Topband: Noise blanker fun

[Tree]

During the first morning of the Stew, someone's electric fence was making
lots of noise.  I decided that something had to be done - and given the 
fun I have had dealing with my neighbors with electric fences - I thought
I would try a different approach.

I have two TS850s - and have often thought about using the noise blanker
in one - tuned outside the band - to trigger the noise blanker in the 
second radio.  I did some looking at the schematics and discovered that
Kenwood had made this really easy to do.  They have two different noise
blanker circuits - each generates a positive going pulse when they detect
a noise event.  The signals from each are simply wired OR'd together and
then drives the base of a transistor on the RF board that will turn off
the receiver as requested.  

All I had to do was connect this signal to the same place on the other 
receiver and I could use the noise blanker from one radio on the other
receiver.  I added a phono jack to the back of both of my radios and 
rand some RG174 to the signals and RG8X between the rigs.

Around noon, I had the rigs modified and was ready for a test.  It 
appeared to work wonderfully. 

Well - that was - until strong signals started showing up on the band.

The problem I ran into was easy to understand once I figured it out.  

The receiver with the NB turned on obviously didn't see any of these
stong signals.  It just kept creating pulses that were in sync with
the electric fences (there were probably at least 3 of them creating
noise).  By sending these to the other radio - I was essentially 
modulating the signal with the pulse - which has a very fast attack.

When you detect a noise pulse in a receiver - you want to turn off the
receiver as quickly as possible - before it can react to the pulse.  
However, this edge is probably worse than an unmodified FT1000MP as far
as producing a click goes.  However, instead of there being one click at
the start of the dit or dah - there are a bunch of clicks during the 
dit or dah.

This quickly turned into a real mess on the band.  I must apologize to
KI7Y, who I though had a transmitter problem as he was the first loud
signal on the band.  When I heard a second signal doing the same thing,
I finally figured out what the root cause was.

I guess in order for this method to be effective, it would require 
some kind of signal conditioning to round off the leading edge of
the pulse.  I do see that Kenwood is doing that on the trailing edge
which is a good thing.  However, the problem with doing it on the 
leading edge is that time works against you.

That is - unless you have some amount of delay in your receiver that
you can use to your advantage.  

There may, or may not, be a common solution to being able to make the
receiver turn off and back on "nicely" so that clicks are not generated - 
AND make the the pulse short enough that it doesn't affect the ability
to copy the signal.  It will still be a comprimise in any case - and 
maybe the best solution is to keep working on finding the electric 
fences.

I did have my last neighbor respond very well - and he even replaced the
fabric "wire" he was using with real wire to solve the problem.

I was lucky and the noise seemed to be less of an issue the second night
and also my trusty R4-C was asked to help out.  It seems to do a better
job dealing with the noise than the 850 does.  This is probably because 
of the 600 Hz filter before the NB.  It does impare the ability of the 
NB to react to the noise - but it still works enough to help and not
be affected by signals that are more than a kHz or two away.

I also wondered if a DSP would be a better method for reducing this type
of noise - were it would be easy to add enough delay to allow you to could
prepare yourself for the noise event.  Maybe if people have had great 
success with some of the DSP units with this type of noise, they can 
share their observations.

73 Tree N6TR / K7RAT