Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Seeking techniques for receiving CW (non-computer) improvements  (Read 16491 times)
AK0B
Member

Posts: 275




Ignore
« on: November 25, 2014, 10:58:31 PM »

I have been searching without success so far.  For articles, designs, etc for receiving CW that would be an improvement over basic ham receiver design that has been around since the dawn of radio.

I would like to learn about techniques and methods that assist the human ear.

Anyone know of anything -- Huh  I would think someone would have done research on hardware design techniques.

Let me know if you know of ideas, articles, etc.

thanks,   Stan AK0B
 
Logged
N3QE
Member

Posts: 5021




Ignore
« Reply #1 on: November 26, 2014, 03:19:47 AM »

Broadly there are two schools of thought:

1: Narrow narrow brickwall filtering. Problem is, many steep sided filters introduce truly wacked out group delay that causes audible ringing. Extended use of narrow brickwall filters is for me, acutely uncomfortable.

2: As little filtering as possible. Having no filtering means no wacked out group delay and the brain gets to exercise its skills using all information - including, importantly, the original phase information - to reconstruct the signal. This is an extremely enjoyable way for me to listen to CW at or below the noise level.

#2 is commonly used by weak-signal guys who are trying to pick signals out of background (in some cases only thermal!) noise. #1 is commonly used by contesters in tough QRM conditions.

There is a happy #3: A narrow filter but carefully designed so that the group delay happens outside the passband. This way the nasty group delay ringing phenomenon happens outside the passband. One common catchphrase for these is "gaussian" or "transitional" filters.

The #3 type filter has become increasingly common in DSP equipment past few years. The very first generations of DSP equipment went for brick wall filters that sounded like crap. More recent rigs with recent DSP does much better because they actually pay attention to reduce group delay variations in the passband.

The #3 type filter can also be done in crystal filter or in outboard audio filtering. Many of the common outboard audio LC filters that CW enthusiasts have published in past 20-30 years do a very good job moving the group delay nasties outside the passband and can sound very nice.

I also like to disable AGC wherever possible (or set it to be "ear-saving only") so that loud signals are loud and soft signals are soft. Most rigs let you achieve this by backing down on RF gain and turning up AF gain.
« Last Edit: November 26, 2014, 03:25:00 AM by N3QE » Logged
G4CMY
Member

Posts: 19




Ignore
« Reply #2 on: November 26, 2014, 04:40:57 AM »

One technique that was suggested many years ago in the G-QRP-Club magazine was to generate pseudo-stereo with lowest audio fequencies in one ear and highest frequencies in the other ear. So the audoio pass band is spread across the stereo image and the signals move from one side to the other as you tune through them.

This moves the QRM spatially and helps you to concentrate on the signal you are interested in.

It sounds great but I've not heard anything about it since then.
Logged
NI0C
Member

Posts: 2947




Ignore
« Reply #3 on: November 26, 2014, 04:54:18 AM »

All good comments from N3QE above.  During the past couple of years, I've been using the 700 Hz crystal filter in my K3 a lot (as opposed to the narrower ones I have available: 400, 250, 200),  I have my DSP filters setup to more or less match the crystal roofing filters.  As Tim points out, sometimes "less" (selectivity) is often better for copying signals at the noise level.

I'm not sure about other transceivers, but the K3 DSP uses finite-impule-response filters (FIR) in all but the two narrowest bandwidths; infinite-impule-response (IIR) filters are employed for the latter.  The FIR filters have virtually perfect linear phase characteristics (constant group delay) in the passband, thus avoiding the problems that Tim mentions.

Another very useful tool is the "Audio Peak Filter," (APF), that was or is employed in some Yaesu (and/or Kenwood) rigs, and possibly some outboard audio filter units.  After the K3 came out, Elecraft's DSP designer actually emulated this filter in the K3 DSP in what was for me one of the most important firmware upgrades.  This filter gives a final "boost" of very narrow selectivity that I have found very helpful in marginal copy situations. 

A few years ago, I ran a frequency response on the APF in my K3.  If I can figure out how to copy and paste pictures in this forum, I will share that.

73,
Chuck  NI0C 

Logged
9V1VV
Member

Posts: 57




Ignore
« Reply #4 on: November 26, 2014, 06:00:24 AM »

Something old and trusted - the resonant speaker. Quite a few guys are doing this now, including me. You can make a tuned speaker using a plastic tumbler or cup, move a small 1.5 or 2 inch cheap speaker  facing down at the opening until you hit the sweet spot. Then mount it in that position. Depending on the size of the tumbler you can get from 400-700 Hz resonance. Weak signals just leap out of the QRN.

Do a google on resonant cw speakers and you will find lots.

Average price $ 4 each.

I use the Elecraft K3 which has "quasi" stereo, and use two,   made of blue plastic tumblers and $3 dollar speakers.

Attenuation above and below resonance is pretty steep curved, but there is no "ringing" such as you get with expensive crystal filters.

Cheers

Logged
W1VT
Member

Posts: 2585




Ignore
« Reply #5 on: November 26, 2014, 06:44:17 AM »

Gaussian crystal filters are rare, but it is possible to design your own.  

I've used Zverev's Handbook of Filter Synthesis, along with a Wes Hayward computer design program, to design a narrow 4 pole CW filter with no ringing.  I even have a scope photo somewhere that shows what it looks like when you run a square wave through it.  It has been a long time since you could buy film for scope cameras.

As I recall, the design used crystals with a specific frequency offset--I measured a lot of crystals and was able to find ones that would allow the construction of the filter.  Not exactly reproducible, but I wasn't worried about being able to make another one.

Zack W1VT
« Last Edit: November 26, 2014, 06:47:21 AM by W1VT » Logged
AK0B
Member

Posts: 275




Ignore
« Reply #6 on: November 26, 2014, 11:52:33 AM »

Thanks for the comments. Gave me a few more things I can use when I google.

I have used the resonant speaker technique 9V1VV talked about. I have never tried tuning one slightly higher and one slightly lower in frequency. 

Hopefully, a few more will have also tried different things.  We need a smart DSP we could tune while listening, maybe??

stan ak0b
Logged
NI0C
Member

Posts: 2947




Ignore
« Reply #7 on: November 26, 2014, 12:31:27 PM »

Re: my previous post regarding the K3 APF--

I was not able to paste in the frequency response plot here (apparently images need to be first stored on a website, which I have not done.)  However, Barry, N1EU, has a really good web page on the subject, which includes a frequency response plot:  http://n1eu.com/K3/K3_apf.htm

His plot looks similar to mine, even though we probably used different techniques to obtain them.

73,
Chuck  NI0C

 
Logged
VE3WMB
Member

Posts: 325




Ignore
« Reply #8 on: November 28, 2014, 04:52:38 PM »

What you are describing sounds like Binaural reception.

The idea is to use a wider filter (at least 600 Hz) and distribute the audio signals between the two ears according to frequency, creating sort of an audio 3D effect. With PowerSDR my Flex-1500 can be switched to provide binaural CW reception and it can be quite useful at times to help separate multiple stations, especially if you don't want to use narrow filtering. It literally feels like the signals are in different places in your head.

Michael VE3WMB


One technique that was suggested many years ago in the G-QRP-Club magazine was to generate pseudo-stereo with lowest audio fequencies in one ear and highest frequencies in the other ear. So the audoio pass band is spread across the stereo image and the signals move from one side to the other as you tune through them.

This moves the QRM spatially and helps you to concentrate on the signal you are interested in.

It sounds great but I've not heard anything about it since then.
Logged
K7KBN
Member

Posts: 3490




Ignore
« Reply #9 on: November 29, 2014, 11:43:48 AM »

QST, April 1976 -- good article about a VERY narrow audio filter using a harmonica reed.  Mentioned also in an eHam discussion of the subject in 2006.
Logged

73
Pat K7KBN
CWO4 USNR Ret.
KF5KWO
Member

Posts: 55




Ignore
« Reply #10 on: December 01, 2014, 01:23:58 AM »

I'm not a CW guy, but this sounds fascinating.  Anyone use it?  Does it work?

One technique that was suggested many years ago in the G-QRP-Club magazine was to generate pseudo-stereo with lowest audio fequencies in one ear and highest frequencies in the other ear. So the audoio pass band is spread across the stereo image and the signals move from one side to the other as you tune through them.

This moves the QRM spatially and helps you to concentrate on the signal you are interested in.

It sounds great but I've not heard anything about it since then.
Logged
KH6AQ
Member

Posts: 7718




Ignore
« Reply #11 on: December 01, 2014, 11:16:22 AM »

The Elecraft KX3 has several CW audio effects features including the one mentioned in the previous post. The KX3 manual says it "maps signals from left to right to left according to pitch." I find that mode just about impossible to listen to.

The simulated stereo mode creates "the illusion of greater acoustic space" and I run that 100% of the time.

Another great feature is the Audio Peaking Filter for weak CW signals. It pulls weak signals out of the noise and is a lot like a feature in the TS-570D.

And being able to dial the filter bandwidth down to 50 Hz with no ringing and invoking the DSP noise reduction both contribute to very comfortable CW copy.  
« Last Edit: December 01, 2014, 11:19:45 AM by WX7G » Logged
KC8FQD
Member

Posts: 27




Ignore
« Reply #12 on: December 01, 2014, 12:40:17 PM »

Gaussian crystal filters are rare, but it is possible to design your own.  

I've used Zverev's Handbook of Filter Synthesis, along with a Wes Hayward computer design program, to design a narrow 4 pole CW filter with no ringing.  I even have a scope photo somewhere that shows what it looks like when you run a square wave through it.  It has been a long time since you could buy film for scope cameras.

As I recall, the design used crystals with a specific frequency offset--I measured a lot of crystals and was able to find ones that would allow the construction of the filter.  Not exactly reproducible, but I wasn't worried about being able to make another one.

Zack W1VT

I just did a five-pole, 500Hz, Gaussian-to-6db filter with a random batch of off-the-shelf crystals. All of the crystals have the same nominal frequency. The filter shape and width are determined by the coupling reactances and terminating impedances. The crystals should be reasonably well matched, which these were right out of the box (to my delight).

Zverev is a great resource, but I found the section on this subject (8.17) a bit puzzling in spots. The section seems to be based mainly (or entirely) on a paper by Dishal (citation below). It's available from the IEEE and, in my opinion, is easier to follow. Also, where Zverev provides formulas only for the upper sideband case, Dishal has formulas for the lower sideband case as well, which is often the preferred realization.

M. Dishal, "Modern Network Theory Design of Single-Sideband Crystal Ladder Filters," Proc. IEEE, Vol. 53, pp. 1205-1216, Sep. 1965.

Don't let the name fool you. They're called SSB filters because they're asymmetrical and great for that application, but they're an easily realized type of crystal filter and often used for other applications as well.

As an aside, I'd love to know what the transfer functions are for the Gaussian-to-6dB and other transitional filters and where Zverev came up with them. He provides charts and the k and q values, but that seems to be it. I tried to chase clues from the references section, but came away unsatisfied.
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!