Modified C-band LNBF

[Marcus D. Leech]

Here's a pic of the mods I've made to a Titanium Satellite C1W-PLL LIte
C-band LNBF:

http://www.sbrac.org/files/c_band_mods.jpg

The device is based on the RDA3570 high-integration C-band downconverter
chip, which takes a garden-variety 25MHz reference clock to drive the
integer-N PLL, which operates at a fixed frequency of 5150MHz (that
seems standard with C-band LNBFs). Mixing is done inside the chip.
Externally
there's an LNA on each of the input probes (H and V, or L and R with
polarizer), and then a microstrip filter.

These boards are fairly "airy" compared to the Ku-band equivalent. I'm
not super happy with the grounding arrangement for the incoming 25Mhz coax,
but it's only 25MHz. The next one I do, I'll perhaps do it slightly
differently.

[Don Latham]

Lovely! If all are driven with the same 25 MHz reference, how good do you
think the delta phase differences of the lo's will maintain--they probably
will not be zero?
Don

Marcus D. Leech

> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to the Google Groups
> "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sara-list+...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>
>


--
"If you don't know what it is,
don't poke it."
Ghost in the Shell
-------------------------------
"Noli sinere nothos te opprimere"

Dr. Don Latham, AJ7LL
Six Mile Systems LLC, 17850 Six Mile Road
Huson, MT, 59846
mailing address: POBox 404
Frenchtown MT 59834-0404

VOX 406-626-4304
CEL 406-241-5093
Skype: buffler2
www.lightningforensics.com
www.sixmilesystems.com

[Marcus D. Leech]

On 07/30/2015 07:56 PM, Don Latham wrote:
> Lovely! If all are driven with the same 25 MHz reference, how good do you
> think the delta phase differences of the lo's will maintain--they probably
> will not be zero?
> Don

They're integer-N synthesizers. So assuming you're careful about
line-lengths providing the 25MHz clock, it should be pretty good.

On my 2.8GHz MMDS downconverters, with an 80s-era integer-N synthesizer,
they maintained exactly-identical phase (at least to the resolution of
the oscilloscope) essentially forever.

[Marcus D. Leech]

On 07/30/2015 08:30 PM, Walter Clark wrote:

Marcus,

I think I've asked you this before about any use of your [SARA] group email letters in the SBMS newsletter.  This letter seems very interesting to me and I'd like to put it up with the picture you provided. As you may know there is a ham band near five centimeters.
But of what interest to SARA types is 5 cm?

These are C-band LNBFs, intended for the 3.4GHz to 4.2Ghz satellite band.   Scientifically, it's "just another" continuum band. As long as one avoids
  the Clarke belt (where all the satellites are), you can use the satellite band for continuum interferometry, and Indiana U has been doing C-band
  continuum observations with their 4.7m dish for a very long time.

The modification you did was to introduce an external LO? How stable does it have to be for your use?

Not an external *LO*, strictly speaking, but an external *reference* that the downconverter chip uses to drive its synthesizer.  Two synthesizers that
  use the same reference will have mutual phase coherence.  If they are integer-N synthesizers, or have other special features, then they also have mutual
  zero phase *offset*.  The former is absolutely necessary for interferometry, the latter, a "very nice to have".  The common clock provided should
  ideally be derived from at least a TCXO, but for simple experiments, even a simple crystal oscillator will suffice.  In interferometry, at least at the
  amateur scale, it's not the absolute phase-noise and stability that matters, but rather, the *mutual* phase noise and stability.  You could have a
  fairly "dirty" reference clock, and as long as both sides are seeing the same "dirt", it won't matter much.

Walt

[spectrahm]

Some time ago I read this paper regarding some experiment at 4GHz: http://www.ukaranet.org.uk/projects/Radio_Astronomy_Experiments_at_4_GHz.pdf
There is apparently no phase coherence in their setup, the two LNAs just feed a mixer, which is puzzling to me.

Something else puzzled me; the C-band is way too high for Bremsstrahlung or synchrotron radiation, yet it seems quite low for blackbody radiation. Apparently they get some radiation from cold clouds of gas weakly heated by nearby stars. Is there anything else of interest to observe at those frequencies?

[Marcus D. Leech]

On 07/31/2015 12:41 PM, spectrahm wrote:

Some time ago I read this paper regarding some experiment at 4GHz: http://www.ukaranet.org.uk/projects/Radio_Astronomy_Experiments_at_4_GHz.pdf
There is apparently no phase coherence in their setup, the two LNAs just feed a mixer, which is puzzling to me.

Yes, that's Ken Tapping's older C-band interferometry setup.  There *IS* phase coherence.  The front-end is just LNAs--no downconversion involved at
  all.  And once you realize that a mixer is just an analog multiplier (single-lag correlator), then it all makes sense.  He and I set up a USRP-based version
  of the same thing a year ago, and he's busy taking data with it. 

Something else puzzled me; the C-band is way too high for Bremsstrahlung or synchrotron radiation, yet it seems quite low for blackbody radiation. Apparently they get some radiation from cold clouds of gas weakly heated by nearby stars. Is there anything else of interest to observe at those frequencies?

You'll be able to observe many of the same sources as at lower frequencies, but at lower brightness.   There's no sharp cutoff--the universe is awash
  in electromagnetic radiation at all wavelengths.

[spectrahm]

Yes, that's Ken Tapping's older C-band interferometry setup.  There *IS* phase coherence.  The front-end is just LNAs--no downconversion involved at
all.  And once you realize that a mixer is just an analog multiplier (single-lag correlator), then it all makes sense.  He and I set up a USRP-based version
of the same thing a year ago, and he's busy taking data with it.

Ah, indeed, no intermediate frequency and local oscillator to deal with here. I still thought it was required for the mixer, I stand corrected.

You'll be able to observe many of the same sources as at lower frequencies, but at lower brightness.   There's no sharp cutoff--the universe is awash
in electromagnetic radiation at all wavelengths.

So, some low energy light from very cold black bodies. But then, why the C-band instead of the Ku? Surely the brightness would be stronger and the angular resolution better?

I thought I would try the Ku band at some point since it could theoretically make the most of my offset dish, but so far I've found the HI line and the VHF band more interesting. I keep an eye on anything new in amateur radio-astronomy though, hence my interest in C-band.

How is your Ku-band interferometer getting along by the way? Did you receive the missing parts?

[Marcus D. Leech]

On 07/31/2015 02:32 PM, spectrahm wrote:

Yes, that's Ken Tapping's older C-band interferometry setup.  There *IS* phase coherence.  The front-end is just LNAs--no downconversion involved at
all.  And once you realize that a mixer is just an analog multiplier (single-lag correlator), then it all makes sense.  He and I set up a USRP-based version
of the same thing a year ago, and he's busy taking data with it.

Ah, indeed, no intermediate frequency and local oscillator to deal with here. I still thought it was required for the mixer, I stand corrected.

You'll be able to observe many of the same sources as at lower frequencies, but at lower brightness.   There's no sharp cutoff--the universe is awash
in electromagnetic radiation at all wavelengths.

So, some low energy light from very cold black bodies. But then, why the C-band instead of the Ku? Surely the brightness would be stronger and the angular resolution better?

Consider that even for a small dish (1.6m in Ken's case), getting both absolute and mutual pointing to under 0.1 beamwidth is a significant challenge,
  doing so at Ku-band is even worse.  That's one of the "traps" one falls into in contemplating shorter wavelengths--pointing accuracy even for a single
  dish is hard, and then getting the mutual pointing to adequate levels for an interferometer is even harder.

I thought I would try the Ku band at some point since it could theoretically make the most of my offset dish, but so far I've found the HI line and the VHF band more interesting. I keep an eye on anything new in amateur radio-astronomy though, hence my interest in C-band.

How is your Ku-band interferometer getting along by the way? Did you receive the missing parts?

I was able to get solar fringes with my small, 0.55m dishes.  And then I decided to "improve" my clock delivery to my LNBFs, and broke them both in
  the process.  They're spectacularly delicate, both due to the tiny circuit sizes, and the teflon-based substrate.  I have another pair of them on the
  way from Penticton, where the local electronics guy graciously offered to modify another pair for me...

[spectrahm]

Le vendredi 31 juillet 2015 22:43:56 UTC+2, Marcus a écrit :

Consider that even for a small dish (1.6m in Ken's case), getting both absolute and mutual pointing to under 0.1 beamwidth is a significant challenge,
doing so at Ku-band is even worse.  That's one of the "traps" one falls into in contemplating shorter wavelengths--pointing accuracy even for a single
dish is hard, and then getting the mutual pointing to adequate levels for an interferometer is even harder.

Good point, especially with offset dishes. I have yet to find a reliable way to determine where mine points at, even though with its 120cm aperture its beam is about 15° at 21cm. Some people use some reflective tape laid in a cross shape but that only works for the Sun. If the dish is static and the Sun never crosses it (my case), it can't be used to calibrate its coordinates.

I was able to get solar fringes with my small, 0.55m dishes.  And then I decided to "improve" my clock delivery to my LNBFs, and broke them both in
the process.  They're spectacularly delicate, both due to the tiny circuit sizes, and the teflon-based substrate.  I have another pair of them on the
way from Penticton, where the local electronics guy graciously offered to modify another pair for me...

Good luck with that. Let us know if you manage to detect anything fainter.  Wolfgang Herrmann had some surprisingly good results at these frequencies, even resolving M17. His dishes were a bit larger though (2x130cm).

Regards,
Damien

[Dave]

Slightly off-topic; I bought a 1.2m dish this morning, and am wondering if those of you with solid dishes drill a drain hole in them so they don't become a bird bath when pointed at the zenith?  I'd also be interested to know how much attenuation a cover causes.

[Marcus D. Leech]

On 08/01/2015 05:33 PM, Dave wrote:

Slightly off-topic; I bought a 1.2m dish this morning, and am wondering if those of you with solid dishes drill a drain hole in them so they don't become a bird bath when pointed at the zenith?  I'd also be interested to know how much attenuation a cover causes.

I few small drain-holes is a good idea, and I did that with my 0.9m offset, for precisely the same reason.

I wouldn't expect an official cover to cause much attenuation.  Although with radio astronomy, one usually doesn't mount the dish in an out-of-the-way
  area, so you can always manually clear snow off it.

On Saturday, August 1, 2015 at 7:05:18 AM UTC-4, spectrahm wrote:

Le vendredi 31 juillet 2015 22:43:56 UTC+2, Marcus a écrit :

Consider that even for a small dish (1.6m in Ken's case), getting both absolute and mutual pointing to under 0.1 beamwidth is a significant challenge,
doing so at Ku-band is even worse.  That's one of the "traps" one falls into in contemplating shorter wavelengths--pointing accuracy even for a single
dish is hard, and then getting the mutual pointing to adequate levels for an interferometer is even harder.

Good point, especially with offset dishes. I have yet to find a reliable way to determine where mine points at, even though with its 120cm aperture its beam is about 15° at 21cm. Some people use some reflective tape laid in a cross shape but that only works for the Sun. If the dish is static and the Sun never crosses it (my case), it can't be used to calibrate its coordinates.

I was able to get solar fringes with my small, 0.55m dishes.  And then I decided to "improve" my clock delivery to my LNBFs, and broke them both in
the process.  They're spectacularly delicate, both due to the tiny circuit sizes, and the teflon-based substrate.  I have another pair of them on the
way from Penticton, where the local electronics guy graciously offered to modify another pair for me...

Good luck with that. Let us know if you manage to detect anything fainter.  Wolfgang Herrmann had some surprisingly good results at these frequencies, even resolving M17. His dishes were a bit larger though (2x130cm).

Regards,
Damien

--

[David Ocame]

I put a couple of very small holes along my elevation path. This helps with rain. Snow and ice are a different story.

 

This message was NOT sent from my Blackberry
***********************************
*****I'm not dead, yet!**********
***********************************

Dave Ocame, WS1ETI

Stony Creek Observatory
FN31og
-72.834 longitude
41.272 latitude

Member: The SETILeague, Inc. (Awards Chair), Civic Orchestra of New Haven, and
The Connecticut Symphonic Winds

[James Abshier]

Dave,

You didn't say what frequency you were going to use the dish for.  At Ku band there can be substantial attenuation from water on the cover when it rains.

Jim Abshier

On 08/01/2015 05:33 PM, Dave wrote:

--

[Dave]

Good point, I didn't think of that. I've got some Ku LNBFs, the seller gave me a C-band LNBF, and I'm watching eBay for a Ka-band LNBF.  I'll have to tarp or no tarp accordingly.  BTW, the tarp is not official.  The Cad Tire near me had their low-end polyethelene tarps on sale for $3, so I plan to use that.

 I had to try it out ASAP, so I've got a homemade LFA 3-element yagi for 611 MHz stuck in the feed holder at the moment.

[Dave]

I meant to mention that, other than up, I have no idea where it's pointing.  I'll Google for instruction, but if anyone has some tips, please share.


On Saturday, August 1, 2015 at 8:38:01 PM UTC-4, Dave wrote:

Good point,

[Marcus D. Leech]

Those offset dishes usually have an offset angle of between 22 and 25 degrees.  So, when pointed horizontal, it's actually "seeing" 22 to 25 degrees
  higher up than its physical orientation would suggest.

The manufacturer usually provides offset-angle information.

It's useful to get them with a cheap LNBF and the appropriate mounting hardware, just so you can see where you feed plane is.

I've found these to be useful:

http://www.satsig.net/22-deg-offset-dish.htm

http://www.satsig.net/pointing/finding-dish-offset-angle.htm

[James Abshier]

Dave,

You might try pointing the dish at the Sun with the detected output indicated on a meter.  When the antenna beam is centered on the Sun (as indicated by maximum output), see if any part of the feed casts a shadow on the dish.  Most small Ku band dishes that I have used show a small shadow of the edge of the feed when centered on the Sun.  A line from the shadow edge on the dish to the feed edge producing the shadow then points in the direction of boresight.

Jim Abshier

On 08/01/2015 08:39 PM, Dave wrote:

[Dwight Shackelford]

Regarding the reflective tape, that sounds like a good idea.  I can see how it would be easy enough to place the vertical strip down the center of the dish, but how do you determine where to put the horizontal strip?  Do you site from the stock LNB mount, and then chop down the LNB mount after?

--

[spectrahm]

Usually about a third above the bottom side of the dish, where the feedhorn points at.

Here's Marcus dish, if you want to have a look:
http://www.sbrac.org/files/dish_front.jpg

Here's also some help to find your offset angle:
http://www.satsig.net/pointing/finding-dish-offset-angle.htm

Mine is advertised at 22°, but I found 23° with the tool. I guess my measurements aren't accurate enough, but this shouldn't be troublesome given the huge beam width at 21cm (15° in my case).