Testing RSPduo receiver sensitivity with noise generator

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

Re: Testing RSPduo receiver sensitivity with noise generator

Post by glovisol » Sun Jul 01, 2018 6:23 am

More pics.
Attachments
DSC_1112 (Medium).JPG
DSC_1112 (Medium).JPG (70.37 KiB) Viewed 28081 times
DSC_1111 (Medium).JPG
DSC_1111 (Medium).JPG (140.83 KiB) Viewed 28081 times

Reason: No reason

vk7jj
Posts: 206
Joined: Thu Mar 02, 2017 9:56 am
Location: Tasmania

Re: Testing RSPduo receiver sensitivity with noise generator

Post by vk7jj » Sun Jul 01, 2018 7:38 am

Great pics, Gianfranco, lovely day for you and us, beautiful winter sun all day. We have our house on the market and today was an inspection which kept us busy in preparation.

BTW it was good of you to put out that extra couple of chairs for us, sorry we couldn't make it. We are settling in to a nice bottle of soft McLaren Vale Merlot though, a contrast to the big bold brassy reds of yesteryears.

It will be interesting to see if you are able to tell the difference with your trial with the lower feed point, it is very hard subjectively isn't it, with variable band conditions one's hopes can play tricks.

The move does seem worthwhile from a modelling point of view. I got the chance to add a 2m extra wire from the top of the horizontal through your window as I think you have it. It makes a difference, the table speaks for itself. I suppose I should do the same at the 560 ohm termination end but I'm not quite clear how that is organised.
g-b-tail.png
g-b-tail.png (131.45 KiB) Viewed 28077 times

Reason: No reason

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

Re: Testing RSPduo receiver sensitivity with noise generator

Post by glovisol » Sun Jul 01, 2018 8:12 am

Dear Phil, Thank you for the updated calcs., but about this technical aspect later. I am sitting in the sun, it is going to be a very hot day, just protected by the big umbrella. Preliminary report is that there is no question about not one, but two very big improvements.

1) The antenna lead does not come off the window. I pulled the cable off the hole and just added 2.5 m of thick copper wire to the operating table. Then I connected a 1 m. same copper wire to the earth stud and came straight to the RSP. (a) direct connection either to the Hi Z or to the 50 Ohm input simply does not work (80, 40, 20 & 15 m). Floor noise level goes up by 10 to 15 dB and signals are eaten away. (b) By putting in the wireline balun and using the proper taps the system started working, PULLING IN CROWDED 80 / 40 & 20 m bands in DAYTIME as I have never seen before I started testing the RSP's. It is going on still now, 10 hrs local time. I also heard, for the first time, signals on 15 m, a Greek guy coming in S9.

2) But friday also came the material 43 toroids I had ordered for higher frequency baluns, hoping to get to 30 MHz. Yesterday afternoon I quickly made one, separate windings, not wireline and yesterday I got such results on noise I did not dare to post until I had more confirmation. The results are too good to be true, but I shall post the details next week. I am enclosing a pic., just to show what I mean.

Kind regards,

Gianfranco
Attachments
DSC_1114 (Medium).JPG
DSC_1114 (Medium).JPG (69.66 KiB) Viewed 27852 times

Reason: No reason

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

Re: Testing RSPduo receiver sensitivity with noise generator

Post by glovisol » Sun Jul 01, 2018 6:22 pm

QUOTE REMOVED
glovisol

Reason: No reason

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

Re: Testing RSPduo receiver sensitivity with noise generator

Post by glovisol » Sun Jul 01, 2018 6:26 pm

Dear Phil, pse look again at the last antenna modeling table: it seems to me impossible such an incredible variation caused by just two m. of added length!

It is 20:30 local time & still reception goind strong. Details tomorrow.

Kind regards,

Gianfranco

Reason: No reason

vk7jj
Posts: 206
Joined: Thu Mar 02, 2017 9:56 am
Location: Tasmania

Re: Testing RSPduo receiver sensitivity with noise generator

Post by vk7jj » Mon Jul 02, 2018 12:02 am

Hi Gianfranco,
The results are too good to be true
I felt vicarious pleasure being to be able to copy and paste that happy comment!

Regarding the last table of results I posted that you have asked me to check, the significant difference would likely be feeding it from the bottom directly against ground as per your test. That latest 'guesswork' model was with the feed arrangement at the top of the vertical, all my other figures and models were with the feed point at the grounded end of the vertical.

Trying to guess the layout of your earth wire and the 2m extra horizontal piece wouldn't help, but then I'm also not sure of your wire layout at the far termination end.

One of the Beverage characteristics is that short bottom fed vertical wire radiator (in your case being 6m) so a meter or two of spare wire is a reasonable percentage of that distance and thus may influence behaviour more than expected.

Feeding it at the bottom seems the logical thing to do, you could stick to HF and put your favourite balun / transformer outside at the feed point and use a nearly invisible run of RG174 or similar for the run inside, it's useful stuff, I use a bit of it.

Alternatively bring in a run of 300 ohm ribbon from the bottom feed point so as to keep your frequency options open and play balun swapping inside?

BTW when playing with creative antenna layouts I always keep reminding myself that
the radiated electric field strength E is proportional to the integral of the current distribution along the antenna. The current at the center is just the driving current I=I0e−it and the current must drop to zero at the ends of the antenna, where the conductivity goes to zero. For a short antenna, we can make the approximation that the current declines linearly from the driving current at the center to zero at the ends
From the link below, couldn't find a shorter one but the whole article is interesting, skipping the maths and just going for the meat!

https://www.cv.nrao.edu/course/astr534/ ... heory.html

In other words the part of an antenna that radiates the most is the part where the current is at it's highest and as the current swings to zero, the voltage swings to it's maximum. So the radiation diminishes, following the sine wave voltage/current transformation curve along the antenna. High current + maximum radiation at the centre of a dipole, high voltage and no radiation at the ends of a dipole.

In your antenna the high current part is supposed to be at the bottom of and in the short vertical because that's where it is meant to be fed, my guess is that's why it's vertically polarised as per the model though that would only be the majority, there obviously would still be plenty of horizontal. And as we know transmit and receive current flows and hence radiation patterns behave the same way.

That's why the ends of a dipole can be bent around or drooped without too much compromise and also why loading coils at the bottom of shortened verticals make such antennas incredibly inefficient - not only do they consume lots of precious RF by turning it into heat by virtue of their radiation resistance / impedance losses but also they simply eliminate a huge swathe of the antenna's ability to radiate anything. Even the best screwdriver or mobile antenna on HF radiates almost nothing, all it has left is just the tiny tip! So much for mini-whips on HF! / end soapbox

--

So you are now officially on notice about the importance of those loose wires coming in your window and the part they play, the mathematically perfectly equal current flow† on that insignificant looking blue ground wire is radiating just as much RF as the current flow on that insignificant looking extra 2m of "antenna" wire, starting at the current maximum at, your balun ;-)

Sincerely, Phil

† except for the other currents we'll politely ignore (both transmit and receive) on any coax braid inside the house and USB braid and internal earth wires, all also happily radiating.

Reason: No reason

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

Re: Testing RSPduo receiver sensitivity with noise generator

Post by glovisol » Tue Jul 03, 2018 11:30 am

Dear Phil,

I enclose the data / graph resulting from your modelling table of Sat. June 30. I am still working on the noise improvement data, but wish to have a second confirmation and this I can have only when there is excess noise on air.

Kind regards,

Gianfranco
Attachments
Beverage 2-14 Graph.jpg
Beverage 2-14 Graph.jpg (94.55 KiB) Viewed 27903 times
Beverage 2-14 Data.jpg
Beverage 2-14 Data.jpg (48.16 KiB) Viewed 27906 times

Reason: No reason

vk7jj
Posts: 206
Joined: Thu Mar 02, 2017 9:56 am
Location: Tasmania

Re: Testing RSPduo receiver sensitivity with noise generator

Post by vk7jj » Tue Jul 03, 2018 11:42 pm

Dear Gianfranco, so sorry I missed your message, I don't log in unless I wish to post and so it remained invisible, please accept my apologies for that.

Your chart above is an interesting view of the data, well done. Great to hear that signals are good too, despite what are at the moment very poor band conditions globally.

Of the two models built so far, the first tables came from the Beverage as illustrated on page 3 with the feed point at the bottom of the vertical wire at the earth point. The last table was based on the top fed arrangement with the extra two wires as per the pic below.

You can see my approximation of what I thought looked like the arrangement of the wires into your house, the bold black wire is 2m long, the feed point is the red dot.

Just for fun here is a closeup of the (red) current distribution on 144.1MHz. That frequency makes the current characteristics more obvious and I have zoomed them so they are exaggerated.
gb-original-2m-current.png
gb-original-2m-current.png (32.23 KiB) Viewed 27893 times
The following two pics give an insight into the vertical v's horizontal components of the 40m radiation pattern of the second model:
gb-original-40m-Vpol.png
gb-original-40m-Vpol.png (84.73 KiB) Viewed 27893 times
Horizontal:
gb-original-40m-Hpol.png
gb-original-40m-Hpol.png (76.38 KiB) Viewed 27893 times
So we need to decide on what antenna configuration needs to be modelled next before putting any reliance on the data, and correct errors in regards to the feed wire arrangement and the termination at the other end.

Hoping that helps!!

Best regards, Phil

Reason: No reason

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

Re: Testing RSPduo receiver sensitivity with noise generator

Post by glovisol » Thu Jul 05, 2018 4:18 am

Dear Phil,

In effect I can confirm the receiving patterns you have modelled. The antenna follows a true north-south direction, with the termination at the south point. In the last 15/20 days the prevailing stations have come from Southern Italy, Spain, France, Greece and South Balkans. I shall now complete the antenna work by REMOVING the termination to check if stations in the north direction will also start coming in.

To complete the modelling question, in my opinion the antenna to be modelled for the time being is the one with the feed point at the bottom of the vertical lead to the earth stake, e.g. the model for which I did the graphing of the parameters.

I have now concluded the study on low noise reception and some conclusions, in view of the activities described in this thread, will perhaps surprise you. For this study I shall open a separate thread.

Kind regards,

Gianfranco

Reason: No reason

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

Re: Testing RSPduo receiver sensitivity with noise generator

Post by glovisol » Sat Jul 07, 2018 3:38 pm

The "PERFECT" wideband antenna??

Dear Phil,

Yesterday I received a brand new Rig Expert's AA-30.ZERO Vector Impedance Anayser. It took 24 hours to make it work, because of a small problem on the PC board, but now it works. Enclosed is the Z = R + JX plot obtained by connecting the antenna leads on my desk to the analyser, range 3.6 - 15 MHz. Unfortunately I had disconnected the antenna termination resistor, to check back directivity, so this is a plot of a "wide open" bidirectional antenna: it really look like a travelling wave. Tomorrow I shall re-connect the termination.

The Real part oscillates between 600 and 200 Ohm. Reactance is always capacitive and fairly constant with frequency. More testing, study and thought necessary....

Kind regards,

Gianfranco
Attachments
The AA-30.ZERO.JPG
The Vector Analyzer in operation
The AA-30.ZERO.JPG (144.41 KiB) Viewed 27810 times
First Beverage R+JX plot.jpg
First 3.5 - 15 MHz R + JX plot
First Beverage R+JX plot.jpg (113.53 KiB) Viewed 27818 times

Reason: No reason

Post Reply