|
Hello Jerry and group, The DVB-T dongles are not really intended to be used as I/Q capture devices, as-designed - and so the 40dB dynamic range isn't necessarily an issue, within the receive passband of the unit.? Since they're designed as TV tuners, they have a tuner before the demod (which contains the ADC) that has a tracking filter of sorts - depending upon the tuner architecture, this may be a zero-IF baseband filter, or a SAW filter, etc. In short, the 40dB dynamic range should theoretically only affect the 8MHz-ish span of the tuner's frontend filter - but the frontends aren't always terribly robust on consumer devices, and the filter isn't infinite-stop-band for adjacent frequencies.? For amateur use, this is obviously a huge swath of signals (perhaps even encompassing the second harmonic!), but for the intended use of TV reception you're only looking at the adjacent TV channels down in the skirts of the passband filter. I hope this explanation makes sense - to be honest I'm speaking in broad generalities, because I've never looked closely at how the popular dongles behave.? However, I am rather familiar with how modern TV tuner/demod chipsets behave. As an aside - knowing and understanding the limitations of your test equipment can enhance their useful-ness considerably.? For instance, you might be able to look farther than 40dB down, harmonic-wise, if you can get the fundamental signal outside of the tuner's passband while measuring the harmonic content - but this assumes that the front-end isn't overloading from the fundamental.? This should be able to be determined by varying the input level of the fundamental, and then watching for the harmonic levels to increase/decrease by the same amount (that is, add a 3dB pad and verify that the harmonics decreased by 3dB, and not 6dB as you might see from input overload).? I would suggest using different valued external pads/attenuators, so that all measurements are made with the same internal input attenuation setting on the tuner itself - to try and minimize the error caused by (ab)using a consumer device as test equipment. 73, Josh, KB8NYP On Mon, Jan 21, 2019 at 11:33 AM Jerry Gaffke via Groups.Io <jgaffke=[email protected]> wrote: If anybody can remember more about this 40dB dynamic range get around, I'd be very curious. |
