BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Tue Jul 23, 2019 6:42 am

Thank you for directing our attention to these two commercially available pre-selectors. Both of them, in the frequency ranges they cover, will protect the RSP front end by reducing the bandwidth seen by the front and therefore the chance of overload and spurious.

My only regret is that these units, which are electronic hardware, are presented without any specification worth its name. Let me better explain my opinion: the MFJ-1046 retails for $129.95. The Cross Country (with transport) for $120.00. The RSP-1A retails around $ 120.00. The paradox is this: you can receive a signal with the RSP-1A alone, but, with a preselector, at the same price, without a receiver you receive nothing.

Yet you buy an RSP-1A with performance guaranteed by one set of specifications which are among the best and most complete that I have ever seen, so the specifications first, just by being given, tell you exactly what you are buying and second, by the qualit they show, tell you that you are buying incredible value for your money. On the contrary, with the proposed pre-selectors, you buy a piece of hardware without knowing what you are buying and worse, if you have to choose among various models, you have no criteria and all reduces to the beauty of the box in the eye of the beholder.

As an ex industrialist I very well realise that if you manufacture a product in limited quantities, also a product that can be hand manufactured only, then it is going to cost a lot. What I do not understand and surprises me is why the manufacturers did not take the little time and effort required to prepare a complete set of specifications or, if they did, why these are not shown in their adverts. The screens shown for the Cross Country unit tell you absolutely nothing.

The lack of availability of test instrumentation is no excuse, considering that nowadays with an RSP-1A (again!) and the fantastic Steven's software, plus an $12.00 Noise Generator you can fully characterise the pre-selector's performance at an incredibly low instrumentation cost.

Finally, what I would like to convey to many Forum members is the fact, that, just because we have the luck of the availability of such low cost outstanding quality instrumentation (never happened before in Human History) if to this you add a solderig iron and a few tools, then you enter the magic and rewarding world of building something for yourself, which is liable to give you the best gift of all: joy and knowledge through experience.
Last edited by glovisol on Wed Jul 24, 2019 12:15 pm, edited 1 time in total.

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Tue Jul 23, 2019 10:27 am

It has come to my mind that with my previous post I have perhaps given the impression that the pre-selectors introduced by mike0agner are not technically effective and therfore not useful. If this is the case I sincerely apologize. On the contrary, my only criticism was the lack of published specs, but this does not mean these pre-selectors can be not only useful, but even indispensable in some situations. I am somewhat short on time now, but later I shall come back on this subject and, having the facilities on the ready, will produce a set of computed, likely performance curves for these units.
Last edited by glovisol on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Tue Jul 23, 2019 7:05 pm

2.3. A DIGRESSION - NARROW BAND PRE-SELECTORS COMPUTER SIMULATION

In general a narrow band pre-selector consists of a tuneable single tuned bandpass L/C circuit placed between antenna output and receiver input. The inductor is wound on a Powdered Iron toroid core and the capacitor is an air variable tuning the L/C to the desired receive frequency. Two basic configurations are possible: (a) one toroid inductor with multiple switched taps or (B) multiple switched toroid inductors, one for each frequency band, to accommodate all frequency ranges between 1.8 and 30 MHz. Solution (B) is preferable as it allows different toroid materials, each with the best Qu for that frequency range, while (A) is more of a compromise. Under the above hypothesis, the following basic parameters have been chosen.

2 - 4 MHz .........Qu=350..........L=18uH
4 - 8 MHz .........Qu=320..........L=5uH
8 - 20 MHz .......Qu=180...........L=1.4uH
20 - 30 MHz ......Qu=140..........L=0.4uH

Powdered Iron materials considered are Amidon #2, #3 & #6. Q values have been obtained by the publication: Amidon RF6 Q Curves and the inductance values have been calculated in order to optimise Qu in the frequency range under consideration. For these conditions, the loading value for the L/C circuit has been selected to obtain maximum possible selectivity, but keeping the insertion loss always below 3 dB. The air variable considered has a standard capacitance range of 50 to 360 pF.

For each frequency range two selectivity curves have been produced for each boundary frequency. A NB, Narrow Band curve and a WB, Wide Band curve, so that adjacent frequency attenuation can be seen, as well as far frequencies up to +/- 800 KHz. Overalaps are also included, in order to appreciate the difference when switching between bands.

I hope other Forum members, equipped with pre-selectors, will upload performance data to compare with the simulation results. The uploaded plots show the relevant parameters under the top title. As an example the pre-selector, tuned at 2 MHz will have an insertion loss of 2.2 dB, a NB selectivity of -8 dB @ 1980 and 2020 KHz (Figure 1) and a WB selectivity in excess of -35 dB @ 1800 and 2200 KHz. The tuning capacitor would be set @ 352 pF.

More curves in the next posts.
Attachments
1. 2MHz-NB.png
FIGURE 1. Narrow band selectivity @ 2 MHz
1. 2MHz-NB.png (18.99 KiB) Viewed 35179 times
2. 2MHz-WB.png
FIGURE 2. Wide band selectivity @ 2 Mhz
2. 2MHz-WB.png (24.32 KiB) Viewed 35179 times
Spreadshhet.png
FIGURE 3. Sreadsheet calculation sample
Spreadshhet.png (31.06 KiB) Viewed 35179 times
Last edited by glovisol on Wed Jul 24, 2019 12:19 pm, edited 1 time in total.

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Wed Jul 24, 2019 5:50 am

2.3. A DIGRESSION - NARROW BAND PRE-SELECTORS COMPUTER SIMULATION

Uploaded below Figures 4 & 5 showing simulated performance @ 4 MHz. Figure 4 (under top title) shows the following parameters:

CF- Center Frequency: 4 MHz
Qu - Unloaded Q: 350
QL- Loaded Q: 80
IL - Insertion loss: 2.3 db
L - Coil inductance: 18 uH
CT - Tuning capacitance: 88 pF
Attachments
3. 4MHz-NB.png
FIGURE 4. Narrow band selectivity @ 4 MHz
3. 4MHz-NB.png (18.33 KiB) Viewed 35147 times
4. 4MHz-WB.png
FIGURE 5. Wide band selectivity @ 4 MHz
4. 4MHz-WB.png (21.9 KiB) Viewed 35147 times
Last edited by glovisol on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Wed Jul 24, 2019 5:55 pm

2.3. A DIGRESSION - NARROW BAND PRE-SELECTORS COMPUTER SIMULATION

Uploaded below Figures 6 & 7 showing simulated performance @ 4 MHz. This is an overlap, having switched from the 2-4 MHz to the 4-8 MHz band. Figure 6 (under top title) shows the following parameters:

CF- Center Frequency: 4 MHz
Qu - Unloaded Q: 320
QL- Loaded Q: 50
IL - Insertion loss: 1.5 db
L - Coil inductance: 5 uH
CT - Tuning capacitance: 316 pF


As we go to higher frequencies pre-selector selectivity diminishes for two reasons:
- Unloaded Qu available from Powder Iron toroids goes down and so loaded QL must also go down to limit insertion loss.
- Bandwidt, being a percent of Center Frequency, in numerical terms increases as center frequency increases.
Attachments
5. 4MHz-NB.png
FIGURE 6. Overlap - Narrow band selectivity @ 4 MHz
5. 4MHz-NB.png (18.1 KiB) Viewed 35111 times
6. 4MHz-WB.png
FIGURE 7. Overlap - Wide band selectivity @ 4 MHz
6. 4MHz-WB.png (22.95 KiB) Viewed 35111 times
Last edited by glovisol on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Thu Jul 25, 2019 3:44 pm

2.3. A DIGRESSION - NARROW BAND PRE-SELECTORS COMPUTER SIMULATION

Uploaded below Figures 8 & 9 showing simulated performance @ 8MHz. Figure 8 (under top title) shows the following parameters:

CF- Center Frequency: 8 MHz
Qu - Unloaded Q: 320
QL- Loaded Q: 50
IL - Insertion loss: 1.5 db
L - Coil inductance: 5 uH
CT - Tuning capacitance: 80 pF

As we go to higher frequencies pre-selector selectivity diminishes for two reasons:
- Unloaded Qu available from Powder Iron toroids goes down and so loaded QL must also go down to limit insertion loss.
- Bandwidt, being a percent of Center Frequency, in numerical terms increases as center frequency increases.
Attachments
7. 8MHz-NB.png
FIGURE 8. Narrow band selectivity @ 8 MHz
7. 8MHz-NB.png (18.24 KiB) Viewed 35069 times
8. 8MHz-WB.png
FIGURE 9. Wide band selectivity @ 8 MHz
8. 8MHz-WB.png (22.87 KiB) Viewed 35069 times
Last edited by glovisol on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Fri Jul 26, 2019 6:16 am

2.3. A DIGRESSION - NARROW BAND PRE-SELECTORS COMPUTER SIMULATION

Uploaded below Figures 10 & 11 showing simulated performance @ 8 MHz. This is an overlap, having switched from the 4-8 MHz to the 8-20 MHz band. Figure 10 (under top title) shows the following parameters:

CF- Center Frequency: 8 MHz
Qu - Unloaded Q: 180
QL- Loaded Q: 40
IL - Insertion loss: 2.2 db
L - Coil inductance: 1.4 uH
CT - Tuning capacitance: 283 pF
Attachments
9. 8MHz-NB.png
FIGURE 10. Overlap - Narrow band selectivity @ 8 MHz
9. 8MHz-NB.png (17.08 KiB) Viewed 35020 times
10. 8MHz-WB.png
FIGURE 11. Overlap - Wide band selectivity @ 8 MHz
10. 8MHz-WB.png (22.65 KiB) Viewed 35020 times
Last edited by glovisol on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Fri Jul 26, 2019 6:23 am

2.3. A DIGRESSION - NARROW BAND PRE-SELECTORS COMPUTER SIMULATION

Uploaded below Figures 12 & 13 showing simulated performance @ 20 MHz. Figure 12 (under top title) shows the following parameters:

CF- Center Frequency: 20 MHz
Qu - Unloaded Q: 180
QL- Loaded Q: 40
IL - Insertion loss: 2.2 db
L - Coil inductance: 1.4 uH
CT - Tuning capacitance: 45 pF
Attachments
11. 20MHz-NB.png
FIGURE 12. Narrow band selectivity @ 20 MHz
11. 20MHz-NB.png (14.39 KiB) Viewed 35020 times
12. 20MHz-WB.png
FIGURE 13. Wide band selectivity @ 20 MHz
12. 20MHz-WB.png (19.99 KiB) Viewed 35020 times
Last edited by glovisol on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Fri Jul 26, 2019 4:57 pm

2.3. A DIGRESSION - NARROW BAND PRE-SELECTORS COMPUTER SIMULATION

Uploaded below Figures 14 & 15 showing simulated performance @ 20 MHz. This is an overlap, having switched from the 8-20 MHz to the 20-30 MHz band. Figure 14 (under top title) shows the following parameters:

CF- Center Frequency: 20 MHz
Qu - Unloaded Q: 140
QL- Loaded Q: 40
IL - Insertion loss: 2.9 db
L - Coil inductance: 0.4 uH
CT - Tuning capacitance: 160 pF
Attachments
13. 20MHz-NB.png
FIGURE 14. Overlap - Narrow band selectivity @ 20 MHz
13. 20MHz-NB.png (13.99 KiB) Viewed 34960 times
14. 20MHz-WB.png
FIGURE 15. Overlap - Wide band selectivity @ 20 MHz
14. 20MHz-WB.png (19.97 KiB) Viewed 34960 times
Last edited by glovisol on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

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

Re: BUILDING SELECTIVITY PROTECTION INTO THE RSP FRONT END IN LF/HF

Post by glovisol » Fri Jul 26, 2019 5:03 pm

2.3. A DIGRESSION - NARROW BAND PRE-SELECTORS COMPUTER SIMULATION

Uploaded below Figures 16 & 17 showing simulated performance @ 30 MHz. Figure 16 (under top title) shows the following parameters:

CF- Center Frequency: 30 MHz
Qu - Unloaded Q: 140
QL- Loaded Q: 40
IL - Insertion loss: 2.9 db
L - Coil inductance: 0.4 uH
CT - Tuning capacitance: 70 pF
Attachments
15. 30MHz-NB.png
FIGURE 16. Narrow band selectivity @ 30 MHz
15. 30MHz-NB.png (15.04 KiB) Viewed 34960 times
16. 30MHz-WB.png
FIGURE 17. Wide band selectivity @ 30 MHz
16. 30MHz-WB.png (19.91 KiB) Viewed 34960 times
Last edited by glovisol on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

Post Reply