OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

General discussions
glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Wed Oct 03, 2018 2:49 pm

An immediate result of the performance assessment of the Receiving System, see:

https://www.sdrplay.com/community/viewt ... f=5&t=3685

is the availability of information which allows us to set the RSP Receiver up for optimum sensitivity and dynamic range in presence of noise. Table 1 of the previous thread has been modified, by directly showing typical receiver's Noise Figure obtained by the value of the terminated Noise Floor.

This post interrupted to allow editing. Go to next post.

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Wed Oct 03, 2018 3:05 pm

Let us consider the following example at a frequency of 3.6 Mhz: here on the HI Z balanced input we have a Noise Figure of 11.4 dB derived from a measured Noise Floor Rnf = -130 dBm. With these values, the theoretical Receiver Minimum Detectable Signal:

12/10/2018 I have added in red colour the values pertaining to the HI Z input impedance of 1 KOhm and uploaded dBm/voltage conversion tables in todays' post.

Rmds = -Rnf + 3 = -127 dBm.
Just to be on the safe side, in the following tractation we shall always consider Rmds = -Rnf + 6. So, to carry on:
Rmds = -124 dBm, which corresponds, in a 50 Ohm system, to a sensitivity of 0.141 uV and in a 1000 Ohm system to a sensitivity of 0.631 uV.

When we connect the antenna, suppose the Noise Floor increases to: Anf = -104 dBm, or a noise level of 1.41 uV (6.31 uV). Our Minimum Detectable Signal, which has to overcome noise by 6 dB, will be:
Amds = -104 + 6 = - 98 dBm, which corresponds, in a 50 Ohm system, to a sensitivity of 2.82 uV and in a 1000 Ohm system to a sensitivity of 12.6 uV.

All the above means that under these conditions we can insert between antenna and RSP receiver's input an attenuator with a value of:
At = -98 + 124 = 26 dB
WITHOUT CHANGING THE SIGNAL TO NOISE RATIO OF THE RECEIVED SIGNALS, BUT INCREASING THE RECEIVER'S OVERLOAD THRESHOLD (OR DYNAMIC RANGE) BY 26 dB. This is possible because our receiver, due to excessive antenna noise, is much more sensitive than actually necessary.

But this procedure has yet another important advantage when operating above 10 MHz, removing one obvious limitation of the RSP Receiver.

As many users have noticed, the I.F system noise figure deteriorates when above 10 MHz and to avoid S/N degradation, one has to operate the receiver at maximum RF gain, thus notably reducing the dynamic range. If antenna noise level is high, the operator is between a rock and a hard place: to avoid overload, gain must be reduced, but in doing so sensitivity suffers. The external attenuator reduces strong signals to an acceptable level, without altering the receiver's excellent sensitivity at all. In fact when you compare attenuator/no attenuator, it works like magic and you see the available signals just jumping off noise. Not only this, but when noise bursts come by, it is much more difficult for them to overload the receiver and receiver's recovery, if has to happen at all, is much quicker.

Another important advantage is improved Noise Blanker operation with impulse noise. By attenuating all signals coming from the antenna, also the absolute level of the impulse noise (whose step transients can be at a very high voltage, even if they contain little energy) is drastically reduced. With QRN it is a pleasure to turn the NBW on and hear the pulses go away entirely with a Blanker level as little as 160/150!

In the next post 500 Ohm balanced attenuator/balun data will be given.
Attachments
RX Noise table-Noise figure.jpg
Table 1A - RSP Receiver typical terminated Noise Floor & Noise Figure
RX Noise table-Noise figure.jpg (223.1 KiB) Viewed 24685 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Thu Oct 04, 2018 12:56 pm

Analogous calculations, as shown in the example above, can be done to determine the optimum atteuation for any frequency band, for any environmental noise area and for any antenna type, depending on the decile evaluation of SDR system performance, as described in the relevant thread.

For a long wire Beverage antenna having a load impedance of approx. 500 Ohm, here below "O" type attenuator data for Receiver System optimization is shown. Attenuation and resistor values have been calculated and selected in order to have relatively accurate attenuation/impedance values even if using standard 5% readily available resistor values. A low capacitance, toroid transformer type balun, 500 to 1 KOhm, should be used, as described here:

https://www.sdrplay.com/community/viewt ... f=5&t=3436

Condensed data: 10 mm dia. toroid core, F 43 ferrite material. Enamelled copper wire dia. 0.35 mm.

Additional attenuator information in a future post.
Attachments
Attenuators.jpg
Figure 1 - Balanced attenuator information
Attenuators.jpg (37.65 KiB) Viewed 24629 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Thu Oct 04, 2018 6:29 pm

Antenna is L=135 m Beverage, H = 6m

Uploaded below screens of 20 dB attenuated / non attenuated RSPduo reception on the 20 m band.
Attachments
14 MHz Att. Oct. 04 20.12 (1).jpg
14 MHz attenuated antenna
14 MHz Att. Oct. 04 20.12 (1).jpg (196.28 KiB) Viewed 24604 times
14 MHz No. Oct. 04 20.12 (2).jpg
14 Mhz straight thru antenna
14 MHz No. Oct. 04 20.12 (2).jpg (195.77 KiB) Viewed 24604 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Thu Oct 04, 2018 6:30 pm

Antenna is L=135 m Beverage, H = 6m

Uploaded below screens of 20 dB attenuated / non attenuated RSPduo reception on the 40 m band.
Attachments
7 MHz Att. Oct. 04 20.11.jpg
7 Mhz attenuated antenna
7 MHz Att. Oct. 04 20.11.jpg (189.16 KiB) Viewed 24604 times
7 Mhz No. Oct. 04 20.11.jpg
7 Mhz straight thru antenna
7 Mhz No. Oct. 04 20.11.jpg (199.75 KiB) Viewed 24604 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Thu Oct 04, 2018 6:31 pm

Antenna is L=135 m Beverage, H = 6m

Uploaded below screens of 20 dB attenuated / non attenuated RSPduo reception on the 80 m band.
Attachments
3.6 Mhz Att. Oct. 04 20.10.jpg
3.6 Mhz attenuated antenna
3.6 Mhz Att. Oct. 04 20.10.jpg (194.27 KiB) Viewed 24604 times
3.6 MHz No. Oct. 04 20.10.jpg
3.6 MHz straight thru antenna
3.6 MHz No. Oct. 04 20.10.jpg (191.52 KiB) Viewed 24604 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Fri Oct 05, 2018 9:53 am

In summary, to find the required attenuation:

1) Connect termination to input and find Terminated Receiver Noise Floor -Rnf
2) Find Receiver minimum detectable signal Rmds = -Rnf + 6
3) Connect antenna and find -Anf
4) Find Antenna minimum detectable noise floor Amds =-Anf + 6
5) Find attenuator value: At = -Amds + |Rmds|
6) Use a high value or even maximum RF gain for optimum S/N

Once the noise level of a frequency /antenna combination has been monitored for at least 10 days, the average noise floor found will allow the stable use of a front end attenuator.

For other input / output impedance values more data in Table 2 below. Resistor references are the same as in Figure 1 of previous post. For unbalanced attenuators, use a series resistor with a value equal to 2*R3.


REFERENCES:
Electronics for Engineers, Markus & Zeluff, 1° edition, McGraw-Hill 1945.
Reference Data for Radio Engineers, 5th Edition, ITT 1968.
Attachments
Attenuators (2).jpg
Table 2 - Additional attenuator values
Attenuators (2).jpg (143.16 KiB) Viewed 24531 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Fri Oct 05, 2018 2:53 pm

Here below a modified attenuator table based on multiples of 6 dB so that it will be easy to correct the "S" units indication by the RSP receiver. For instance, if you have inserted 24 dB of front end attenuation and your meter reads 2 "S" units, you are receiving a signal with "S" = 2+4 = 6 signal strength.
Attachments
Attenuator calc. (S units).jpg
Table 3 - "S" units attenuator table
Attenuator calc. (S units).jpg (190.06 KiB) Viewed 24498 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Fri Oct 05, 2018 5:46 pm

It is also possible to recover the correct reading of the noise floor, of the display and of the "S" meter by altering the DISPLAY CALIBRATION as shown in the enclosed screen. If, for instance, you install a 20 dB attenuator, the receiver's front end will still receive 20 dB less and be 20 dB farther away from overload, but all parameters will be shown at the +/- 20 dB level set on the calibration window.

Restoring the original screen levels has the advantage that improvements in antenna noise can be immediately noticed and one can then remove the attenuation as needed.
Attachments
Display calibration.jpg
Screen showing true levels despite a 20 dB attenuator between antenna and front end
Display calibration.jpg (308.3 KiB) Viewed 24483 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: OPTIMIZING RSP RECEIVER OPERATION WITH ANTENNA NOISE

Post by glovisol » Sat Oct 06, 2018 4:00 pm

It is possible to switch the same antenna between two, or even all three antenna inputs (with a third transformer) using the RDSPlay switch only. This setup allows very quick switching for comparing different noise/propagation conditions. Of course this is only possible when attenuation is necessary because of noise. It is also possible to feed separate front ends with the same antenna. The schematic is uploaded below.

The balanced nature of these multiplexing systems guarantees minimum possible noise feedthru.
Attachments
Switched inputs.jpg
Fast RSP input switching.
Switched inputs.jpg (21.5 KiB) Viewed 24436 times

Reason: No reason

Post Reply