Keyboard shortcuts could be added easily using a dedicated plugin if someone is motivated enough to cover all the controls - and there are dozens. The challenge here is to avoid messing with the different settings from third party plugins that may use the same shortcuts.

For the frequency xml database, you can use mklink as administrator to create links to the same target. Note that by doing so, if you manipulate the file in instance 1 of SDR#, other instances are not aware of the changes and may just overwrite the file causing data loss. If you just need one master file for all instances in read only mode, this should work.

Maybe binocular ferrites can be a good compromise for the size, performance and reproducibility.

The transformer can have a single turn on the primary. In
that case, make a short tube that fills most of the inner
diameter of the toroid. Distribute the secondary winding
evenly over the toroid. That would minimize the leak
inductance.

You can even use the loop itself as primary, like in this photo...
In this case the transformer had only two turns as secondary, may be more would have resulted
in a greater efficiency...


 

Yes, with a transformer that lifts the very low impedance
to about 50 ohms one can get a fairly good sensitivity.

The transformer can have a single turn on the primary. In
that case, make a short tube that fills most of the inner
diameter of the toroid. Distribute the secondary winding
evenly over the toroid. That would minimize the leak
inductance.

Go to the frequency of interest. Listen to a stable signal,
then see how much it changes when you add a 50 ohm load
in parallel to the receiver input (assuming the rx is 50 ohms)

If the signal drops by much more than 6 dB you need fewer turns
on the secondary, if the signal drops by much less than 6 dB,
you need more turns on the secondary.

In case you want more sensitivity you can tune the loop
with a capacitor in series with the primary winding.

If you have an impedance bridge, look at the impedance
seen by the receiver. It should be resistive at the frequency
where the capacitor resonates with the loop. Make it 50 ohms
by changing the turns ratio.

If you put flanges on the single turn to have just a very
narrow air gap for the secondary on three sides you can
reduce the leak inductance further which is good for
the bandwidth and makes the choice of ferrite material
even less critical.

A test: Check that the impedance on the secondary is
much higher than 50 ohms when the loop is disconnected.
(Measure or compute the inductance.)

73

Leif

Hi all.
1. is there a way to change mode (e.g. from USB to LSB) without going in the "Radio" collapsible section on the SDR# left pane?
I mean, some  shortcut keystroke...
2. is there a way to change the location of frequencies.xml ? I would like to put it in a folder visible from many instance of sdr#.
After googling a lot I could'nt find those answers... Thanks in advance.
Bye
Gio

p.s. if there is no simple way to do these... can sdr# team add these in the wishlist ?

On a quick glance I can see this person got it right. One other example I saw fed into the gates of J-Fets or something like that - high impedance. Going into common base amplifiers is good. And he's running them at high current, which makes the input impedance even lower.

{^_^}

A very good and cheap receiving loop antenna can be made from one turn coaxial cable with diameter of about 1,5 meter and gap in the shield in the middle of the antenna (at the top of the loop)

Why not rearrange the same one turn to achieve two turns with good shielding and balancing?
For the amplifier, I'm trying to keep things simple. I'll probably just rely on the low NF of the HF+ to exploit the small loop.

We had a 'somewhat' similar problem at LTO (Little Thompson Observatory:  <starkids.org>) in setting up our H1 receiver.  Our setup:

     1)  14-foot diameter off-center-fed dish

     2)  1 GHz HPF (to get rid of local RF noise)

     3)  LNA (nominal +22 dB gain, NF = 0.5 dB)

     4)  Bias tee

1 through 4 are at the feed which is a loop 1/4-wavelength in front of a splash plate.

     5)  75-feet 1/2-inch hardline

     6)  Another +20 dB LNA

     7)  bias tee
     8)  AirSpy

     9)  PC running Radio Sky Pipe (freeware at <http://radiosky.com/skypipeishere.html>)

We had to include the second LNA in the warm room (item 6) to have enough gain to get the 'signal' significantly out of the noise for RSP.  Suggest you add another gain stage in front of the AirSpy. 

Our system noise temp runs - measured - about 100 k.  We should do better, but the requirement of the 1 GHz HPF dictates that initial loss to get rid of RF dirt from a close-by cell tower. 

Dave - W0LEV

Our own setup for 21cm is:

0.85m offset dish-->circular-waveguide-feed-->20dB, 0.5dB NF LNA--->25m of RG6--->1400-1427MHz filter---->AirSpyMini

We produce live spectra every few minutes here:

http://www.ccera.ca/files/live_data/h1/live_spec.png

If I were going to be running more than 25m of RG6, I'd put a line-amplifier (doesn't need to be an LNA) in-line.  In fact, when we convert this to
  a "binocular" survey instrument, that's what we'll do, because those dishes are further away from our lab.

Have you considered placing the second LNA in the middle of your 75' feed rather than at the input for the Airspy? It might purchase a few easy more degrees of noise temperature.

{^_^}

The idea is to avoid any complex electronics and extra power supply.

Yes, that is possibly and I also tried this already by changing the amplification, but that would only zoom 10 dbFS.
So I know now, that it is not an easy bit changing, changing the config file would not help and have to look further, into other possible solutions.

Many thanks for all the help and ideas around.

Jochen

OK.  I understand.  Using my spectrum analyzers, there are many times I employ the 5 dB per major division, and, on occasion, even the 1 dB per major div.

While we're at it, using either the radio astronomy application aimed at the H1 line or the spectrum analyzer (SpectrumSpy), it would be extremely helpful if we had something as small as 10 kHz span and choice of resolution bandwidths.  Of course, there is always something more that would be 'absolutely' needed......

Personally, I have recommended the AirSpy to many customers (EMC and RFI) who do not have the resources to buy or rent  a full-up spectrum analyzer.  For the purpose of reducing radiated emissions, the AirSpy is just what is needed!  It's a wonderful piece of test gear if you don't have a spectrum analyzer (which I do).  But, this is getting off the subject.

Dave - W0LEV   

Jumping right into the middle of this thread.

The H1 "line" is usually only a MAX of about 2dB above the surrounding continuum noise, which itself corresponds to a sky temp of perhaps 10-20K
  at 21cm.

I usually set my display on my own apps to about 0.5dB/div, but of course, I also log all the data, so I can later inspect and process it at whatever level
  of detail I might desire.

--

--

Isn¹t the original also out of phase, input to output? Or is that part of
the design?

On 8/5/17, 1:06 PM, "Leif Asbrink" <airspy@groups.io on behalf of
leif@...> wrote:

The new drawing is quite different. It shows a simple
two turn loop with its midpoint grounded.

The old image was NOT a loop, try follow the yellow line
from one side of the transformer to the other side.
Impossible. Some connection is missing.

The way the transformer is drawn is misleading. The primary
winding should have few turns and the secondary winding
many turns if you connect to a 50 ohm system.

Look again. The loop has two turns in series with a center tap.
The loops swap between the inner and outer condictors of the coax.
I simplified the diagram a bit:

It is clear now:-)

73

Leif

The problem is that "bit" also affects SDR# and all the spectrum plugins, since it's based on the same reusable component.
The easiest trick is to increase/decrease the RF gain by 1 tick (3 dB) which will align your signal in the right region.

I only removed the other side of the shielding from the same diagram.
Now this build technique ensures the currents induced by the E-fields will be canceled at the grounding point, which is a desirable property.

Hi Dave

Thanks for your answer and the description of your setup. "prog" explained it right, I have not a problem with the signal strength, I would like to have the possibility to scale the signal strength in smaller steps than only 10 dbFS.
In the moment, I can also see the 1.5 dBFS lowered noise floor, due to the coronal hole facing to us. This is also the reason, why the level drops below -70 dbFS. You may have a look to my animation on youtube  https://youtu.be/4FC_6LQjpHA where you see the signal level over 24 hours.
My Setup is a 2 Meter Dish, a HB9BBD LNA with 40 dB gain and 0.19 dB NF. Than through 5 Meter low loss Koax Cable CS29 into my Airspy Mini. I agree, I do need a bit more gain but the next days, I will replace the Bi Quad by a Linear Horn or Septum Horn Feed.

So I hope, the developers could easily switch a bit and all is working ;-)

vy 73's de HB9DRT, Jochen

The new drawing is quite different. It shows a simple
two turn loop with its midpoint grounded.

The old image was NOT a loop, try follow the yellow line
from one side of the transformer to the other side.
Impossible. Some connection is missing.

The way the transformer is drawn is misleading. The primary
winding should have few turns and the secondary winding
many turns if you connect to a 50 ohm system.

Look again. The loop has two turns in series with a center tap.
The loops swap between the inner and outer condictors of the coax.
I simplified the diagram a bit:

It is clear now:-)

73

Leif