Ku-band LNBFs
Marcus D. Leech
Ku-band LNBFs?
They often advertise as "0.1dB Noise FIgure" which is decidedly
preposterous. My thinking is that
they're likely over 1dB.
Eduard Mol
--
--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/e01d2304-8929-48ed-b6ea-906e0df7d854%40gmail.com.
Marcus D. Leech
I have not measured the noise figure of the LNB itself but I have done some rough estimates of the Tsys of my 12GHz setup, which uses a 1 metre satellite dish and one of those Ku band LNBs. It was around 100 K. Assuming a reasonable spillover contribution of 20-40 K and ~10K atmospheric and background noise, I would estimate that the LNB has a noise figure on the order of 50- 70 K. So indeed not even close to the 0.1dB they are claiming but also really not that bad considering the low price of those units.
I was just in a conversation with Ken tapping on the subject, and when he was a young radio astronomer, working up at
Algonquin Radio Observatory, their 10 GHz LNA, with a noise temp of about 270K, cost more than many modest homes
back in the day...
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CAFcm1Q5Q%2BKvrP_%2B%2BV%2B0RsYbYTsQP5%3Dme73sbCc-fPk3Pe7JW9A%40mail.gmail.com.
Marko Cebokli
Many (>20) years ago, I was active on 10GHz EME, where low noise was paramount. I made my own LNAs, and the lowest I could get was 0.8dB NF, with a waveguide design (https://lea.hamradio.si/~s57uuu/emeconf/eme98.htm). The best HEMTs then had a declared minimum NF of 0.35dB. When I looked at datasheets a few years ago, there was not much improvement in ~20 years, the declared minimum NFs were about 0.3dB. Looks like the limits of this technology were reached already about 20 years ago. Note that these datasheet values are "de-embedded" values, you can never reach them in practice.
I also played with commercial LNBs for 12GHz. A WR75 input LNB declared 1dB NF also measured very close. With a feedhorn for 0.35 f/d dish, I tried sky/ground, and got about 5.5dB. Assuming about 30K of antenna temperature for this antenna, this agreed with the declared and measured 1dB NF.
Kuhne later sold a 10GHz LNA he specified at 0.7dB. I was quite skeptical about that. Well, maybe with selected transistors?
Years later I bought some LNBs, marketed as "0.3dB NF". These had integral feedhorns, so I could only do ground/sky, and it wasn't better than 5dB. Considering the narrower pattern (these feedhorns were for offset dishes with f/d >= 0.6), this would indicate a NF slightly higher than 1dB.
The 0.1dB advertised today is total nonsense. As noted above, today's best HEMTs have NFmin of 0.3dB DE-EMBEDDED!
A realistic value is around 1dB or slightly more, considering these are made down to a price. The value printed on the box is determined by the marketing department, not engineers.
Marko Cebokli
2024-10-19 23:35, je Marcus D. Leech napisal
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/37bda1e4-8318-4c7f-8e00-cff101890447%40gmail.com.
Marcus D. Leech
Many (>20) years ago, I was active on 10GHz EME, where low noise was paramount. I made my own LNAs, and the lowest I could get was 0.8dB NF, with a waveguide design (https://lea.hamradio.si/~s57uuu/emeconf/eme98.htm). The best HEMTs then had a declared minimum NF of 0.35dB. When I looked at datasheets a few years ago, there was not much improvement in ~20 years, the declared minimum NFs were about 0.3dB. Looks like the limits of this technology were reached already about 20 years ago. Note that these datasheet values are "de-embedded" values, you can never reach them in practice.
I also played with commercial LNBs for 12GHz. A WR75 input LNB declared 1dB NF also measured very close. With a feedhorn for 0.35 f/d dish, I tried sky/ground, and got about 5.5dB. Assuming about 30K of antenna temperature for this antenna, this agreed with the declared and measured 1dB NF.
Kuhne later sold a 10GHz LNA he specified at 0.7dB. I was quite skeptical about that. Well, maybe with selected transistors?
Years later I bought some LNBs, marketed as "0.3dB NF". These had integral feedhorns, so I could only do ground/sky, and it wasn't better than 5dB. Considering the narrower pattern (these feedhorns were for offset dishes with f/d >= 0.6), this would indicate a NF slightly higher than 1dB.
The 0.1dB advertised today is total nonsense. As noted above, today's best HEMTs have NFmin of 0.3dB DE-EMBEDDED!
A realistic value is around 1dB or slightly more, considering these are made down to a price. The value printed on the box is determined by the marketing department, not engineers.
Marko Cebokli
PLL LNBF. I used to have quite a number of these, and they've "drifted away" in projects that I've lost track of,
and give-aways to various people.
We have a *very* tentative plan to add a X/Ku band feed to our dish, with a possible project to do either an entire
Northern Sky map at about 10.65GHz, or just a map of the galactic plane.
Back in the late 1970s, GB did a sky-map at 4.85GHz with the 300ft telescope, using a technique called
"driving on the meridian" which is a relatively lazy way to build up sky-maps, by just driving in elevation
all day long. Slowly moving elevation up and down between limits, while the sky moves in front of you.
You basically raster-scan the sky.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/a67d0effd719ae9b288efe3b5e640f7a%40hamradio.si.