Some ears work better at copying below the noise level than others.
W5WXV didn't do EME but he did long range terrestrial 2m CW contacts
regularly and in the lab he claimed he could detect a signal 20 dB below
the noise level. I know I've been in his shack while he was rag chewing
and I listened on the same speaker and heard nothing to copy. One time
at a CSVHF conference in Cedar Rapids, WB0TEM was demonstrating his
portable 1296 EME chatting on CW with K2UYH a big signal on the band.
Mark acted like he copied solid, I copied about half and many listened a
while and walked away shaking their heads, they copied nothing.
Ringing was a problem in the S-line vintage mechanical filters. They
would turn lightning into a craaaash while my old Q5er aligned to have a
pretty much gaussian passband heard clicks on the same antenna. The
vintage Tentec filters are good that way. They don't have as steep
skirts and as sharp corners as those vintage Collins filters, but that's
why they don't ring and why lightning is a click instead of a craaaash.
The time response of a filter is important, and wasn't always considered.
There is considerable EME still done on CW, some have the view that the
computer modes sometimes make contacts with the transmitter turned off.
5UN also made up for ringing with excess antenna and preamp gain. The
DSP audio filters are better narrow than most analog about ringing from
transients and keyed signals.
I had a 400Hz Inrad filter in a couple Kenwoods and it was ringing with
CW above about 30 wpm.
Hardware noise blankers needed 100 kHz or more to truly discriminate
against ignition noise and power line noise compared to desired signals.
The narrower the filter the longer a pulse and the closer it is to the
characteristics of desired signals and so its harder to discriminate
between signal and pulse noise. The pulse detection has to be down
before the ultimate bandwidth filtering in either case.
73, Jerry, K0CQ
On 2/26/2011 5:44 PM, Rsoifer@aol.com wrote:
Jerry,
Sounds reasonable to me. I have no data on the effect of fatigue. I do
know that in the good old days (1980s-1990s), when men were men and EME was
on CW, some of the best EME operators, like VE7BQH, kept their rx
bandwidths at 3 kHz or more, to minimize ringing. Others, like W5UN, kept
theirs
narrow. Dave used a QF-1A. I used a QF-1. Different ops, different
strokes.
On another subject, I've found that the O II NB works best on 160 with a
roofing filter of 6 kHz or more, regardless of where the DSP BW is set.
Seems counter-intuitive, but there it is. Tnx to k8IA for tipping me off
about it.
73 Ray W2RS
In a message dated 2/26/2011 9:16:59 P.M. GMT Standard Time,
geraldj@weather.net writes:
On 2/26/2011 12:52 PM, Rsoifer@aol.com wrote:
Jerry, Lee, and others,
It may be useful to draw a distinction between digging weak signals out
of
the noise and improving the SNR on stronger signals so they sound better.
As we know, most of the intelligence in (male) human speech is below
about
2400 Hz. The human ear is very good at disregarding higher
frequencies, so
passing the signal through a low-pass filter will make it sound better
but, for most good operators, won't make much of a difference in their
ability to dig it out of the noise.
When the operator is fresh, yes. After several hours the operator can
become fatigued and then needs all the help the hardware can give. But
the brain extraction of weak signals from noise is sort of a correlation
process and if the noise bandwidth is too narrow, just like correlation
noise reduction in a DSP it works less well with narrow band noise.
Same for CW. One year at FD at our club station the CW rig was a TS-430
owned by a ships sparks bought overseas with a factory narrow CW filter.
That radio didn't seem to have the option of selecting the filter or not
for CW and so it was always in the circuit. So I couldn't switch to a
wider filter and that filter rang enough on noise and was narrow enough
the noise had a pitch to it, so copying CW I had a constant tone to
discriminate against which wore me out in less than 4 hours of operating.
In my FT-857D, I've found the audio DSP CW filters do nothing to improve
S/N of a CW signal below the noise level or to make it easier for my
brain processing to do it, but the Collins mechanical CW filter does
improve the S/N of a CW signal below the noise. The difference between
copying and not copying on long VHF paths.
73, Jerry, K0CQ
73 Ray W2RS
In a message dated 2/26/2011 6:16:09 P.M. GMT Standard Time,
geraldj@weather.net writes:
That's where I find my passive speaker filter shines. It passes no audio
section noise and no IF noise, an few DSP HF artifacts.
A fundamental of receiver design is that selectivity works best as close
to the antenna as possible. Unfortunately that ignores the noise
contributions of all the stages after that. The typical product detector
is double sideband so the IF noise of the image is there along with the
signal and the RF noise that passed through the filter plus the same
sideband noise much wider than the filter that was up front. Receivers
would benefit from having a SSB filter at the product detector, but I
know of only one design that way, called the Hohentweil, a 2m
transverter kit. Then they would benefit from making the audio output
stage, often essentially a power op amp into an active low pass filter.
In tube receivers a simple capacitor from audio output tube plate to
ground combined with the tube and the audio output transformer to make a
rudimentary low pass filter. In the 75S-3B, it was effective enough to
make using 2125/2975 tones for 850 shift RTTY (and for all recorded
history, the standard tones for 850 shift RTTY due to an AT&T standard)
difficult until the capacitor was removed from the circuit.
73, Jerry, K0CQ
On 2/26/2011 11:57 AM, kc9cdt@aol.com wrote:
I think one of the reasons the Drake R-4B, Hallicrafters SX-117 and
many others are beter in a noisy condition is simply they do not have
all the high frequency respnse in the audio, or maybe it is the tube
amp??. I wish there was a HF cutoff on the OII, full EQ like Bob Heil
recommended day one to TT way back may have helped.
Interestng...last nght, on 40 I was working a really nice guy in St
Kitt. There was quite a lot of QRN, He was just above the noise floor
I found that if I used the old Hallicrafers SX-117 to receive
him...copy was more clear!!!!
OMG, Maybe we need to go back to the older stuff (I have both) Unless
of course it is contesting at a high level...where you need lot of
speed
I use the Collins S line& KWM-2
Drake C line (all Sherwood mods)
Halli SX-117/HT-44
Halli SX-115/HT-32B
Along with the OII of course.....
73,
Lee
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|