Sunday, December 23, 2018

WQ6X SOUND PROCESSING: SOUNDING OFF about SOUND

I C O M  R - 7 1 A + M F J  7 5 2 - B / 1 0 2 0 / 1 0 2 6

 
When you think about it, radio, and radio communication is ALL about sound; yet amazingly,
this is an area all too frequently overlooked, even with some of the "name brand" Super Stations across the country (you know who you are).  Amateurs and contest groups spend10's & 100's of thousands of dollars on more-gain antennas at 200' to pull the weakest signals into the receiver and super-KW amplifiers (powered by buffalo-driven treadmill generators) to be heard across the galaxy.   They follow all that with artifact-creating DSP signal shaping with horriblly filtered or distorted audio; obliterating proper copy of that weak signal they spent $1,000's of dollar$ to bring into the receiver's front-end.


Over the decades there have been evolving solutions to the poor audio
and noisy-audio "problem". 

As far back as the early 1950's, National's Select-O-Ject audio Q-multiplier provided significant audio improvements.  You may also remember the I-F based Q-multipliers, like the Heathkit QF-1 & GD-125 units.

In the 1970's, a number of external audio filters were produced by various manufacturers; my favorites of course being the Autek QF-1A and the MFJ-752 models.


When we fast forward to the mid-80's the size-reduction in CMOS devices and DSP chips heralded
a clear path for DSP units to appear in amateur radio; first in the AF-world, eventually working its way toward the IF (and even RF) circuits.  Be these circuits analog or digital, any attempt at "processing"  a signal (be it RF, IF or AF) we run the risk of artifact (I.e. some unintended alteration or distortion of the signal).

Noise-blanking is a perfect example of artifact introduction.  With many transceivers in the last 20 years the NB1/NB2 style of noise-blanking often exhibit signal "distortion" when turned past the 10 O'clock position (or thereabouts).  Surprisingly, the early 80's noise-blanking circuits are amazingly effective with virtually NO artifact; one of the reasons I loved my ICOM 740 "back in the day".



During sailboat trips from L.A. to Catalina island and back, running the 740 into a vertical wire up the main mast made the radio susceptible to QRN from distant lightrning storms.  In most cases the 740 NB blanker rooted the noise out completely.



Today, my vintage Icom R-71A receiver being from that era enjoys the same noise blanking success; who knows,
it may well be a near-identical circuit. 

When my current R71-A makes its way into "daily" service, it will be matched with an MFJ 1020 (Active Antenna),
an MFJ-752C, and of course, my favorite, an Autek QF-1A; all currently sitting on the auxiliary backup shelf.

Preliminary tests of the R-71A and the 752-C/QF-1A show them to be quite compatible, the same as with my FT-1000mp (@W7AYT).  Bottom-line: "audio out" is audio out, be the source an Icom R-71A, an ICOM-7000 or an FT-1000mp; it ALL processes the same.



You may remember one of the earliest noise-blanking devices - AEA's "Moscow Muffler" from the early 1980's. 

What is AMAZING about the Moscow Muffler is its design for operation in the FRONT-END of the radio/receiver; not the A-F or even I-F for that matter (as most noise blankers currently do). 

There is a 6db receiver preamp as well as dual blanking widths to knock rhythmic noise pulses out of the front-end BEFORE those pulses have an opportunity to overload the receiver's pre-amp and mixer stages.

Because the signal "processing" is done PRIOR to the Ant-IN socket, no "modification" to the equipment is needed; as is the case with IF-based NB's and Q-MULT's.  It recently occurred to
me that the WB-1, MFJ-1026 and the JPS ANC-4 "noise canceller" units may well be operating on different but co-operative approaches to removing the noise PRIOR to the radio - Absolutely GENIUS!

I've extoled the virtues of "Stereo CW" for shifting the audio "inside my head"
as an aspect of audio-QRM reduction.  ([CLICK HERE] to re-read that.)

I have learned to incorporate Q-Filtering into my SO2-V operations from W7AYT.  This is what I said about it in this last summer.

And then of course my favorite: Q-Filtering for Fun and Profit: Part-1 and Part-2.

Most of todays HF amateur transceivers are so complex and loaded with convoluted CMOS CPUs there is little we AMATEURS can do (internally) to improve radio performance.  However we DO have control over what comes IN to the radio and what goes OUT the audio exit lines.

What do YOU DO  to improve signal intelligibility?

Inquiring modems want to KNOW!


No comments:

Post a Comment