1 - yes
2 - possibly, but not sure yet
See you all!
Hi Mike, G0MJW - my dish is a 60 cm TRIAX with f/D 0.6. My measurements does not show that the POTY is better without the dielectric lens - on the contrary, as others have reported. When I assemble my next POTY, I'll measure the lens effect more closely.
My Y-factor measurement was done on POTY both with and without dielectry lens. I make these measurement on the LNB+feed combination alone - so no dish involved. Ideally the Y-factor should not change from narrowing the beamwidth, as long a the LNB sees either cold sky or hot ground within the actual beamwith. I believe the small reduction in Y-factor with the lens is due to losses in the dielectric material. (theoretically could also be due to changes to side-/backlobes)
The POTY also have a 10 GHz "side". I have measured LNB performance with the original LNB feedhorn and with the POTY feed on the LNB - after cutting off the original horn.
First measurement is Y-factor (difference in cold sky/hot ground).
Original feed: 4.9 dB
POTY (without dielectric lens): 3.7 dB
POTY with dielectric lens: 3.3 dB
Second measurement is transponder noisefloor (60 cm offset dish)
Original feed: 2.9 dB
POTY with dielectric lens: 1.8 dB
Noise measurements are done with averaging, so accuracy is about 0.1 dB. Both Y-factor and transponder noise floor involves other parameters than just LNB NF or antenna gain. So there is no direct one-to-one dB relation to the POTY performance.
The degradation in performance with the POTY is measureable, but any dual band antenna is a compromise. This is compensated by only having to use one antenna.
Anyone else with similar observations?
Rain attenuation in dB/km can be seen below. From the graph it is clear that 2.4 GHz is not much affected by rain. (except when water gets into the feed itself and ruins the matching).
For 10 GHz it can be a problem, but then it has to pour down. There are lot of information on Sat-TV and rain attenuation on the internet. One report stating that: "For Western Europe rain attenuation only excees a couple of dB in 0.1% of time".
The transmission line - copper track - transforms the capacitive impeadance of the tuning flap. A quater wavelength away and the impeadance is inductive.
So if you want inductive impeadance, you place a capacitance one quater wavelength away.
ich würde mir über "driftenden" LNB erstmal nicht zu viele Sorgen machen, Du bist ja auch "Doppler-Shift" erfahrener OSCAR-User... Ich würde mir eher Sorgen um Deine 60cm Schüssel (für den Empfang) machen, die ist eigentlich unterhalb der Empfehlung für unsere Region.. da musst Du schon wirklich alles optimal ausrichten und auch einen performanten LNB haben... aber solange Du das Transponderrauschen noch siehst ist natürlich alles gut! Ein größerer Empfangs-Spiegel macht dann auch überhaupt keinen Sinn, da sich das Signal-to-Noise Verhältnis der Nutzersignale nicht verändert..
Dear Peter, DB2OS
I know it's a bit out of topic regarding "Umbau Octagon", but...
When you write that "as long as you see transpondernoise all is well" and "a bigger dish makes no difference to S/N ratio" this is not quite true
I have seen this statement in other postings as well, but seeing the transponder noise floor, does not mean reception cannot be improved.
In case of a 70 cm dish we see a typical transponder noise floor of around 3 dB - this is compared to the noise level outside the transponder bandwith. Actually what we see, is that the transponder noise - and the background noise within the transponder bandwith - are adding up and the sum of the two is 3 dB stronger than the background alone (or transponder alone if that could be measured).
A bigger dish will receive more transponder noise, but the same amount of background noise (assuming identical LNB and antenna noise temperature). The background noise becomes less important and the S/N ratio increases. With a really big antenna, the transponder noise dominates over background noise and we can receive the satellite with up to 3 dB better S+N/N than in the "3 dB noise floor" example above.
The noise coming from the transponder and from the background (LNB + galactic) are unrelated, so the noise power adds up in the transponder bandwith. This could also be shown on a spread sheet.
A 3dB noise floor using a minimum size dish is in my opinion a good compromise between small setup and reduction of S+N/N ratio.
I bought a GLOBO LSP04H from local market.
It has a 25 MHz XTAL with a 3566 PLL
I plan to use an external GPS lock from a NEO7 GPS.
Is it importand to this PLL chip which x-XTAL pin I will use for external clocking?
I have seen OZ2OE mods but in my case the XTAL traces are below the PCB and I am not sure which 3566 pin goes to each XTAL pin.
73 de George SV1BDS
If you look at the xtal terminals with an oscilloscope - before removing the xtal of course - it is easy to see which is input and output of the onchip oscillator. The terminal having lower amplitude and cleaner signal is the input. Where you can inject external LO.
Here is a schematic. I have changed the design from a Butterworth ( picture in my previous posting) to a Chebyshev - the schematic below. Both works fine, but Chebyshev has slightly better cutoff characteristics. I have also changed the attenuator to allow for two receivers. Both designs have 20 dB attenuation. Component tolerances are quite uncritical. Coil diameter is inner diameter.
This have no relevance for QO-100, but I have set the split frequency at 110 MHz. The lowpass will pass any frequency from DC to 70 MHz and highpass will pass signals from 1 - 2 GHz down to 144 MHz. So this LP/HP filter can also be used for feeding reference signal to a mast mounted microwave transverter with 144 MHz IF.
I'm also using single output LNB's - Octagon and other types - modified for LO injection through the downlead coax cable. I was puzzeled by the amount of LO power needed for the PLL to lock. In my case something like +3 dBm to + 7 dBm. It turns out that with single output LNBs, where the 25 MHz shares F-connector with DC supply voltage, the LO signal is loaded down by a capacitor - 0.1 to 0.33 uF - placed at the input terminal of the 78L06 regulator. Only a meander line RF-choke prevents a total short. This circuit is used on all LNB I have seen.
My solution was to put a 10 ohm SMD resistor in series with the meander line. It reduces the loading effect of the 0.1uF? capacitor and increases LO injection sensitivity by 15 dB. So now only -10 dBm is necessary. It also gives a 1 V voltage drop, but with 12 V supply to a 78L06 this is not a problem.
The LNB is not an Octagon, but a cheap 7 Euro PLL type bought locally.
Diplex filter in the shack to separate IF, 25 MHz and 12 V DC is homemade. Just a few components in a metal box.
Both beacon about +7 dB S/N in 2,5 kHz BW. This is considerably lower than most SSB signals during the last two days. We will have to see if it stays that way?
Really enjoyed yesterday!
RX using 85 cm offset dish with Octagon OSLO LNB modified with remote LO - 25 MHz OCXO - inserted in the shack through a simple diplexer. RTL dongle plus HDSDR. Transponder noise level 11 dB above "background".
TX using 1 m prime focus and linear feed. Power is 30 W through 20 meter coax - 6 dB attenuation - giving 7,5 W at the feed.
With this setup my own signal is 13 dB S/N in 2,5 kHz BW on the downlink. I can copy SSB signals down to 4-5 dB S/N. A few excessive strong signals are 20 dB S/N or more, but most in the 10 - 15 dB S/N range.
As a test I have received - seen on the waterfall - my own carrier down to a level of 5 mW (measured in the shack). Thats 1,25 mW at the feed! But that is just few dB above noise in 2,9 Hz BW.
73 Ole OZ2OE
Listening to my own transmissions I have estimated latency on an RTL dongle using HDSDR to around 600 ms. Playing around with sample rates and buffer size doesn't seem to change this. Any idea on how to reduce latency on this type of RX??
Here in JO45 I have Es'hail-1 11199,8 MHz beacon with 22,6 dB S/N in 500 Hz BW. Using 85 cm offset dish and Octagon OSLO (no frequency stabilization). Using RTL dongle and HDSDR for demodulation.
I believe most S/N reports are referenced to 500 Hz bandwith, although not always stated. Otherwise difficult to compare numbers
Hallo Peter - DB2OS
Thank you for the information about spelling Es'hailSat and Es'hail-2 correctly.
But how do we pronounce this arabic word? Is "hail" pronounced like the English word "mail" or "snail" - or is it two syllables "ha-il".
Looking forward to meet you all on the satellite.
73 Ole OZ2OE