RSP-1A INPUT IMPEDANCE IN THE RANGE 0-60 MHZ & SPECTRUM ANALYSER OPERATION
Posted: Tue Apr 23, 2019 8:14 am
RSP-1A INPUT IMPEDANCE IN THE RANGE 0 - 60 MHz
I would like to focus the attention on the structure of the input circuitry of the RSP class receivers as compared to that of "conventional" receivers and/or "modern design transceivers". Most of the latter have (and rightly so) bandpass tuned circuitry at the input, to obtain as much selectivity as possible at the reciver's input, in order to attenuate, as much as possible, interference of strong nearby unwanted signals. This is the reason why the input impedance of all these units can be defined as "nominal", because it depends on the variable characteristics of the input tuneable circuitry.
By virtue of necessity, the RSP class Receiver are different and very much so, even if there is some resemblance to the very old RACAL RA17 receiver. Referring to the uploaded original blocks diagram, here the protection of the wideband front end cicuitry is obtained by an attenuator "Gain Control", followed by a set of four switched band-pass passive filters. Nothing is being tuned here, it is all fixed, save for the variable gain control circuit.
Considering the receiver's input with MW, FM and DAB filters not switched in, the antenna input has to go throuh the "gain control" variable attenuator. If the attenuator is in, the input impedance will be mostly determined by the attenuator resistive impedance. In position 0 the input is connected directly to one of the bandpas filters, more of this later.
From position 1 to position 6 attenuation moves from 6 dB to 61 dB: this means that the input impedance is not only resistive, but becomes more and more independent from what follows. It is easy to calculate the relationship between attenuation and the value of the input impedance assuming an open circuit after the attenuator and I shall not do it here, but still the presence of the attenuator makes the input impedance very near 50 Ohm, with the resistive and reactive components getting nearer and nearer to 50 Ohm resistive as attenuation increases.
If we now look at attenuator position 0, we see that the receiver's input impedance is totally determined by the impedance of the bandpass filter actually switched in. The response of these filters is fully described and pictured in several pages of the detailed specification. It is quite reasonable to suppose these filters have been designed to work between 50 Ohm terminations, so even at maximum gain (no attenuator in between) THE INPUT IMPEDANCE SHOULD BE VERY NEAR 50 Ohm.
I did a test with a Network Analyser in the frequency range 3-10 MHz (e.g. sweeping the input impedance of filter 2-12 MHz) with maximum gain (no attenuator in) and obtained a result fairly corresponding to 50 Ohm.
https://www.sdrplay.com/community/viewt ... 4&start=10
In conclusion the Spectrunm Analyser software works with some input attenuation most of times, so we can be fairly sure there will be no detrimental effects on measurements due to a non-defined and/or variable Analyser input impedance.
I would like to focus the attention on the structure of the input circuitry of the RSP class receivers as compared to that of "conventional" receivers and/or "modern design transceivers". Most of the latter have (and rightly so) bandpass tuned circuitry at the input, to obtain as much selectivity as possible at the reciver's input, in order to attenuate, as much as possible, interference of strong nearby unwanted signals. This is the reason why the input impedance of all these units can be defined as "nominal", because it depends on the variable characteristics of the input tuneable circuitry.
By virtue of necessity, the RSP class Receiver are different and very much so, even if there is some resemblance to the very old RACAL RA17 receiver. Referring to the uploaded original blocks diagram, here the protection of the wideband front end cicuitry is obtained by an attenuator "Gain Control", followed by a set of four switched band-pass passive filters. Nothing is being tuned here, it is all fixed, save for the variable gain control circuit.
Considering the receiver's input with MW, FM and DAB filters not switched in, the antenna input has to go throuh the "gain control" variable attenuator. If the attenuator is in, the input impedance will be mostly determined by the attenuator resistive impedance. In position 0 the input is connected directly to one of the bandpas filters, more of this later.
From position 1 to position 6 attenuation moves from 6 dB to 61 dB: this means that the input impedance is not only resistive, but becomes more and more independent from what follows. It is easy to calculate the relationship between attenuation and the value of the input impedance assuming an open circuit after the attenuator and I shall not do it here, but still the presence of the attenuator makes the input impedance very near 50 Ohm, with the resistive and reactive components getting nearer and nearer to 50 Ohm resistive as attenuation increases.
If we now look at attenuator position 0, we see that the receiver's input impedance is totally determined by the impedance of the bandpass filter actually switched in. The response of these filters is fully described and pictured in several pages of the detailed specification. It is quite reasonable to suppose these filters have been designed to work between 50 Ohm terminations, so even at maximum gain (no attenuator in between) THE INPUT IMPEDANCE SHOULD BE VERY NEAR 50 Ohm.
I did a test with a Network Analyser in the frequency range 3-10 MHz (e.g. sweeping the input impedance of filter 2-12 MHz) with maximum gain (no attenuator in) and obtained a result fairly corresponding to 50 Ohm.
https://www.sdrplay.com/community/viewt ... 4&start=10
In conclusion the Spectrunm Analyser software works with some input attenuation most of times, so we can be fairly sure there will be no detrimental effects on measurements due to a non-defined and/or variable Analyser input impedance.