A way to push back the hiss?

[Tom Wells]

I was thinking about a humorous imaginary product I could design a panel for, with lots of clever button and switches
intended to create a "quiet" zone in a station's listening area if they were being interfered with as Bob Savage's WYSL.
I was going to call it Dieboc Technologies AM-Q1 and just try to make it a funny thing.
But then, of course I got to thinking like an engineer.
Of course it's not "legal" to have sidebands as I'm about to describe, but then some of us feel there's nothing
"legal" about the iboc sidebands, either.
So if we have a blank carrier , let's say it's 1040, and we are getting creamed by iboc, what would happen if we then put clean
strong empty carriers ON adjacent and 2nd adjacent frequencies? Well, there'd be the 10khz whistle, but besides that there wouldn't there
be a reduced amplitude of the offensive hiss? Similarly, the 2nd adjacent would be an audio tone at 20khz, but wouldn't
be audible to almost anyone, while providing "quieting" over the hiss from that channel.
Now, it would be trouble to put 4 more "carriers" on the air, if we consider we'd need new transmitters, but we don't.
All we need to add is a precision audio oscillator. It would need to be pretty precise and stable to not cause moans and groans
on the 2nds.


It would not be necessary to install new RF equipment, but only to "waste" some power with injection of a 20 khz audio tone.
This would put the same RF equivalent of another, totally unmodulated signal on a stations' 2nd adjacents.
And if those stations are throwing iboc, so much the better. It would require sufficient system bandwidth, and a willingness
to waste some power, but I'd love to hear how it works. Sorta like running a calibrated, controlled parasitic oscillation.
10khz would likely cause mushy tuning and too much whistle to be acceptable, but not 20khz.

 

[nd2023]

That won't push back the hiss, that will just create more noise.

 

[mmnassour]

Here's a new product....it's called a gun. Use it to put a bullet in each and every HD xmtr in the country.

Noise problem solved....and not before that happens.

 

[kenglish]

You could just offer to upgrade the old HD Radio transmitter (for free) to something in another band, where it can run full-power without walking on any adjacent channels.
And, update all the radios to tune, say, 76-88 MHz.

 

[Tom Wells]

That won't push back the hiss, that will just create more noise.

I don't think you've quite reasoned this out.

Please consider this for a moment. If there's no variation in that 20,000 hz signal, it is as if the "quieting station" has no audio on it.

If you've ever dxed and wished a splattering adjacent would be "quiet" just for a second so you could get an ID, you'd know a
quiet carrier next door is far better than a thrash metal station.

So how is this going to make more noise? I can only imagine it will diminish sideband hash noise,
and improve S/N ratio of your (local) signal.

I haven't tried this yet, but I plan to, since part 15 AM requires no NRSC mask.

Love to hear input from others who've played or worked with (or thought about) AM superhet design, IF mixing, AM modulation, etc.

 

[Play Freebird]

So how is this going to make more noise? I can only imagine it will diminish sideband hash noise,
and improve S/N ratio of your (local) signal.

I haven't tried this yet, but I plan to, since part 15 AM requires no NRSC mask.

Love to hear input from others who've played or worked with (or thought about) AM superhet design, IF mixing, AM modulation, etc.

I haven't tried this either, but my initial reaction is that it won't do any good because the added signals would be so far outside the IF passband of most receivers.

Keep in mind that it's the carrier that controls action of the RF AGC. When you tune to the upper and lower first-adjacent channels of an AM IBOC station, the carrier then shifts 10 kHz off-center of IF and most of it is filtered out. This causes the AGC to call for a LOT more gain, bringing the noise of the digital sidebands WAY up and increasing drive to the detector stage, where it comes out as obnoxious hiss.

Sad to say, but the IF filter skirts on most receivers are probably down at least 30 dB at +/- 10 kHz and maybe 60 dB or more at +/- 20 kHz.

Even an AMAX-compliant wideband filter designed to closely track the NRSC pre-emphasis curve would be around 10 dB down at +/- 10 kHz and (hopefully) steeper beyond that point, because the FCC does allow second- and third-adjacent stations to operate fairly close to another.

 

[ajc_trw]

You could just offer to upgrade the old HD Radio transmitter (for free) to something in another band, where it can run full-power without walking on any adjacent channels.
And, update all the radios to tune, say, 76-88 MHz.

With some radios it serendipity. That is to say my Aiwa portable can be set to what it calls "Japan FM" where it tunes from 76-108MHz.

I'm sure many radio manufacturers have this capability and don't use it in the US because there's nothing (yet) on 76-88MHz. However, if stations started appearing I'm sure it won't take a major overhaul to turn those HD radios to 76-108MHz tuning.

Then again I fear if those slots go up with full digital virtually all of them will be packed with so many multicasts each one will be no more than 48kbit.

Yea, CD quality. :

 

[Tom Wells]

OK, I tried this out.
(Officer Spock face with cocked eyebrow)
Fascinating......
The regular air computer did not give a clean 20 khz output, but had many SUBharmonics in the output.
With a different computer, and no audio, adding the 20 khz audio at -10 db gave impressive quieting 2 channels away.
With audio, it did seem like there was going to be some benefit. No increased 10 khz whistle...
But the "intended" audio is now also "mixed" and modulated against/with the 20 khz "inaudible" audio.
This means there is an overmodulated version of the audio ON the phantom carrier 20 khz removed.
This could either be "real" RF products or simply the results of using one audio card for both audio and the 20 khz signal.
As these modulation products "mirror' the original, at 10 khz offset, there is a sort of dual-sideband supressed carrier thing happening, because both of the 20 khz offset products meet each other in the middle. This sounds almost like there are multiple "stations"
with the same audio, at least listened to in the very near field.
At a distance, other things seem to be apparent. The definition of the "main audio" now seems limited and defined by the 20 khz
audio power subtracted from it, instead of by the 1620 khz. Much less crisp and punchy.
In exactly the same way that the iboc hiss subtracts from the resolution of the main audio, the crispness was mushed.
Any quieting gained by producing phantom carriers seems to be offset by the loss in "useful" sidebands
produced in regular modulation, and the DENSITY of signal that could be present if all audio power available goes into producing them.

Maybe a diplexed smaller transmitter set to quell a specific offending sideband could be useful at 20 khz offset,
as it could be free from any audio bypoducts from the main signal.
I would still expect the images and the dual sideband supressed carrier image 10 kc out as predictable products.
How well it would "work", I can't guess, but it shouldn't detract from the resolution of the main audio done that way.

Now what will happen if I try the same thing on the just the receive end by using a signal generator?
I suspect I will just desensitize the radio with too much quieting. We'll see.

 

[Tom Wells]

More thought on this... It seems the real-time noise (hash) can ONLY be quelled by a blank carrier
having NO relationship or coupling to the signal we'd like to protect and/or its audio.
Meaning somewhere else, more or less. Like miles away, Dad.
By mathematics, any coupling would detract, so the results would vary by reception point in "real time" only.
Why? What the heck does that mean? ( Drops dangling keys/keychain)

There was a problem presented to us at Valpo Tech.
Why can't you just hook up a converter to a VCR, convert 0-6mhz to say channel 3, record 2 hours of it, and then
play it back, feeding the output back to 0-6mhz, and have 2 hours of 0-6 mhz "hold still" for you to dx?
There's 6 Mhz of bandwidth, but not sufficient resolution within that bandwidth.

Time to try a calibrated 1640 injection aganst my 1620. I get iboc on 1630 from 1640 Dizzy-ny Radio in Milwaukee.
I suspect I won't be able to decouple 1620 enough.