[Elecraft] Noise Blankers

[Bob McGraw K4TAX]

Depending the type of noise, for repetitive pulse type noise only,  the 
Noise Blanker does it job.  Now one must understand that all pulse noise 
is not alike.  Therefore, different values of NB, both with the  IF 
values and with DSP values, will require some adjustments.    If the 
noise is not repetitive then the NB is not the better choice but the NR 
is the better choice.  Again, depending on they type of noise, certain 
combinations of NR values will be necessary.

Yes, that is correct in that the Noise Reduction function is not 
available or can not be activated in the DATA mode.  However the Noise 
Blanker can be used in the DATA mode.   Again, the Noise Blanker is for 
repetitive pulse type noise.

I've spoken with many hams and find their understanding of Noise 
Reduction  being described as often does not work as expected. Further 
discussion seems to indicate they want NE........Noise Elimination.    
Well, NR........Noise Reduction does in fact work when the values are 
selected based on the type of broad noise spectrum being encountered.  
NR has very little effect on pulse type noise.

One fact not clearly understood, is that most hams operate with too much 
RF Gain.   Thus the correct application of ATTENUATION and RF Gain 
reduction will greatly improve receiver performance in the face of noise.

Optimizing receiver gain compared to band noise and the noise floor of 
the receiver is explained in a quote from Rob Sherwood, Sherwood 
Engineering:

"If receiver noise floor is 10 dB below band noise, the receiver is 
contributing less than 0.5 dB of the total noise.
Band noise varies by band over 30 dB, 160-10 meters. It also varies by 
direction and time of day, plus what the sun is doing. In an noisy urban 
environment it is anybody's guess as to your band noise level.
A simple test is to see how much the noise coming out your speaker 
increases when you switch between a dummy load and your antenna, when 
tuned to a dead spot on the band.
Example on 10 meters at my rural QTH, IC-756 Pro III: preamp OFF, noise 
goes up 3 dB. That means the receiver is contributing half the noise. 
Preamp 1 ON, band noise goes up 9.5 dB.   Almost all legacy receivers 
are designed for 10 meters, and attenuation is desirable on the low HF 
bands."

In Rob's first statement, how do we get the band noise to be 10 dB above 
receiver noise?  Use Attenuation and RF  Gain reduction. As an example, 
if receiver noise floor is -130 dBm and the band noise, no signal, is 
S-5 or -97 dBm, the difference being 33 dB. This would then indicate one 
should employ 15 dB of Attenuation and 8 dB of RF Gain reduction.  Or 10 
dB of Attenuation and some 13 dB of RF Gain reduction.   Of course the 
band noise will be comprised of different noise components, depending on 
band and many other factors as he suggests.

73

Bob, K4TAX


On 9/3/2017 3:06 PM, Keith Onishi wrote:
> In my experience, NR is much better than NB against power line noise.
> However, NR does not work on DATA mode. I use BHI Compact In-line DSP Noise Eliminating Module, which is inserted between K3 line out and USB audio module.
>
> For effectively eliminating noise, tuning RF gain and AGC level in addition to NB and/or NR would give you better results.
>
> de JH3SIF, Keith
>
>
>> 2017/09/04 4:38、Wes Stewart <wes_n7ws at triconet.org> のメール：
>>
>> I agree. The adjustments are far too many and arcane.  I'm currently plagued with powerline noise that has yet to be fixed, although they are supposed to be working on it. The K3S blanker is slightly better than useless. When I set it aggressive enough to be audibly useful, FT8 decodes multiple signals at 120 Hz intervals, which upsets sequencing.
>>
>> I'm finding that my SDR-IQ that I normally use only as a bandscope, has better noise blanking on its demodulated audio than the K3.  Too bad I can't use it with WSJT-X.
>>
>> I suspect, but do not know for sure, (maybe the designer can enlighten us) that there is insufficient delay in the (analog) signal path and the noise gets through before blanking takes place.
>>
>> Wes  N7WS
>>
>>
>>
>> On 9/3/2017 9:18 AM, K9MA wrote:
>>> I have never found the K3 noise blanker to be effective.  On the other hand, in side-by-side comparisons, the noise blanker in my old FT-1000D can be very effective on my ever present power line noise.  However, like all noise blankers of that design, it only works if there are no strong signals within many kHz of the operating frequency, so it's completely useless on a crowded band.  The only way around this I know of is to use a second receiver, tuned to a clear spot nearby, to control the noise blanker.
>>>
>>> BTW, the K3 noise blanker uses a filter with a bandwidth of about 100 kHz.  The idea is that, in such a wide bandwidth, the amplitude of the noise pulses should be large, making them easier to distinguish from signals.  (The FT-1000 uses something like 15 kHz.)  However, I've never been able to get the K3 NB to work, even on a completely dead band.  Under the same conditions, I can often get a 20 dB reduction with the other radio.  That I don't understand.  I should investigate that sometime.
>>>
>>> 73,
>>>
>>> Scott K9MA
>>>
>>> ----------
>>>
>>> Scott Ellington
>>>
>>>   
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:Elecraft at mailman.qth.net
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to jh3sif at sumaq.jp
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Elecraft at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to rmcgraw at blomand.net