We seek to understand and document all radio transmissions, legal and otherwise, as part of the radio listening hobby. We do not encourage any radio operations contrary to regulations. Always consult with the appropriate authorities if you have questions concerning what is permissable in your locale.

Author Topic: What's all this regenerative loop stuff?  (Read 693 times)

Offline jFarley

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 5746
    • View Profile
    • Email
What's all this regenerative loop stuff?
« on: April 20, 2018, 1311 UTC »
I have just verified the operation of a resonant NDB loop which uses positive feedback to enhance the Q which can be attained with commonly available real world components, namely, tuning diodes and Litz-wound ferrite rods.  A working prototype is up and running in the basement here and is showing great promise.

I have re-tooled my ferrite loop so that it is now tuned remotely by a pair of back to back NTE618 tuning diodes (4 diodes total) via a control signal of 0 - 15 VDC.  These diodes resonate a wound ferrite rod (160 turns of Litz) over the NDB band.  The loop output is amplified by a balanced JFET differential amp.  Tuning diodes work great with one caveat: they have a Q which increases with frequency, and vicey-versey.  At the top end of the band (around 400 kHz) they have enough Q to produce a reasonable but not great loaded Q of around 180.  I would really like this to be higher; Q = 250 would be acceptable .  At the low end of the band (around 200 kHz) the loaded Q is only around 90, and the loop, well, the loop sucks.  There is just too much loss due to series resistance in the loop tank, and this needs to be reduced for the loop to perform as desired.  One could use lower loss components in the tank, but this is not practical.  Instead, one can trick the loop into believing that it has lower losses.  If one replaces some of the energy burned up in the lossy tuning diodes and ferrite rod with some energy derived from the loop amp output, then the loop Q should increase.  The loop now becomes a regenerative loop which employs positive feedback for Q augmentation.

The best way to see this is through an understanding of what Q represents.  We have all probably seen that the Q of a resonant loop is the ratio of the loop's resonant frequency to the width of the response curve between its -3 dB points.  As the Q is increased, the loop tunes more sharply, and its -3 dB points move in towards the center frequency.  This is really a result of Q and is not a formal definition of Q.  In electrical engineering, the Q of an LC circuit is defined as the ratio of the energy stored in the L and the C per cycle to the energy dissipated in losses per cycles.   If one can replace some or all of the losses dissipated as heat, then the loop Q must increase.   As one adds controlled positive feedback to the LC tank, one is in effect adding negative resistance to the loss elements in the LC tank.

In practice, I take the output of the loop amp and split it using a toroidal splitter.  One port goes to the coax back to the SDR.  The other port is attenuated in a 75 ohm 20 dB attenuator and passed to a variable attenuator comprised of a PIN diode and load resistor.  This voltage is then applied to a one turn tickler winding on the ferrite rod.  The biasing of the PIN diode is set up so that a 0.7 - 5V control signal will cause a current of 0 to 150 uA to flow through the PIN diode, thus changing its effective AC series resistance.

I have observed 3 regions of operation.  Below a PIN diode current of 1 uA, the loop behaves like the underlying loop without regeneration, and has a generally depressed Q due to the Q of the NTE618 diodes.  If the PIN diode current is increased to a value of typically 100 uA, the loop breaks into oscillation (as it should) and the SDR waterfall goes bat-guano crazy.  Between these 2 points, one can observe the loop Q increasing smoothly above the default level with increasing PIN diode current.  The response in the spectrum pane of the SDR begins to peak higher, and the response skirts narrow.  The object viewed near the resonant peak (carrier or offset) in the waterfall begins to brighten with increasing PIN diode current.  I can typically see an increase of up to 20 dBm in this middle region with increasing PIN diode current before it breaks into oscillation.

The regeneration process seems to be well behaved and stable as the Q operating point is varied, and this bodes well for the future. I am not anticipating running the loop at insanely high Qs such as 10000; I simply would like to enhance the Q of a deficient loop to around 250 which experience has shown is about what is required to make the loop competitive with the amplified whip which I also employ.

Interesting sidebar: if the sense of the tickler winding is reversed - negative feedback - then the regeneration will produce increasing attenuation of received signals as the PIN diode current is increased.

What I now have is a resonant NDB loop with software programmable (and knobless) control of:
    - The loop azimuthal position, with an accuracy of better than 1 degree.
    - The loop resonant frequency with a resolution of typically 100 Hz.
    - The loop Q and thus its Effective Height.

Further details will be posted after it is sited outdoors and fully shaken down.

Joe Farley, Near Chicago
SDR-IQ / R8 / R7
Remote Resonant Loops for HF and LF / ALA 1530
Active 60" Whip / PA0RDT
QSLS appreciated to:    jfarley44@att.net

Offline ChrisSmolinski

  • Administrator
  • Marconi Class DXer
  • *****
  • Posts: 22093
    • View Profile
    • Black Cat Systems
Re: What's all this regenerative loop stuff?
« Reply #1 on: April 20, 2018, 1440 UTC »
What I now have is a resonant NDB loop with software programmable (and knobless) control of:
    - The loop azimuthal position, with an accuracy of better than 1 degree.
    - The loop resonant frequency with a resolution of typically 100 Hz.
    - The loop Q and thus its Effective Height.

Further details will be posted after it is sited outdoors and fully shaken down.

This indeed looks very promising.  I have several antenna projects queued up for summer already (assuming winter ever ends), but can certainly add one more to the list.
Chris Smolinski
Westminster, MD
eQSLs appreciated! csmolinski@blackcatsystems.com
NRD 545 / netSDR / AFE822x / AirSpy HF+ / KiwiSDR / 670 ft horizontal loop / 500 ft northeast beverage / 270 ft west-south-west beverage / 300 ft south beverage / 43m / 20m / 10m  dipoles / Crossed Parallel Loop

Offline jFarley

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 5746
    • View Profile
    • Email
Re: What's all this regenerative loop stuff?
« Reply #2 on: April 20, 2018, 1523 UTC »
This is working out better than anticipated, and I am shocked how well the regen loop seems to behave.  I really expected the worst, that being an unstable hard to control beast.  Not so, apparently.  I am also happy to have most of this work done by January 110th, leaving a lot of time for summer testing.

So far, testing has been done in the noisy confines of my basement, but reception of a couple of NDB locals and the Mequon DGPS has been very promising. 

I have been thinking about other uses.  The resonant loop I use for SW tends to have a Q of around 270 in the funny band.  A couple of years ago I built a higher Q version from copper tubing and a small gap silver plated cap which was giving me a Q of around 600: a BW of around 11 kHz.  Not enough to split close stations, but a very good effective height.  I had planned to try regen on that guy but never got around to it.  Now I am wondering if regeneration would work on the loop I use currently.

 
« Last Edit: April 20, 2018, 1536 UTC by jFarley »
Joe Farley, Near Chicago
SDR-IQ / R8 / R7
Remote Resonant Loops for HF and LF / ALA 1530
Active 60" Whip / PA0RDT
QSLS appreciated to:    jfarley44@att.net

Offline ChrisSmolinski

  • Administrator
  • Marconi Class DXer
  • *****
  • Posts: 22093
    • View Profile
    • Black Cat Systems
Re: What's all this regenerative loop stuff?
« Reply #3 on: April 20, 2018, 1553 UTC »
So far, testing has been done in the noisy confines of my basement, but reception of a couple of NDB locals and the Mequon DGPS has been very promising. 

Listening to DGPS stations??? Sure, you say just for testing purposes... but that is a slippery slope!  ;D

Quote
I have been thinking about other uses.  The resonant loop I use for SW tends to have a Q of around 270 in the funny band.  A couple of years ago I built a higher Q version from copper tubing and a small gap silver plated cap which was giving me a Q of around 600: a BW of around 11 kHz.  Not enough to split close stations, but a very good effective height.  I had planned to try regen on that guy but never got around to it.  Now I am wondering if regeneration would work on the loop I use currently.

For my uses, something resonant around the upper end of the MW band would be interesting, for trying to find pirates in the TIS infested haystack.
Chris Smolinski
Westminster, MD
eQSLs appreciated! csmolinski@blackcatsystems.com
NRD 545 / netSDR / AFE822x / AirSpy HF+ / KiwiSDR / 670 ft horizontal loop / 500 ft northeast beverage / 270 ft west-south-west beverage / 300 ft south beverage / 43m / 20m / 10m  dipoles / Crossed Parallel Loop

Offline Josh

  • DX Legend
  • ******
  • Posts: 2534
    • View Profile
Re: What's all this regenerative loop stuff?
« Reply #4 on: April 20, 2018, 1721 UTC »
Can we use it to transmit?
Conveniently located near Vincennes Indiana.

Offline jFarley

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 5746
    • View Profile
    • Email
Re: What's all this regenerative loop stuff?
« Reply #5 on: April 20, 2018, 1750 UTC »
I'm sure it's spewing all kinds of junk when it oscillates!
Joe Farley, Near Chicago
SDR-IQ / R8 / R7
Remote Resonant Loops for HF and LF / ALA 1530
Active 60" Whip / PA0RDT
QSLS appreciated to:    jfarley44@att.net

Offline jFarley

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 5746
    • View Profile
    • Email
Re: What's all this regenerative loop stuff?
« Reply #6 on: July 16, 2018, 1236 UTC »
Doing some testing with the loop deployed outdoors.  I grabbed a couple of screen shots which exhibit how the loop Q can be modified remotely to enhance reception characteristics.  The first shows the loop Q with very little regeneration, and this is more or less the native Q of the antenna at this particular frequency.

https://imgur.com/oyEjjvR

The 2nd image represents a moderate amount of Q, and is the Q with which I would most likely find most useful.

https://imgur.com/a/CzDDCdO

The 3rd image represents a rather extreme Q level.  Whether of not this is useful remains to be seen when things quiet down and propagation returns on the LW band.

https://imgur.com/a/EjU6ANr

I have added a a wideband noise genny to the loop head, and all 3 images were capped with the noise genny enabled.  The response peaking is clearly seen in the FFT panel, as well as a brightening of the response in the waterfall.

Early indications seem to suggest that the regen loop can usually be adjusted to give greater sensitivity than the amplified whip which is being used for comparison.  This whip has been the workhorse for me as I have used it extensively for SDR recordings and has produced NDB logs from 30 countries in the past 2 seasons.

 
« Last Edit: July 16, 2018, 1258 UTC by jFarley »
Joe Farley, Near Chicago
SDR-IQ / R8 / R7
Remote Resonant Loops for HF and LF / ALA 1530
Active 60" Whip / PA0RDT
QSLS appreciated to:    jfarley44@att.net

Offline jFarley

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 5746
    • View Profile
    • Email
Re: What's all this regenerative loop stuff?
« Reply #7 on: July 16, 2018, 1344 UTC »
A few comments about how the loop Q is controlled...

https://imgur.com/a/3eDNM2y

The above image is a cap of the main form of the app which controls the loop and rotator.  This app has a number of tab pages to organize common tasks, and the RNA tab page is currently displayed below a collapsible pane.  This pane can be collapsed to display the current loop resonant frequency and rotator bearing only, or can be expanded to expose some controls to directly manipulate the resonant frequency and loop Q.  The control with the black background is the Q control and its functionality somewhat resembles a scroll bar.  Clicking anywhere within the bar will send a value (  0 to 255 ) to adjust the loop Q.  This control will also respond to mouse wheel events.  At any level, the current value is displayed in the text box to the right of the Q control.  The vertical yellow bar represents the value above which there is actually a regenerative state in the loop.

The bottom part of the screen displays the current RNA database in a ListView control.  This is an owner drawn control, and colors are used to add extra info to enhance the DXperience.  In this scenario, domestic NDBs are rendered with a white background, Canadians with a red BG, and Foreign with a green BG.  The selected item is rendered with a blue BG. Double clicking on any item in the database will tune the loop to that frequency,  and spin the rotator to the proper bearing.  Note that in the info panel that the bearing for the currently selected NDB is ~270 degrees relative to Chicago, whereas the rotator has been spun to 88 degrees.  This app can use "Smart Rotation" and employs the fact that the loop response pattern is a symetrical Figure 8 pattern.  This minimizes the amount of work (and rotation delay) required to get to the assumed proper azimuth.

« Last Edit: July 16, 2018, 1351 UTC by jFarley »
Joe Farley, Near Chicago
SDR-IQ / R8 / R7
Remote Resonant Loops for HF and LF / ALA 1530
Active 60" Whip / PA0RDT
QSLS appreciated to:    jfarley44@att.net