connector--there are types to make solder connection availible called "bulkhead", but you can use male-male F-81 barrel too. Probe should be exactly 1/4 wavelength from backplane of can if foiled--doesn't really matter where if you don't foil the back.

I find that a 1 by 2 fits in the standard Ku band LNB holders--might want a little bicycle tape on the wood to make fit tighter and to protect surfaces. You could use a U or L shaped support for the oatmeal can. The use of existing LNB support is easy to do but requires that you offset the dish EL upwards if you are mounting the feedhorn above the bracket. You might want to experiment with putting your Ku band LNB on the back plane of the can to see if you can use it to allign your dish (there are C band LNBs that have Ku band LNB mounted in this way). The wood support for the feedhorn should have some gimble to it to allow you to point the thing at center of dish as well as you can.

[swingbozo]

I found this site. It has some great pointers to other sites along with info:



http://www.uhf-satcom.com/
http://www.uhf-satcom.com/lband/


There L-Band setup uses the 13cm/23cm pre-amp from here:


http://www.minikits.com.au/kits2.html

And it looks like most people fab up their own feed horn for these things. Some dish pictures here. Scroll down for the L-Band ones:


http://www.uhf-satcom.com/rx/

[fleckstar]

Hi swingbozo you can pick one up at down east microwave, good and price is right..............

[swingbozo]

I had to dig through their price list to find it:

1691LNAC High IP3, 1691MHz LNA, Weatherproof <0.7dBNF, >16dBG, DC Coax, Type "N" connector. 120.00
1691LNA High IP3, 1691MHz LNA, Weatherproof <0.7dBNF, >16dBG, Feed-Thru, Type "N" connector. 120.00
1691LNAHC High IP3, 1691MHz LNA, Weatherproof <0.7dBNF, >25dBG, DC Coax, Type "N" connector. 150.00
1691LNAH High IP3, 1691MHz LNA, Weatherproof <0.7dBNF, >25dBG, DC Coax, Type "N" connector. 150.00 <- assuming they meant "feed-thru"
1691ULNA PHEMT 1691MHz LNA, <0.4dBNF 15dBG, Weatherproof and type “N” connectors 140.00

These were in their "commercial and non-ham band" products. They appear to be assembled only pre-amps. They also appear to sell "transverters," or up/down converters, but I haven't found the correct model.

[jvvh5897]

Radio Shack has an in-line L band amp for $15US, doesn't say the noise figure, but odds are they are just using a standard microwave part and odds are that it is, by default, low noise. Part is just called satellite amplifier.

BTW, at L band an 18 inch dish has 16 dB gain and 1.5meter dish has 26 dB.

I thought it might be of help to see what a Ku band LNBF looks like in cross section so I drew up the attached--close as I could use a meta paint program and measure. As the Ku band wavelength is about 1 inch, the inch dimensions could be directly interpreted as fraction of wavelength.

[Terryl]

OK Swing, I found a place that sells the "L" band LNA down-converters, you wont like the price but maybe you can now look at some of the industrial and lab equipment scrap yards on the web or maybe even fleaBay for the model number they use, you might get lucky.


http://www.qcom.com/datasheets/prices.pdf

Look at page #2 the 99R0011203, 204. 205, 206 part numbers, these are the LNA and frequency down converters for the EMWIN signals.

[Terryl]

I had to dig through their price list to find it:

1691LNAC High IP3, 1691MHz LNA, Weatherproof <0.7dBNF, >16dBG, DC Coax, Type "N" connector. 120.00
1691LNA High IP3, 1691MHz LNA, Weatherproof <0.7dBNF, >16dBG, Feed-Thru, Type "N" connector. 120.00
1691LNAHC High IP3, 1691MHz LNA, Weatherproof <0.7dBNF, >25dBG, DC Coax, Type "N" connector. 150.00
1691LNAH High IP3, 1691MHz LNA, Weatherproof <0.7dBNF, >25dBG, DC Coax, Type "N" connector. 150.00 <- assuming they meant "feed-thru"
1691ULNA PHEMT 1691MHz LNA, <0.4dBNF 15dBG, Weatherproof and type “N” connectors 140.00

These were in their "commercial and non-ham band" products. They appear to be assembled only pre-amps. They also appear to sell "transverters," or up/down converters, but I haven't found the correct model.

Well that's a start, too bad the LNA's skip over the frequency band you need.

You could use that amp with a good yagi antenna (or can antenna) cut to the correct frequency, then point that at the center of an old "C" band 10 footer, this may get you the gain needed to get a good usable signal.

[swingbozo]

The GOES EMWIN freq is right around 1691 Mhz, I thought?

I pulled the feed horn off of the "commercial" emwin setup I have access to. I don't want to pull apart the LNA this thing has on it. No doubt I'll create a leak and blow the thing up. If in the end it doesn't work I can open it up and not feel bad.

They attached the LNA directly to the can with the feed probe coming right out of the side feed of the LNA into the can. I took some quick measurements:

The outside diameter of the can is 12.8cm. The length of the can is 17cm. The probe is 7.4cm from the bottom of the can, and the probe is sticking 3.4cm into the can. The can itself is approximately 2mm thick, made out of aluminum.

It appears the reason this wasn't set back up again was because the person that "moved" it dropped the thing and bent up the feed horn mounts. I'm in the process of straightening them out and getting the dish setup again. The dish itself is a pansat 90cm center feed with a custom non-intrusive mount.

This system is an older EMWIN WX-13 with dish/feed/LNA setup from here:


http://www.big-z.com/wx13.asp

I'm going to point it to GOES-11 and get the older signal till they kill that one off.

[swingbozo]

Here's some pics of the dish I'm working on

[Terryl]

It dosent look that bad, but is that rust inside the can on the first photo?

If it is you had better stop it now befor it causes a signal drop.

Rust doesnt reflect the signal as good as nice clean stuff, use some good silicone spray and wipe it out good and clean.

[Terryl]

The 1691 amp is 250 KHZ above the lower frequency you need, if the amp will go that far down then its OK.

But if thats the bottom end and its tight it may not, you will have to contact them and ask.

[Terryl]

Here is another source for amps, and this one may work.


http://minicircuits.com/cgi-bin/modelsearch?search_type=model&model=ZRL-2400LN&tb_no=

[swingbozo]

Unfortunately not rust, it's a dent from when the dish was dropped.

[swingbozo]

Just to reiterate, the new EMWIN signal off of either GOES bird is at 1692.7 MHz

[jvvh5897]

General rule of thumb, defects in the surface of a dish smaller than 1/10 of the wavelength have little effect on gain of dish. Paul Wade info says defects at 12GHz should be less than 1mm, but defects up to 1cm will only cause 1dB loss of gain. Since the 1.7GHz signal is 7 times longer wavelength, defects 7 times the prev should have little effect.

[pupp1977]

these two sites might help just type the www. and paste link
members.inode.at/576265/index.html
prapro.com/nubes.htm

[jvvh5897]

There might be a good chip to use for this type of stuff too--the dsPIC33 looks to be a microcontroller with DSP features--ADC of 1.1Msps is good enough for the LRIT signal and I think I saw an application note about getting 2.2Msps out of the chip which would be good enough for HRIT. With UART, I2C and SSP I/O and even a microstick that can do development using USB for $25US. THere is a set of C source for it to do FFT--maybe combine with GNU radio stuff...? Might be other better chips out there (or will be by the time HRIT gets active)

[Terryl]

Have you seen this place?


http://comblock.com/product_list.html

They have an L band receiver.

This is the 3003.

8556

[swingbozo]

Thanks for the new pointers!

If I can get the signal into the PC around sound card levels then I can do all the FFT stuff in software on the computer itself, which is what the GOES/EMWIN site was suggesting. I straightened out the dish and need to put some time in pointing it to see if the old setup is still working.

[swingbozo]

Stumbled across another place:


http://www.rfbayinc.com/LNA/LNA-1620.pdf

[jvvh5897]

I came across a homework project by a student that applies to this thread IMO. It also has the schematic for a sat finder. Called "Radio Astronomy at the backyard" by Eduardo Manuel Alvarez, it uses a 12 foot dish to attempt to do solar obs at 1.4GHz. So pretty close to the 1.7GHz signals from GOES sats.