[Elecraft] Help with IF Noise, DSP Noise, NR settings

[Bob McGraw K4TAX]

David et al;

My statement is almost word for word from the Elecraft K3 manual.   And 
some from work done by Art Collins and company at Collins Radio.  They 
prescribed the function of a Noise Blanker as follows:

"The following operational requirements were kept in mind:

 1. Reduction of ignition noise from vehicles.
 2. Reduction of power line corona noise occurring at 120 CPS repetition
    rates.
 3. Reduction of local thunderstorm disturbances.
 4. And, in general, reduction of any man-made noise which is impulsive
    in nature.

Basically, all the above forms of noise interference are impulsive 
functions with repetition rates than can extend up to 100 KC in the case 
of the strokes in a thunderstorm."

Those are my sources.

Again both descriptions use repetitive pulse rates which are impulsive 
in nature.    In order for a NB to function efficiently it is best 
suited in a wide band signal path that is not restricted by filters.   
Hence a wide IF stage of the receiver before any filtering.

As to thunderstorms, since lightning contains many pulses in a single 
stroke, the NB is suited to minimize those pulses while at the same 
time, the bulk of the strike energy is affecting the receiver in other 
means. Namely AGC.  Many receivers suffer grossly from this phenomenon. 
   Fortunately Elecraft and Tentec took actions with their designs to 
minimize this phenomenon based  on the work of Rob Sherwood.    And 
regarding thunderstorms, there is a clear difference in the stroke and 
content of such for  a "local thunderstorm" as compared to the noise 
from distant thunderstorms several hundred miles away.   The distant 
thunderstorm is affected by propagation and may have several wave fronts 
with different arrival times where as a local thunderstorm only has a 
single wave front.   Hence the waveform is quite different and the means 
to suppress such will be different.


73

Bob, K4TAX


On 6/14/2019 12:24 PM, David Woolley wrote:
> The below statement surprised me.  That is not how noise blanking 
> normally works; normally it will suppress  any wide band pulse. The 
> typical arrangement uses a high bandwidth, low delay, path, to detect 
> the pulse.  It can then mute the signal path by the time the pulse 
> gets through the narrower filters on the main path. That will happen 
> with single, or randomly spaced spikes.
>