RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Thu Jul 12, 2018 7:52 am

ANTENNA INDUSTRIAL NOISE SUPPRESSION AT 3.5 & 7 MHz WITH HIGHPASS FILTER

More illustrations below.
Attachments
Noise filter schematic.jpg
Filter example for initial testing
Noise filter schematic.jpg (58.95 KiB) Viewed 25261 times
Noise filter.JPG
Badly constructed prototype is still effective
Noise filter.JPG (88.16 KiB) Viewed 25261 times

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Thu Jul 12, 2018 9:45 am

...But this Butterworth n=6 in a balanced configuration seems to work much better! Industrial noise has gone, 3.5 to 14 MHz. Noise floor is now close to - 135 dBm on all these bands! Await confirmation from other Forum dwellers.

glovisol

IMPORTANT ERRATA CORRIGE as of 13/07/18: The original balanced filter schematic had a serioud error in that all capacitor values were wrong. I have removed the wrong schematic and replaced it with the revised version with correct capacitance values. Sincere apologies, glovisol
Attachments
Noise filter schematic.jpg
REVISED Balanced noise filter schematic
Noise filter schematic.jpg (59.2 KiB) Viewed 25214 times

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Fri Jul 13, 2018 11:48 am

Please be aware that in the balanced n=6 filter schematic in the previous post all capacitance values were wrong. I have now replaced the wrong schematic with the correct version. Sincere apologies,

glovisol

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Fri Jul 27, 2018 11:16 am

The advantages of the high pass noise filter can be extended to receivers equipped with the 50 Ohm input only. The uploaded schematic is for an umbalanced 50 Ohm input / 50 Ohm output. Balanced antenna input can still be achieved by placing a balun at the filter's input and isolating the grounds, as shown in the previous posts. The balun could accommodate a variety of antenna impedances by adjusting the PRI/SEC turns ratio. This would bring the advantages of balanced feed to single ended input receivers like the SP1A.

All capacitor values are 5% standard. Capacitors are polyester, stable enough for this non-critical requirement filter. Attenuation of the broadcast band is over 40 dB, thereby removing, in addition to noise, the danger of front end overload by local strong signals.

COMPONENTS:
L1 - 5uH
L2 - 1.4 uH
L3 - 1.9 uH
C1 - 680 pF
C2 - 68 pF
C3 - 470 pF
C4 - 82 pF
C5 - 1000 pF
C6 - 1000 pF
Attachments
Umbal. noise filter n=6.jpg
Umbal. noise filter n=6.jpg (44.55 KiB) Viewed 25133 times

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Sat Jul 28, 2018 6:52 am

High pass n = 6 calculated filter performance is shown below. Attenuation in blue & return loss in red.
Attachments
ScreenHunter_157 Jul. 28 08.50.jpg
FILTER ATTENUATION / RETURN LOSS TABLE
ScreenHunter_157 Jul. 28 08.50.jpg (24.56 KiB) Viewed 25091 times
N=6 filter graph.jpg
FILTER ATTENUATION / RETURN LOSS GRAPH
N=6 filter graph.jpg (53.34 KiB) Viewed 25091 times

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Tue Jul 31, 2018 10:30 am

Application of the high pass noise filter to build a balanced input for the RSP1A, as anticipated in the previus post.

This scheme allows for best ground separation. One side of the balun's input is connected to the antenna's ground. Connection between antenna and balun may be also done using a balanced line.

The ideal situation, as far as noise intake is concerned, is to place the RSP1A + noise filter right at the long wire antenna output terminal/ground stake and to use a long (buffered) active USB cable to connect to the PC.

Below is winding table based on a 10 mm dia. toroid core, F 43 ferrite material. Enamelled copper wire dia. 0.35 mm.
Attachments
Balun ratios.jpg
Balun construction data table
Balun ratios.jpg (47.73 KiB) Viewed 24977 times
Xfrmr coupled noise filter n=6.jpg
Balun input for the RSP1A
Xfrmr coupled noise filter n=6.jpg (51.51 KiB) Viewed 24987 times

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Fri Aug 17, 2018 6:58 am

Preliminary tests of front end High Pass filter

I enclose two comparisons of reception with and without HP filter on 40 and 20 m. These have been done with a makeshift 15 m long wire antenna and a water pipe ground while on a flat in a sea resort, so they are by no means definitive, but give some indication on possible usefulness of cutting off the lower spectrum while working on HF bands. Far more extensive testing is planned with the Beverage antenna in september next.

The test consisted in quickly inserting/removing the High Pass filter at the balanced input of the RSPduo. The change was done at an average of less than 30 seconds to avoid input signal level variations due to propagation changes during the swap.

The change in the level of the noise floor is impressing, even though a decrease in the received signal level is also noticeable, so there must be a some filter insertion loss, which will also have to be investigated further on.
Attachments
HP filter test 14 Mhz Aug. 4.jpg
HP filter test 14 Mhz Aug. 4.jpg (162.73 KiB) Viewed 24762 times
HP filter test 7 MHz Aug. 4.jpg
HP filter test 7 MHz Aug. 4.jpg (163.27 KiB) Viewed 24762 times

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Thu Oct 04, 2018 1:14 pm

FINAL ASSESSMENT OF HIGH PASS FILTER ACTION AND POSSIBLE NOISE ADVANTAGE

After a series of measurements it has been determined that the biggest part of the noise advantage shown by the highpass filter in HF was provided by the significant attenuation (6-8 db) of the filter itself, depending on operating frequency.

This is not to say that the use of a highpass filter is not useful in itself, in fact it provides a significant contribution in cutting off very strong LF and MW signals which could overload the receiver's front end. In any case significant noise advantage can be more favorably obtained with front end attenuation, as described here:

https://www.sdrplay.com/community/viewt ... f=5&t=3723

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Wed Oct 17, 2018 8:18 am

NOISE BLANKER EFFECTIVENESS DEMO

The Noise Blanker performance of the RSPduo, according to my experience, is extremely good. On the 80 m band, where impulse noise in daytime is sometimes so strong as to blot out weak signals, the NB totally removes it without affecting receiver's sensitivity. Comparison PC screens, showing a notable 8.3 dB improvement, enclosed.
Attachments
NB EffectivenessOct. 17 09.58.jpg
Noise Blanker OFF
NB EffectivenessOct. 17 09.58.jpg (274.57 KiB) Viewed 24199 times
NB Effectiveness Oct. 17 09.58.jpg
Noise Blanker ON
NB Effectiveness Oct. 17 09.58.jpg (248.95 KiB) Viewed 24199 times

Reason: No reason

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

Re: RSP RECEIVER LOW NOISE HF OPERATION WITH LONG WIRE ANTENNAS

Post by glovisol » Wed Oct 31, 2018 9:41 am

RECEIVING SYSTEM ASSESSMENT @ 1.8 MHz

In this and in the following posts we shall use the data and information shown here:

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

to assess the noise performance in HF of an RSP receiver system equipped with the already described L=135 m Beverage antenna. Noise Floor measurements have been performed daily over two weeks and data has been classified according to Table 3 of the cited reference and also downloaded below. Received Noise Floor data has been corrected to take antenna gain/loss under consideration, as follows.

Wavelength Wl @ 1.8 MHz: Wl = 300/1.8 = 166 m
Ratio L/Wl: R = 135/166 = 0.81
From "Beverage Antenna gain Vs. wavelength", Figure 1 in the cited reference: G = -6.5 dB referred to the isotropic. This loss is due to the fact that the antenna is too short when considering a wavelength of 166 m.
Balun insertion loss is 0.8 dB. Total loss is 7.3 dB.

Therefore the Noise Floor measurements taken with this antenna show a noise level of 7.3 dB less than the level we should have obtained if the measurement had been taken with an isotropic antenna. To compare the measurement with data and criteria of Table 3, we must correct by adding 7.3 dB to it.

The graph shown below shows the raw measurements in ligt blue and the corrected measurement in red, compared to the noise leve ranges of "Rural" and "Quiet Rural". The Rural decile range extends for +/-3.5 dB, while the Quiet Rural range extends for +/- 3 dB. By inspecting the graph, it is evident that the antenna is too short at this frequency: in fact in several days the measured noise is notably below the lowest threshold of the Quiet Rural range. The real situation is then shown after the correction.

The noise peaks shown 22/10 to 25/10 and 28/10 to 29/10 have been caused by electrical activity due to stormy weather, which has a very strong influence at this frequency. In general, in average weather conditions, we can classify the receiving system under examination as located in a Quiet Rural situation, therefore little or nothing can be done to improve the receiving system background noise.
Attachments
Beverage Gain.jpg
Figure 1 - Beverage Antenna Gain Vs. Wavelength ratio
Beverage Gain.jpg (284.38 KiB) Viewed 24068 times
Noise floor (3).jpg
TABLE 3 - PREDICTED & CLASSED NOISE DATA
Noise floor (3).jpg (108.92 KiB) Viewed 24074 times
1. Noise meas. 1.8 MHz.jpg
Two week Noise Floor measurement data @ 1.8 MHz
1. Noise meas. 1.8 MHz.jpg (329.33 KiB) Viewed 24074 times

Reason: No reason

Post Reply