Just Sharing This Invacom 0.3dB vs. Octagon PLL vs. SMW PLL Ku LNBs

[s-band]

Provoked by John's thread Just Sharing This - 1.2 Precision / Octagon lnbf mod I got around to fitting an Octagon PLL LNB to an Invacom C120 flange. The id of the Octagon is within 0.2mm of the id of the IRTE feed but I think there's an extra error of the same order due the LNB not being quite concentric with the flange. It is only force fitted at the moment and I may try to re-do it.

I did comparisons on 1W which is probably not the best option due to the 0.5 degree spread but the dish was not moved between tests carried out at the same time. Two tests (SMW vs. Invacom and Invacom vs. Octagon) were done at different times. At about 14dB SNR, QPSK, a 2dB change in SNR is equivalent to about 2.8dB in RF level using the TBS6983.

The table gives the averages for all transponders. It masks the fact that, on some TPs, the Octagon was ~2dB down on the Invacom. The SMW and Invacom were more consistent. XML files for the Octagon & Invacom are attached as .txt.

SNR Average
4/6/16
Skew peaked at mid frequency, horizontal
Invacom H (same time as SMW) 14.35dB
SMW H (with above)           14.83dB

31/7/16
Skew peaked on lowest frequency Horizontal only
Invacom V & H     14.28dB
Octagon V & H     13.57dB
Invacom H         14.60dB
Octagon H         14.00dB

Concluding, the SMW 1dB max NF was a bit better than the Invacom 0.3dB which was 0 to 2dB better than the butchered Octagon 0.1dB.

SMW LNA and IRTE Feed https://www.satellites.co.uk/forums/attachments/smw_lna-bdc_dsc1067-jpg.94158/

 

[John]

Provoked by John's thread Just Sharing This - 1.2 Precision / Octagon lnbf mod I got around to fitting an Octagon PLL LNB to an Invacom C120 flange. The id of the Octagon is within 0.2mm of the id of the IRTE feed but I think there's an extra error of the same order due the LNB not being quite concentric with the flange. It is only force fitted at the moment and I may try to re-do it.

I did comparisons on 1W which is probably not the best option due to the 0.5 degree spread but the dish was not moved between tests carried out at the same time. Two tests (SMW vs. Invacom and Invacom vs. Octagon) were done at different times. At about 14dB SNR, QPSK, a 2dB change in SNR is equivalent to about 2.8dB in RF level using the TBS6983.

The table gives the averages for all transponders. It masks the fact that, on some TPs, the Octagon was ~2dB down on the Invacom. The SMW and Invacom were more consistent. XML files for the Octagon & Invacom are attached as .txt.

SNR Average
4/6/16
Skew peaked at mid frequency, horizontal
Invacom H (same time as SMW) 14.35dB
SMW H (with above)           14.83dB

31/7/16
Skew peaked on lowest frequency Horizontal only
Invacom V & H     14.28dB
Octagon V & H     13.57dB
Invacom H         14.60dB
Octagon H         14.00dB

Concluding, the SMW 1dB max NF was a bit better than the Invacom 0.3dB which was 0 to 2dB better than the butchered Octagon 0.1dB.

SMW LNA and IRTE Feed https://www.satellites.co.uk/forums/attachments/smw_lna-bdc_dsc1067-jpg.94158/

View attachment 95815 View attachment 95816

........................................................................................................................................................................................................................................................................................................................................
Nice job & report SB. Human nature drives you to evaluate all possibilities until ones curious nature is satisfied-- for better or worse lol.
Iv'e got some C120 Invacom stuff somewhere and an old SMW single polarity job so think i'll try and find time to see what they will do. To be honest, i'm a bit of a fan of single polarity devices .... these modern lnbf things are very convenient i suppose but are a bit of a compromise .
** Anything else to report ?

 

[s-band]

More comparisons. I decided to check the noise figure of several LNAs and LNBs. Measurements were done comparing the noise from a “hot” source (a piece of microwave absorber) with noise from the quietest bit of sky I could find. Hot noise minus cold noise (sky) gives the Y factor. From that noise figure & noise temperature can be calculated. There’s a good summary here: https://gm8bjf.joomla.com/images/pdf/noise_figure.pdf

It’s quite hard to eliminate the possibility of ground noise sneaking in on the side lobes of the feeds especially as the garden is surrounded by trees. That will tend to make the devices seem worse than they really are. I didn’t have time or enthusiasm to check on more than one frequency. All will vary across the bands. Most tests were done at an IF of 1.2GHz. In the case of the LNAs at 3.9GHz for the Dexcel and 10.95GHz for the SMW. The figures below are as measured and calculated, there's a fair probability of errors but I think they give a reasonable idea of reality.



Observations:
Norsat - probably meets its spec as it is easy to pick up another 20 deg K on side lobes.
Dexcel - Not far off its 120K, can be forgiven as it is 33 years old.
TIT C1W PLL - Hmm, 15K vs 82K. Not much difference between V & H. Even assuming some side lobe pick up it’s much worse than the Norsat.
HDSKY - Unknown history, came with the 1.8m CM. Good enough for Sky, I guess.
Octagon OTLSO - Hmm again, 0.1dB vs 1dB not much to say except there wasn’t a measurable difference between the C120 modified one and an unmodified unit.
SMW 1dB Max - Meets spec
Unknown Ka - I guess 2dB+ is to be expected from a surplus 5yr old Ka.