Reply to thread

Message

<blockquote data-quote="Old Fred" data-source="post: 22930">I wouldn't put too much faith in noise figures. I've read reports that say, for example, that the Manhattan MTI 0.6dB LNB works better (for Astra 2D) than the Invacom 0.3 or the Syntec 0.3. And consider these points:-1. It's very difficult to measure such low noise figures.2. They are measured under laboratory conditions with &quot;perfect&quot; 75 Ohm cable simulation. In practice, &quot;75 Ohm&quot; cable can be anywhere from 65 to 85 Ohm (worse if you kink it!)3. It's not feasible to measure noise for every LNB across the entire two frequency bands (high and low). So the manufacturers make a couple of &quot;spot&quot; measurements at specific frequencies and &quot;extrapolate&quot; (= guess) that those measurements are &quot;typical&quot; for the whole spectrum. And they measure only a few LNBs in each batch of thousands.4. Even Swedish Microwave *allegedly* have only a dozen little graphs which they print and put in LNB boxes at random! (I don't know if this is true but it makes sense). But at least their graphs really are &quot;typical&quot;.As for &quot;PLL&quot; (Phase Locked Loop), I don't know of any other way to establish an accurate and virtually drift-free Local Oscillator frequency, so I guess ALL LNBs must use a &quot;PLL&quot;.Kinda like boasting &quot;bread made with REAL flour!&quot;As if there was some other kind of bread.</blockquote>

[QUOTE="Old Fred, post: 22930"] I wouldn't put too much faith in noise figures. I've read reports that say, for example, that the Manhattan MTI 0.6dB LNB works better (for Astra 2D) than the Invacom 0.3 or the Syntec 0.3. And consider these points:- 1. It's very difficult to measure such low noise figures. 2. They are measured under laboratory conditions with "perfect" 75 Ohm cable simulation. In practice, "75 Ohm" cable can be anywhere from 65 to 85 Ohm (worse if you kink it!) 3. It's not feasible to measure noise for every LNB across the entire two frequency bands (high and low). So the manufacturers make a couple of "spot" measurements at specific frequencies and "extrapolate" (= guess) that those measurements are "typical" for the whole spectrum. And they measure only a few LNBs in each batch of thousands. 4. Even Swedish Microwave *allegedly* have only a dozen little graphs which they print and put in LNB boxes at random! (I don't know if this is true but it makes sense). But at least their graphs really are "typical". As for "PLL" (Phase Locked Loop), I don't know of any other way to establish an accurate and virtually drift-free Local Oscillator frequency, so I guess ALL LNBs must use a "PLL". Kinda like boasting "bread made with REAL flour!" As if there was some other kind of bread. [/QUOTE]

Verification