My Small Radio Telescope (STC)

[bojan]

Last weekend I obtained the dish on ebay - C-band mesh for satellite reception, for $160 (just a 2.3m dish, no electronics.. still a good buy because it is easy to assemble the contraption in a backyard).
Then I stumbled on this website:
http://www.haystack.mit.edu/edu/unde...srt/index.html (http://www.haystack.mit.edu/edu/undergrad/srt/index.html)
This is almost exactly what I had in mind.. I am not the first one here ;-)

Currently I am still building the LNA and BPF filter
For the receiver and recorder I will use SDR# (http://sdrsharp.com/) and SkyPipe.. I tried it in a dry run, and it works OK (RAW signal, my audio card passes DC signal to recorder it seems..)
Anyway, I am hoping to at least detect the 21cm line, and it seems it will be possible to even do some Doppler measurements (I have high stability frequency reference, so it won't be hard to calibrate the TV stick properly).

Today I simply couldn't resist to preliminary assemble the dish..
First, I thought I will be able to mount it on EQ6.. but it's too cumbersome and too heavy (13kg of what's on the image below, plus I would need a counterweight.. too much).

Feeder is made from one of my wife's unused pots (7" dia, 3" deep - ideal!)
Antenna is helix, 2 turns of 4mm wide copper tape, wound and glued on 68mm dia pharmacy box (krill oil, I think.. on the image near connector (SMA) you can see the matching section, 22mm 75ohm cable plus 4nH of wires - measured return loss is >10dB at 1420MHz).

All in all, getting there...

EDIT: after playing with RFSim99 a bit, I (theoretically) managed to improve the matching (pic attached). It's much more broadband with additional 75-ohm shorted lamda/4 section in parallel to connector.

[atkinsonr]

That's really cool!

I hope it's not a silly question, but what can you do with it once it's operational?

[bojan]

I'll be taking pretty graphs with it ;) !

Now, seriously, there are many things that can be done with this receiver.
Sun observation is only one little thing.. galactic plane observations is what I wish to do. And Magellanic clouds detection.
It is also pure technical challenge!

[Woffy]

I am in awe Bojan. Keep us posted, I am interested in your technical journey!

[bojan]

Work continues today...
20 minutes, and the pole is in the place (60cm deep, 1.5 bags of fast setting concrete).

Another hour or a bit more, and the dish is in place.
Now, the electronics..

And, yes, most likely I will need some anchoring, to prevent damage in case of stronger wind.
I will do it later..

[Astro_Bot]

Oh, come on now, we all know you just want to hack into everyone's WiFi from the comfort of home! Say, you don't happen to live close to a foreign embassy by any chance .... :P

Seriously, though, that is really cool. I used to have an old MW/SW radio as a kid and tune into various terrestrial signals. I always wondered if it'd be possible for an amateur to do the same with signals from space.

[icytailmark]

you must get free foxtel :P

[bojan]

I am looking forward to that myself.. I am pretty sure it will not be too hard (especially to detect Sun activity).
As I mentioned earlier, this time around I will start with single atom Hydrogen line.. maybe later, if I don't loose interest in the meantime, I will try with OH (18cm) maser emissions...



That was the official justification for SWMBO, as alternative - if the primary project fails :D

[robz]

Hey bojan!..............if you pick up the sound of mechanical grinding and thumping at regular set intervals, send me a P.M. and I'll give you Jodie Foster's phone number!:rofl:

I am most impressed mate..........keep us informed!

Cheers,
Rob.

[seeker372011]

In awe, (of your expertise) and hoping it all actually works.

Would be grand if it does.

[bojan]

last Tuesday, I tried the antenna on Sun, but no joy..
After I checked the whole thing once again (cables, matching, feeder.. everything seemed to be in order) the conclusion is: 1.4GHz signal from Sun is still below the noise floor of the receiver (Elonics tuner, <4dB NF).

Well, since the company I currently work for as contractor decided to offer me an extension, I decided to celebrate this happy event with preamplifier from Minicircuits (http://www.minikits.com.au/ZX60-P162LN) . It will arrive on Friday, so weekend wil be busy again (I hope.. in principle, I could easily do it through the clouds, provided I had setting circles on the antenna, which I don't.. yet.)

[Barrykgerdes]

Hi Bojan

Join the club. I had a dish that had a beam width (3db) of 10 degrees and a gain of 24db.
A 30db low noise preamp and a HP spectrum analyzer with a noise floor of -117 dbm. I could not see anything from the sun with this set up.

I had no bother with the 3rd harmonic (1427Mhz) of a little hand held CB radio at about 50 metres. (Did not test at any greater range). I would have thought this was enough to see the sun but apparently it is not. I have not done anything since as the mechanical work became too heavy for me.

Barry

[bojan]

Hi Barry,
Yes, it is curious.. because according to this website (http://astro.u-strasbg.fr/~koppen/10GHz/basics.html), it should be easy(er).
Well, my antenna has 28.7dB at this wavelength (2.3 metres dia), so 20dB amp (at 0.5dB NF) should be a bit better than what you had at hand...
I will keep you posted..

[bojan]

Update:
After unsuccessful attempt at Sun with just antenna and SDR (and spectrum analyser), I purchased LNA from Mini Kits (http://www.minikits.com.au/components/mini-circuits/mini-circuits-low-noise-amplifiers), 20dB gain.

Also,I made interdigital hairpin filter some time ago, the PCB arrived yesterday.
After some adjustment, I obtained 8Mz BW (3dB), 1dB ripple... pretty good.
However, the insertion loss is quite high, 10dB.. so this filter can be used, but after LNA. And there will have to be another line amplifier.. 16dB or so.
In the meantime, I started building the "proper" interdigital filter, I expect much lower losses with this version (<0.5dB? hopefully..)
Anyway.. getting there, but somewhat slowly..
The saga continues..

[Astro_Bot]

Good to see it progressing. :thumbsup:

[bojan]

Update:
LNA and line amplifier (36dB total gain) are mounted, ready for test.
I just need a clear sky, so I can aim the antenna at Sun (I still do not have motors for positioning and tracking (one day, Sound Stepper System will take care of that, but first I need to detect something, before I go any further))

Unfortunately, the filter, as it is now, could not be tuned on 1.42GHz, because the resonators are a bit too long (1mm.. currently, it is tuned ~25MHz lower, and I am still waiting for SMA connectors from ebay).

[bojan]

After some time, I am back to the project :-)
Filter is tuned on the frequency, but I am still not happy with response - it is not flat (ripple is +/-1.5dB, and the attenuation is 3.5dB.. too much for placing this filter in front of the LNA.. so it will have to go after.
Getting there..

[Kunama]

Now you just need to mount the dish on an EQ6 for tracking the Sun.....

[bojan]

Hi Matt,
I was contemplating this.. but it is too heavy for EQ6 (actually there are two counterweights required - one to balance the dish + feeder on DEC axle, and then another one to balance the whole thing around RA axle .. 60-100kg or more it total.).
At the end, I will be using alt-azimuthal mount, with Bartels or SoundStepper software to control it.

[Astro_Bot]

That's some fairly professional looking gear. :thumbsup:

You'll have to update your user description (from "amateur"). ;)

[mswhin63]

Fantastic project Bojan, I have always been interested in building a dish.

As a Radio Technician I wanted to take that step as well, and lo and behold 4 weeks ago found a dish up for auction and bought it for $10.00 but fees. :thumbsup:

I have removed the aluminium mesh panel and the frame though is in excellent condition.

Thanks for the information on the LNA, typical Astronomy LNA are the most expensive :P

One thing as part of the auction I received a heavy duty mounting bracket and the linear motor for adjusting.

I cant really start it yet as the woodwork shop requires my attention along with my rental observatory, only 1 month to go before starting new semester at Uni so I think my time is wearing thin. :(

I am curious though your feed mount is very low, my calculation have place it fair bit higher. I am using the Parabolic Calculator V2 but I only made rough measurements so far, I hope to make a horn antenna as well.

Good luck with yours not sure if I can offer suggestions, I am at the same point in understanding as you.

[leon]

Excellent work indeed, hope it all works for you, I'm impressed.

Leon

[Miaplacidus]

Play cricket in it.

[bojan]

Success.
Finally today, I found some time to try detect the radiation from Sun at 1420MHz.
The proof is here: there are clearly two tracks visible (2dB difference), the top one was obtained with antenna pointed at Sun, the lower one is when antenna was pointed away from the Sun.
The tracks were obtained with Max Hold for 30 seconds, for both cases.
The noise floor of the pre-amplifier is still significantly lower than that of Spectrum analyser (there is no difference in noise level on the screen between preamp ON and OFF), so I can add more gain to the system (so far it is ~36dB).

With SDR (Terratec USB receiver) the signal is clearly visible:
The lowest part of the waterfall display is with Pre-amplifiers OFF, the middle part is Sun, and the top part is away from Sun.
The other screen shot from SDR window is simulated transit (antenna beam was moved across the Sun).

The system works!
This is only the beginning..

[mswhin63]

Good on ya Bojan, great you are having success.

One thing I think might help is a down-converter on the antenna. I think the down-converter could provide better wave-guide transmission down the cable to the receiver. For me it is just a thought, also maybe the RTL SDR will have a lower noise floor at the lower frequencies too.

[bojan]

Yep, downconverter may be useful for higher frequencies, but the cable I have (RG6) is pretty good for 1.4GHz.. so for now it is OK.
What I need now is a bit more gain after preamplifier, and recorder.

[SkyViking]

That is so cool! Great looking radio telescope you've put together there. I remember trying to build this when I was a teenager. I didn't get very far though.
How about Jupiter? It's supposedly easily detectable with shortwave radio at 10-25MHz. There is some good info here: http://www.spaceacademy.net.au/spacelab/projects/jovrad/jovrad.htm

[mswhin63]

Yes I understand. This is where the cost can really sky-rocket too. Have you measured with a signal generator of some sort the actual RF input measurement of the RTL. The indicator is -60-70dbm according to the RTL SDR. Curious how accurate this is.

[bojan]

Yep, the signal is visible at -72dBm (SA), 2dB above noise floor (there is cal factor for this situation.. so my signal will be ~-75dBm). I still have to calibrate the whole thing..
RTL was so-so.. I had to calibrate it using tracking gen, since I do not have SG that can reach 1.4GHz.

The cable has 0.3dB/metre... that is 4.2dB total (14m).

[mswhin63]

It is what I thought when I first tested my RTL, I think with a good quality SDR the reception should be greatly improved, but that is where it get mighty expensive though. Good luck with it all, looking forward to see how it goes.

[bojan]

So far it was $15 for RTL, and ~$100 for preamplifier (20dB)..
Line amplifier is my own (12dB gain), but I need more total gain, so there will be another $100 out of pocket..
The filter I made earlier will be placed between them.

[bojan]

For SDR reading of -30dB (noise floor is at ~-60dB), I had -61.3dBm CW at input (75ohms).. since I do not have SG that goes that high frequency-wise, I used tracking generator in zero span mode (then I can reach 1.6GHz), with external attenuator to supply the test signal level at SDR input.

On the screen plot it is visible how the recorder (SkyPipe, detector set to peak) has shown the drop in audio level when carrier drifted out of audio frequency range (gray area on spectrum plot), set by sample rate.
So, cal factor for my case is -31.3dB.

Which means the SDR level reading is in dBuW..

With additional 40dB in front od SDR (pre-amplifier) the noise floor will be at ~-130dBm (to be confirmed.. because there will be some noise from preamp as well).. which is pretty good receiver all in all..

[mswhin63]

Yes that is the way to go, using the SDR as a middle amplifier stage.

Still as you mention there could be some additional noise in the earlier amplification but -130dbm will be a great achievement for sure.

Best to reduce the bandwidth as much as possible should get a much better S/N ration as well.

I was going to consider a PCB filter for the higher frequencies not sure if 1.4Ghz is too low for a strip-line (thats what I called it years ago) filtering system. Maybe that is for another frequency detection.

Frequency shifting for spectral analysis should be easier to obtain at radio frequencies.

[bojan]

Malcolm,
I'm afraid the bandwidth can't be reduced in our case - the received signal is also noise-like and broadband, so the only way to increase S/N is reducing the losses in front of LNA.. that means larger dish, optimized illumination of the feeder, best possible feeder (I am trying with this helix design, however I am not entirely sure how good it is, so this will have to be checked as well at some stage.. and whenever we are dealing with polarized signals, we have that 3dB loss), impedance matching to LNA and shortest possible cabling in between).
And a lots of signal averarging/integration (appropriate software should be providing this..)
The biggest problem here is experimenting - it takes a long time between changing something and seeing the results of the change, so this work requires a lot of experience and confidence in what has been done already..

[mswhin63]

I understand, good luck with what you have and will look at some updates. Although I have an RTL SDR with me I may be considering a change in direction with a different SDR. It will be a while before I can get started with so many other things happening.

[bojan]

On the graph below I captured half of the Sun's transit through the antenna lobe.
The antenna was pointed at Sun (using the feeder shadow), the Earth rotation did the rest..
The plot is actually the presentation of the the audio signal, coming from audio card.. not ideal method, what I really need is to somehow couple the output from SDR with chart recorder..
The noise bursts visible on the plot are system sounds and random noise, probably from environment.. Couple of magpies trying to nest in my antenna also caused some of them.

[mswhin63]

I can utilise the waterfall graphing better than Skypipe. There is a bit too much noise on the signal to get a reliable reading.

Have you though about auto tracking as part of the project?

[bojan]

Yes, I have.. It will be fully motorised one day, soon.
Waterfall graph is useless in my opinion because there is no numerical data there, only presentation..
Since the SkyPipe saved the data in file, I will process that eventually and "flatten" the noise.. But even this is not right method, I really need direct output from ADC within USB RTL to SkyPipe..

I was thinking of looking into recorded and saved data file from SDR# (in PCM format).. Still a lot of preparation work to do before the system is really up and running..

[mswhin63]

I understand there is a difference here, I was considering the aspect of determining the signal for technical purposes. I don't know how to apply the signal as yet for science purposes but that would be another thing to consider for me after making my system.

I was meaning the spectral analysis as well. The waterfall makes it visually easy to see a signal in amongst the background noise, although your signal is higher than previous.

[bojan]

Yes you are right, waterfall is actually spectrum vs time, and it's very easy to spot features, especially transients.
However, I am not sure this can be recorded with current version of SDR#.. and the data record is essential.
Another problem with recording raw data is the file size..
Since the SDR sample rate is huge, so would be the data file as well..
Maybe SDR# is not the ideal solution after all.. for spectrum analysis that is.
I tried to use my HP 8569B Spectrum analyser for this purpose, but it has limitations as well.. one of them is relatively high internal noise floor, so I need more (variable) gain in front of it (40dB or so).
So for this to do I will have to go for suitable down converter.
Tere are some mixers and VCO's stored in my boxes somewhere..

I am now considering building the AM detector for RF signal, and my own ADC, which can be controlled by SkyPipe.

[bojan]

Just realized that I have couple of AD8310 and AD8314 logamps in my stock :-)
Since I have the theoretical noise level of -130dBm @ 20kHz BW, the output level from preamp stage is ~-90dBm.. and the input range of AD8310 is from -80dB m (@ 50 ohms) to -17dBm..
So I need ~20dB additional gain.
I will also need a down-converter, because AD8310 fr cutoff is at 400MHz.

And ADC.. or some win app that measures DC voltage at MIC input of the audio card.

This configuration will be usefull as radiometer - the precise measurement of radio flux.
However, no spectrum analisys..

[bojan]

After some time, I am back to my radio-astronomy project..

Today I put together two amplifiers (total gain 60dB, NF= 0.5dB) and tried with Sun again.. no problems, signal was >5dB above the system noise floor.

Then I reduced the BW on my Spectrum analyser and tried with antenna pointed at zenith (where Milky Way was at that moment).

Results are summarised on the attached picture:


On the screen, two tracks are visible.
One is with spurious signal at -70dBm (internal spur, harmonic of reference frequency), another (in the centre of the screen) is clearly coming from outside (it disappeared when I switched off the power to pre-amplifiers).
After one hour, when Milky Way was not in the antenna field of view, this signal was also gradually gone.. visible on the screen as well, on the second track (there was some frequency shift during that time, this Spectrum Analyser is drifting a bit - I should have switched it ON much earlier, to allow more time to stabilise).

I am trying now to confirm what I saw.. I hope it's not some radio-amateur experimenting on same frequency..

[bojan]

I think I managed to convince myself I am actually observing the Milky Way at 21cm

The attached image confirms the statement.. The marked "bump" is Milky way, the other track was taken with antenna away (at zenith).
The sharp peaks are external.. the middle one is from my test generator set at 1.4204GHz, the others are of unknown origin.. but definitely from Earth.

In the course of the observation (2 hours, max hold), I noticed the peak frequency of the bump (marked) was changing slightly (lowering..).

All this is increasing my confidence in the setup.. I will be able to prove to myself that Milky Way is rotating ;)

[tonybarry]

Hi Bojan,

A very interesting result. I think the baseline raise around your marked peak is a good indication that what you are seeing is a source composed of lots of little sources. This is some indication that it might be extrasolar.

What is the BW of your SA ? From the header I read 50kHz span, which makes the marked source 1420.3125 MHz, with a span of about 8kHz across the shoulders.

With the galactic rotational velocity of ~100km/sec, we would expect that your detector would pick up the galaxy's hydrogen line anywhere between 1419.8MHz and 1420.7MHz depending on whether the bit you are pointing at is approaching us or receding from us (or not doing either). So what you show is consistent with the 21cm hydrogen line, with a bit of spread due to you sampling stuff around an area rather than a single star. The effective field of view of your radio telescope would be a good thing to know.

Caveat:- I am not a radio astronomer :-)

Keep us posted !!

Regards,
Tony Barry
WSAAG

[mswhin63]

Well done, glad it is still working out for you, definitely an improvement with the extra gain. :thumbsup:

[bojan]

This is result of the second pass of the Milky Way (from last night, antenna was pointing upwards).

I am still not entirely sure about accuracy of frequency calibration (1.4204 is middle peak on the test track), generated by HP8922M, set to 710.2MHz.. so on the screen is first harmonic), but the two peaks of the 21cm signal are very obvious.

Guys, our Galaxy is still rotating! Confirmed :P !

[bojan]

Guys,
I am trying to write a simple (Borland Turbo) pascal routine, which should read ADC values starting at specified time, with specified intervals between readings, for 24 sideral hours.

Before I went too far into that task (I am not a good programmer..), I would like to ask a question:
Does anybody knows of such source code?
It doesn't have to be pascal, anything else will do, as long as I have easy access to LPT port (on which my ADC is attached).

I know that SkyPipe supports ADC via LPT port but I do not have that particular ADC at hand.. and it has to be freeware.
Thank you in advance!

[tonybarry]

Hi Bojan,

Do you have any more information about the ADC ? How it is interfaced ? The control codes needed ? How many bits, sampling rate, S&H triggering etc ?

Regards,
Tony Barry
WSAAG

[bojan]

Currently I am thinking about PCF8591 (I2C bus), as I have it at hand.
And I already wrote the code for it (years ago), but I do not have a decent code for timer.
Actually, I worked out something today.. I can start (reading, or whatever) every second or every minute or every hour (by reading system clock via "GetTime" procedure).. but I still need to be able to program the start time, and more flexibility with interval.

I was also thinking about Attiny85 stick
http://www.ebay.com.au/itm/Digispark-ATTINY85-General-Micro-USB-Development-Board-For-Arduino-New-BY-/221619647153?pt=AU_Gadgets&hash=item33998f72b1
http://www.ebay.com.au/itm/1PCS-Digispark-Kickstarter-Attiny85-Mini-USB-Micro-Development-Board-for-Arduino-/271710155759?pt=AU_B_I_Electrical_T est_Equipment&hash=item3f432ff3ef
(both of them on their way).

[SteveInNZ]

Since you are using I2C, you could use something like the PCF85063A real-time-clock which has an alarm function. You set the alarm time and it pops up a pin. You can either do the sidereal to solar calculation or, since the osc uses the crystal in series resonance, pull it off frequency to run the clock at sidereal rate. The chip will keep time for years on a hearing aid battery.

Excellent project BTW.

Steve.

[bojan]

Hi Steve,
Good idea (PCF85063A), sideral time is something I definitely need as time-stamp, as the conversion into RA is straightforward.

Thank you for the kind comments :)

[bojan]

After a lot of fiddling with various options, I decided the best approach to record data would be to control the HP8569B spectrum analyser via HPIB bus from built-in HP8922 IBASIC controller (I am using this machine as part of the system, because of its very stable and accurate time base).
However, I am having difficulties with programming this beast...
I am using terminal program running on PC (via RS232), and this is very uncomfortable, since there is no text editing capabilities in IBASIC controller.. so I have to re-type the code lines in full every time there is a syntax (or whatever) error encountered, and the IBASIC manual for 8922 is not written for beginners (which I am, after 20+ years not using it).
Is there anyone on the forum that might help me with this?
Thanks in advance!

EDIT:
I found that string variable accepts only 18 characters by default.
So, to proceed, I had to insert the following statement:

5 DIM A$[1923]
String variable A$ can now accept the full length of trace data from SA, and I can go on with developing routine for data collection in numerical form.

Then, next stage was a small IBASIC program:

5 DIM A$[1923]
6 INTEGER Ta(480)
10 OUTPUT 709;"TA"
15 ENTER 709;A$
20 PRINT "Trace transfer from 8569 finished"
30 PRINT "Printing numerical results.. "
50 FOR I=0 TO 480
60 J=4*I+1
70 Ta(I)=VAL(A$[J,J+2])
75 PRINT Ta(I)
80 NEXT I
200 END

The first test spectrum from SA (pilot tone at 1420MHz) in numerical form is attached below.

Watch this space..

[SteveInNZ]

Use your favorite text editor on the PC and then do an ASCII transfer of the file with suitable (found by trial and error) inter-character and inter-line delays. Even the god-awful hyperterminal can do that.

PS. Thanks for nostalgic flash-back of writing data analysis programs on an HP-85. :-)

Steve.

[Astrogeo]

G'day Bojan, for info here's a different approach if you're open to alternative data collection/logging options?

I'm using a USB TV dongle with an alternative to skypipe (RTLSDR Scanner (http://eartoearoak.com/software/rtlsdr-scanner)) which can be run from command line. I've found this pretty flexible and have it triggered using VBScript on a schedule from Windows task scheduler. Output goes to a csv file. It does generate a lot of data, but I read it into XL and reduce it with some VBA code.

Any use?
Tony

[bojan]

Hi Tony,
Yes of course, I will try it, than you.

[mswhin63]

Nice one Tony, been looking for a decent RTL specan software for a while.

I have an L Band Up Down Converter I am looking at using as it has up to 45db gain. Place a LNA and should start something

[bojan]

Observation from early this morning, Crux region (with test signal superimposed).
Signal strength is celibrated to the antenna output.
I still have to figure out the antenna efficiency, to be able to express the vertical axis in terms of flux density.
Residual BW was 10kHz, digital averaging, sample detector.

[mswhin63]

Interesting signal capture, If that is from your spectrum analyser than you have a red shifted signal. Although that will depend on the calibration of the injected source signal though.

I think you may have more of a handle on it than me though.

[bojan]

Malcolm,
I have both red shifted and blue shifted signals :-)
On this particular spectrum, the 21cm line is pretty broad, spread across almost 500kHz BW, indicating velocities between ~60mh/s receeding and ~20km/s aproaching.
I am wandering now why on Earth I didn't start this or similar project earlier.. beinng RF design engineer :shrug:

Probably because I had too much electronics as part of my everyday work..
Also, vintage (but GOOD) equipment is much more affordable these days..
HP8922 that I am using as time base and IBASIC controller was US$25k 20years ago.. and I've got it for $100 on internal auction in the company I worked for. Noone wanted it because it was too heavy and too specialised (GSM testset).. so I said to myself, what the heck... another box in my cubboard won't be a problem, and it may be used for soemthing one day :D

[mswhin63]

I have to 100% agree with you, hence the reason why I have started to look into 1.4GHz as well. My only problem is I am studying to further my existing qualifications as a Radio Technician doing and Electronics and Communication Degree and have really limited time available. I am also battling severe sleep apnoea too and have been temporarily prohibit driving and staying awake grrrr. At least I have my first appointment this week to get it resolved. My health frustrations as starting to get me babbling sorry.

This semester start next week and will start 3rd year course in Electromagnetics. Well in my line of interest.

I am really hankering for test gear and one of my greatest ambitions is a decent spectrum analyser but even though they are cheaper, they are still very expensive on a students income and my wifes disability support pension.

I also have to deal with my wife not understanding what I need. She looks at my current gears and quizzes why? Her disability makes it difficult to explain to her :( I wish my AUSLAN (Australian Sign Language) was a level 4 which doesn't exist in Australia.

[Astrogeo]

Hey Malcolm can I ask what type the LNA is? Is it a bought one, or homebrew? Sounds meaty with 45db Gain. I've been thinking about how to improve that component of my rig and have done some searching but not sure of the best approach since I'm scanning a broad freq range (25-88MHz) ta

[mswhin63]

This is not an LNA, instead it is a amplifier with a downconverter more of a mid stage amplification and converter stage. It still requires a Low Noise Amplifier in front to give it its initial sensitivity.

As it is an intermediate restage amplifier it can have multiple stages to provide 45 db of gain.

I picked to up at an auction. The only unfortunate part is that the impedance is 75 ohms, but it should be rectified with a very simple impedance matching network to provide me with the 50 ohms that I'd like.

[bojan]

Very simple matching cicuit I am using is 1/12 wave circuit, described here:
https://www.cv.nrao.edu/~demerson/twelfth/twelfth.htm
It is not broadbad, but loss is very low (0.5dB in my case)

This circuit is also used to power my LNA (that is why those additional coils and caps are there).
I am also inserting the pilot signal there (via side SMA connector, open) with -33dB coupling, for frequency check.
The reason for all this is, 75-ohms cable (RG6) is very low loss and good quality.. and available at Bunnings.
50-ohms cable (RG58) is much more expensive and not so good at those frequencies (1.4GHz).

I cut this PCB with scalpel...

[mswhin63]

Thanks for the info, things are going to get a little bit busy at the start of semester so I may need to put off this project again until the next holiday break.

I am not sure if I will end up using the up down converter, I thought it was a good alternative to use that and a much lower frequency spectrum analyser like the RTL system.

This is just an easy and quick way to get started with limited funds. Still enjoying your efforts and watching this post very closely.

[bojan]

Initial scan started, using listing as follows below..
Hopefully tomorrow I will have a first scanned band, 7.5° wide, at DEC -70°.

Program listing:
10 DIM A$[1923]
20 INTEGER Ta(480)
30 Interval=300!Interval in seconds
40 Test_duration=24!Observation time in hours
50 Test_duration=Test_duration*60*60
60 Test_end_time=TIMEDATE+Test_duratio n+Interval
80 LOOP! start of the loop
90 Int_end_time=TIMEDATE+Interval
100 Curr_time=TIMEDATE
110 PRINT "Current time = ";Curr_time
120 PRINT Date;Time
130 OUTPUT 709;"TA"
140 ENTER 709;A$
150 PRINT "Trace transfer finished"
160 PRINT "Time = ";Curr_time!Timestamp
170 FOR I=0 TO 480
180 J=4*I+1
190 Ta(I)=VAL(A$[J,J+2])
200 PRINT Ta(I); !Printing track to terminal via RS232 in the same line
210 NEXT I
220 IF Curr_time>Test_end_time THEN GOTO 290
230 REPEAT
240 Curr_time=TIMEDATE
250 UNTIL Curr_time>Int_end_time
260 PRINT
270 PRINT "Interval Loop finished"
280 END LOOP
290 PRINT "END TEST"
300 END

[bojan]

A problem occured in midnight:

It seems that HP8922 internal clock went bacwards 24 hours.. thus disrupting the scan.

Does anybody know anything about this strange behaviour of the RTC ?

In manual I found reference to internal clock:

TIME
Description Sets/queries the TIME of day for the instruments clock.
Syntax CONFigure:TIME?
CONFigure:TIME <real> [:INUM]
Options Refer to Appendix A.
Format = HH.MM in 24 Hour format.

DATE
Description Sets/queries the current DATE for the internal clock
Syntax CONFigure: DATE?
CONFigure: DATE <integer> [:INUM]
Options Refer to Appendix A.
Format = yymmdd
But, whatever I type as command, all I am getting is syntax erro message.. very frustrating..

[bojan]

Problem solved with the following statements, specific to HP8922

120 OUTPUT 814;"CONF:DATE?"
130 ENTER 814;D$
140 OUTPUT 814;"CONF:TIME?"
150 ENTER 814;T$
160 PRINT "System Date & Time",D$,T$
170 PRINT "TIMEDATE = ";Curr_time

All this was actually in the manual.. but it was not cleat to me until give it some more thought.
Now I have time stamp. The format of the returned value is strange, but consistent.

System Date & Time +22715 +1.95300000E+001

This means:
The date is Feb, 27, 2015
Time is 19:53.
:screwy:

[mswhin63]

:lol: US Date system and no colon implemented in the ASCII character set.

Well done, you have done really well and I am so happy to see red and blue shifting in you data. :thumbsup::thumbsup:

[bojan]

Work continues..
But I have a lot of RP *Radio-Pollution)

I thought noone is permitted to transmit on this frequency..

[bojan]

This signal I am getting from the direction indicated by red circle

[mswhin63]

Nice stuff, is that further red shifted than before?

[bojan]

Yep..
But I think this is associated with MW.
There is also a signal from LMC earlier, it is in the collected data but I can't find it now.

[mswhin63]

still a very good result well done :thumbsup:

[bojan]

I wanted to share with you guys the forst results of my work, just one scan, actually a band of sky 7.5° wide at DEC -53°.
On this graph, the horizontal axis is velocity, vertical axis is signal level at antenna output, and "series" is time, or AR.

All features parallel with X or Y axis are external or internal interference..
only that red, random-like blob at the roughly middle of the graph is actually Milky Way signal.

I have a huge work on my plate to do to sort this out properly.. but I am really excited about it.
I am aware that this is a far cry from a "real" scientific work.. but still it gives me a taste of what professionals could be doing and what this is all about (at least I think so.. :P).
Anyway.. my journey into amateur radio astronomy has just begun ..

[mswhin63]

Really nice! Definitely things are speeding up at your end. :thumbsup::thumbsup:

[bojan]

Yep..
Data are pouring in now, and I am wandering how to process all that properly. Mainly data reduction.. removal of known external interference.. and finally, presentation of the information.
There is a lot of info on this here:
http://www.haystack.mit.edu/edu/undergrad/srt

However, I do not want just to follow somebody else's example...

[Steve_Withnell]

Hi Bojan,

Saw your note on the problems you are having. I've built a few of these now and have some observations. (Mine are 1275MHz centre frequency). If you use the Butterworth routines, then the bandwidth tends to be fine. The tapping points I have have found to be a bit too low. The rods seem to me to be slightly too short, as the tuning screws (I use 6mm screws for 6mm rods) are a long way into the case, when the design does not recognise the need for screws. The two characteristics that strongly drive passband attenuation are bandwidth and case depth.

By the time I add in 0.2dB for connector losses (not included in the design tool), then the passband attenuation is not far off.

I have discovered "Dishals tuning method", this is great for these filters, but you do need to be able to measure reflected phase. All you do is terminate the filter with a 50ohm load, wind in all the screws to short the resonators. Then using a VNA or similar, back off the first screw and watch for the phase flipping from -ve to +ve and lock off the screw at 0deg at your centre frequency. Move to the second screw and back that off until the phase is at 90degrees. This is an inverter, so the input impedance should be 0 + J50. If all is well you move onto the 3rd screw and back that off until it is at 180 degrees, then the input impedance will be 50 + J0 and the resonators are all bang on the desired resonant frequency. What is likely to happen, is that the input impedance will be off, but you will have a good guide on how much adjustment is necessary.

The issue I have, is on my current filter, I cannot get the last rod adjusted to resonance, once I've sorted what causes that, then I'll have good grip of these things.

[bojan]

Hi Steve,
Welcome to the IIS forum and thank you for your comments... yes I have access to Agilent E5061B here at work, and will certainly try your method at first occasion
My conclusion after playing with this filter (4 resonators) at home (I have SA with tracking gen and reflectormeter bridge) was coupling was too high

In the meantime I build another one (3 resonators, 10MHz, see here (http://www.iceinspace.com.au/forum/showthread.php?t=148874&highlight=version+2.0)).
Interestingly, this one has too low coupling, and losses are also too high.
I even used copper sticky tape inside but the result is still not good enough for my taste.
Next I will try with thicker rods, they are easy to replace.

[mswhin63]

Which is the most up to date thread on this?

Malcolm

[bojan]

The other one (http://www.iceinspace.com.au/forum/showthread.php?t=148874&highlight=version+2.0) (version 2) :-)