SGL-wide VLBI Radio Astronomy Network???

[Andy McK]

Hi guys, hmm, this does sound interesting.

I have a question. What sort of band width/ frequency will the array be looking at?

Does it fall within the band width of an LNB for the Knackerd Vauxhall, sorry, astra?

Reason I ask, is.

If you notice your terestrial Tv twig, it has elements evenly spaced apart, this is a product of the frequency you are recieving, eg, the higher the freq, the smaller the gap between the elements

Now if the frequency/bandwidth you/we are trying to recieve is in the micro band, the wave length will be tiny, hence the LNB's and polished ducting, instead of a twig and a length of co-ax.

But, if the frequency/wavelength is in the UHF, it is possible to make very hi gain and very sharp antenne out of broom handles and stripped back co-ax.

have a look at this site, gives you an idea whats involved

http://hamwaves.com/antennas/inductance.html

BTW I like the sound of "The SGL Array"

Andy (What happens if I press this?) McK

[Geesus]

Hey guys... Just a quick message to say that this looks like an absolutely awesome idea, and I'll definitely be keeping a good eye on your progress!

I'd love to help out, but I'm leaving for a gap year soon so I wouldn't really get the chance to see it through .

I don't know if this will be of any help, but I saw these while browsing the "B grade" section on Maplins website, and this thread sprung to mind!

Satellite dish motor. Pretty sure someone mentioned moving dishes with GoTo mounts, but this could be a cheaper alternative?

http://www.maplin.co.uk/satellite-dish-motor-47176

I've done a bit of radio signals in both electronics and physics over the last 2 years at college, and whilst I'm sure there are plenty of members here with far more knowledge than me if there's anything I could do to help out then I'd be glad to give you all a hand!

Hope this all goes to plan!

Ginger

[Treeden]

@badgerchap: Great idea and a very interesting project. I can't offer any relevent expertise I'm afraid, but put me down for an element of the array.

Q: What is the correct term for an individual component of an array? Dish, antenna, element? Anyway, I'd like to be part of this, it sounds really interesting.

I would imagine that just getting this thing up and running will be the main focus of this project, but assuming we are successful, has anyone thought about possible uses for the array?

[garethmob]

soooooo we are all bouncing ideas around on here but no actual concrete stuff

id be more than happy to make a prototype device if some one could sugest really cheap materials

also there is the ittybitty telescope how about basing it on that?

[jamespels]

Gareth,

nothing stopping anybody on here trying something like the itty bitty scope - I may try exactly the same thing to get a feel for radio astronomy - but setups like that are going going to be sensitive enough to contribute to an array. From the article you link to, it is clear that the itty bitty scope has a beam width several times greater than the width of the sun. This table (http://en.wikipedia.org/wiki/Angular_diameter) gives a good representation of the angular sizes of various things in space. For reference, in a dark sky the core of the Andromeda galaxy appears as a fuzzy blob about the size of the sun or moon (1/2 a degree); the full galaxy and extends to about 4 degrees or eight times the size of the moon.

For the radio telescope to give a meaningful output, I would suggest a minimum target resolution of around 0.1 degree (6'), with a medium term goal of 1'. Note: this is revised down in terms of accuracy from my earlier comment of 1' and 1" as short and long-term targets! At 6' resolution, Andromeda would appear as an elliptical grid approx 40 wide x 20 high; at 1' resolution, the grid will expand to 240 x 120 approx. For comparison, the best the itt bitty scope could hope to achieve is about 2 x 1, and it would be difficult to tell the image from the background (this assumes the signal : noise is high enough to detect anything, which I strongly doubt).

J.

[garethmob]

ahhhh i see, is there any way to build upon the ittybitty make the dish larger and the res tighter? i didnt actually releise andromedia was that big i havent seen it in a dark sky naked eye yet

[sologuitarist61]

As for the drift scanning, I think that's just about the most sensible idea we've had so far. Much more simple, plenty effective and shouldn't cost nearly as much. One sticky bit though is we'll have to be darned sure that each time we move up or down a notch, we all do it by precisely the same amount. I think either a preprogrammed movement or some kind of synchronicity between dishes would be the way to get around this, and seeing as they'd all have to be linked up anyway, the latter is probably best. I'm not sure how we'd exact this control though. I think we'd all have to build a similar altitude bearing and actuator, and have it controlled simultaneously and remotely.

Tell you what boys and girls, this ain't harf gonna be complicated. Fun though! Might take a few years to get it right....

I am following this thread with interest though I have little knowledge of such things as I come from a Chemistry background, so excuse any silly remarks made, but isn't drift alignment precisely what they do with the Arecibo dish? Which I believe is not track-able.

Also, for the numpties here without an electronics background, could you please when you use abbreviations give us the full title as well when first used? The 'net unfortunately can come up with seemily hundreds of differing answers to each abbreviation I type in. Sorry to ask this but I feel that there are many like me who would like to get into this project but do not understand the jargon :cheesy:

Ta muchly

[garethmob]

i can see the dishs having a argo navis like "goto" or pushto system

i think the only main problem which we will have to over come isnt the actual hardware

its where all the data is to be accumulated (perhaps it might be a good idea to see if big organisations like Seti or a large university network will be able to help) as it will need massive computer system to gather all the data and compile it together. the other question is the local computer software. i think somthing like the seti screen saver like one will be a good idea, it hides in the background working away when its done, you can send it off to be compiled, the compiled data is then sent back out for using computers processers to do it, and it comes back very much like folding ect it could work very very well but it would need an encreadable programing network to do its thing.

[Vince1963]

Hi Guys... I have a couple of question if anyone would like to answer..... If this project ever gets off the ground what objects will you be looking at/for?.. and also as anyone worked out the problem with the geometric delay involved in VLBA (very large baseline array)?. can anyone tell me how (everyone who as the right equipment) is going to connect this lot together simultaneously and the delay worked out for each individual receiver. I'm not trying to poo poo the idea i just want to understand.

[jamespels]

Hi Vince,

1 - to start off with, the best (i.e. simplest and easiest to coordinate) approach has to be drift scanning - everybody to the same location and let the sky drift by with the Earth's rotation. This means we need to set up accurately in longitude (az) once, then we can leave it alone so we only have to play with latitude (alt). This means that there is no specific target - it would essentially gradually build a map of the sky.

2 - timing is the bulk of the debate in these two threads. Agreed, there is no realistic way to synchronise tracks from different scopes in real time, so the only option we have is to record time in each track - hence the debate about GPS and other potential very precise timing signals. What we need is a very accurate (nanoseconds if possible) "tick" that is the same interval on all recorded tracks. We also need to know when each tick is, but this can be out be a few milliseconds. The trick then is to use the recorded times to align the tracks as closely as possible, then use software to move the tracks back and forth until interference fringes appear (this is a huge ask, btw - depends on signals being equivalent, very low noise - signal ratio, etc. to even detect fringes and then possibly filter out false matches). The very accurate tick spacing is vital to ensure the tracks remain synchronised and that each track represents the same amount of time so that they play at the same speed in relation to each other.

Massively over-simplified and I am sure there are 1,001 different things I am missing so far but that's all part of the fun!

J.

[jamespels]

Sorry for double posting but I thought this was equally relevant to both threads...

Very helpful guide to building a RA antenna here: http://www.britastro.org/radio/projects/Construction_of_an_Amateur_Radio_Astronomy_Antenna.pdf

J.

[Vince1963]

Hi James... Thank you for taking the time to explain things, that as helped me a lot, I thought we/us would be tracking objects, which to my mind would be very difficult to almost impossible for us/we with our budget, i thought we/us were going to look for pulsars, nova and what ever we could pick up with a radio signal ( assuming the equipment will be sensitive enough), I'm sure i will still have problems getting aligned or understanding and interpreting the signal and end up with just signals from some satellite or high altitude aircraft lol. Thanks again Vince

[Islander]

Have a look at MSF Rugby. I know that the whole thing is controlled from a caesium clock but I'm not sure what the resolution of the time signal is. However, if memory serves, the carrier frequency is controlled extremely accurately (1 part in a billion or so) from the same clock so it might be possible to do something with this using the time signal as a sync source.

[wella]

Hello all,

This is all very interesting! Im sure I can be of very little help but please excuse me while I butt in anyway! The only daft questions are the ones you dont ask!

Do all the data sources actually have to be connected live or do you need people to actually record data on something and accurately time stamp it. Does anyone know of any software to provide a very accurate time stamp to a data file?

Then I guess all the data files have to be brought together and interpolated.

Seems to me that as a proof of concept you first need to get two data files that can succesfully be time locked together. These dont need any radio frequency data yet, just to prove the time problem.

Or are computers not the right way to gather this data??

Paul

[rouan]

Hello fellow star gazers, what do I stumble upon from my lounge! Great read and I am humbled by your knowledge and friendly community.

I have the data solution for you. I've got a running prototype running live.

Basic front end, no need to login/signup now. You can see the functionality without that.

http://205.186.165.44

or http://8bodev.com

Open up two browsers at the same time, side by side, or ask a friend to try over the phone.

Theres a horizontal line that is drawn by your browser, when someone visits, or posts data (this will be streaming sensor data with additional meta data combined.} Or chat, which makes a yellow line that scrolls. This data is stored and broadcast in realtime. If you stay in the US you should get 50ms or better, this can be spread to more servers in localised areas relatively easily. So give it a try, refresh the second window a few times, chat and click [say] etc. I've got this system storing data as a continous stream in memory., So it will start to toss old data if we dont update. But 512MB RAM can be useful if we synchronise a test.

You'll also see server status. This has basic status data of the physical machine, as well as link to http://8bodev.com/health

Which gives us parseble JSON data easily for remote machines. This keeps updating for every request and is generated in <1ms. Uptime gives us 0.000000001 s accuracy (probably have to reference this against other sources for slip), any unix machine can output this. Refresh the page over and over to see live time we can sync to. in BLUE.

Uptime is the accurate one

Servertime is UNIX standard timestamp format.

Serverdata is formatted timestamp

[color=#ff8c00]{
  "pid": 21595,
  "memory": 29036544,
[/color][color=#00ffff]  "uptime": 5811866.429727584,[/color][color=#ff8c00]
  "osuptime": "2011-05-24T05:36:08.894Z",
[/color][color=#00ff00]  "servertime": 1356451994919,
  "serverdate": "2012-12-25T16:13:14.919Z",[/color][color=#ff8c00]
  "started": "2012-10-19T09:48:48.489Z"
}[/color]

We can also do efficient raw dump of data. http://205.186.165.44/event gives full memory with individual data point and time. Just need to add something like this. Blue is the reference you get from /health "uptime"

{

"name": "Yourname",

"lat": -26.185788469629863,

"lng": 28.014450073242188,

"datestamp": "2012-10-19T09:49:23.782Z",

"timestamp": 1350640163782,

"uptimeref": 5811866.429727584,

"sensora1" = 564,

"sensora2" = 563,

"sensora3" = 567

}

I've got this working locally using my arduino polling the ADC and streaming over serial to the laptop, which is running a local webserver. I've got it plotting out as well. Looks like I've got my work cut out for JAN/FEB So:

SPACE > ANTENNAH(s) > Sensor/Arduino > USB > PC (linux) > CLI script (nodejs/js) > SOCKET.io > Webserver (8bodev.com) > SOCKET.io (broadcast) > CLI script on your side or browser

Also, will have live streaming graphs on the site. CLI is command line.

Busy on FFT and multiple sensors. I'm just using simple omni antennahs for now, so far i can detect my hand moving closer/further. I need to learn from you how to do directional antennah, but i'm trying to first get the aperture working on simple copper wires. Seems promising, lots of signals to practice on So next step is to scale this up and use lat/long from the packets to place global location in 3D space. so we can measure space drift and signal refraction through the planet. Atleast we'll have realtime signal cross referencing so you'll know what to look for as it happens, and when it does, boom! you've got distance and origin. So I'm setting up a city grid with microcontrollers listening and streaming on different scales. I'll have about 10 sensors spread over 100m^2 and then at different location 10mins drive away another bunch, then in other direction a 30mins away, and 45mins away.

For anyone wanting to play around with this while we get this up:

https://github.com/voodootikigod/node-serialport#readme for interfacing with arduino/circuits

ttp://nodejs.org/ this is what is running the data/server (no apache/sql nonsense)

http://processingjs.org/ for live graphing in your browser. can be from you local dish/antennah/sensor, or from the network (with delay), so we need to get to a point where we have an interface to adjust stepper motors/circuits at any point on the planet in sync. So you can choose a 3D coordinate to listen from, and we combine all data (offset for speed of wave) and see if any thing is emitting there. We do this for a volumetric scan (so we image in x, y, z). Accuracy will be good because we are triangulating with known gps, and time.

Sensor/Microcontroller

http://arduino.cc/

I've got several of these older ones and will be getting the DUE as well.

The Arduino Due is a microcontroller board based on the Atmel SAM3X8E ARM Cortex-M3 CPU (datasheet). It is the first Arduino board based on a 32-bit ARM core microcontroller. It has 54 digital input/output pins (of which 12 can be used as PWM outputs), 12 analog inputs, 4 UARTs (hardware serial ports), a 84 MHz clock, an USB OTG capable connection, 2 DAC (digital to analog), 2 TWI, a power jack, an SPI header, a JTAG header, a reset button and an erase button.

Wow, thats alot of information. I dont know if any of you are on irc, but I'm on irc.freenode.net #arrow (just made it now)

Or email me: rouan@8bo.org

I'll be watching this thread and contributing as much as possible. I'll encourage people to start playing around with your sensors and try to get it into some format that we can get into nodejs. What do the big guys use, what bitdepth/Hz? My idea is to allow streaming of raw sensor data, but also play around with local FFT, perhaps FFT needs to happen server side. But we'll need some mayor help with hosting/servers then

As a side note, with pulses out from nodes we can image surface movement as well, which is cool. Distributed radar.

[rouan]

Hello everyone, see no reply yet. So this will be a bump and update with progress. These graphs are screenshots from the webapp, its quite early in development but you can see if it works on your side.

https://github.com/fluentart/arrow#readme for download and install. It is built with nodejs as linked in above post.

This is all from a few minutes of recording on arduino. Not very great ADC, but I think it is a start. I'm doing my radio licence early next year, so i'll get some decent gear and a dish. Will also get sky tracking rig pretty soon, all diy though.

Heres sample of program recording:

As you can see that is 6 sensors (graphs only plotting A0), with 0.00413sec rate (241.696hz). I think it can be sped up alot! But will do directional sun test within coming days so we'll know if it is working.

Hope theres still interest in this!

Happy new year!

[jam1e1]

Sounds interesting idea, not sure if already mentioned, may be worth pulling together funding, if required and can be agreed, via kickstarter?

[rouan]

jam1e1: I'm open to any collaboration/support, and believe this stack can scale to the level needed to do this properly.

[badgerchap]

Hi guys, sorry for the lack of attention over the last few months! Unfortunately, this seems to have been a bit sidelined - personally I have exams (yuck) but on the plus side at least I'm learning the neccessary Physics. Rouan, can't we just wait until the SKA is built and get you to hook a line up to it ?

If people still have interest for this, I propose putting together a forum for it. I have a website which I could happily host a simple forum for the purposes of gettiing our ideas together and (slowly) getting this thing moving?

[nikolifish]

I came here looking for instructions on how to build my first radio telescope and ended up here. I see this has lost some steam but if it kicks up again I'd like to throw my hat in the ring. It would be my first time with radio astronomy however I can code.