Modulation can be too much of a good thing

Discussion in 'Amplitude Modulation' started by K5UJ, Oct 24, 2019.

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: L-MFJ
ad: Left-2
ad: MessiPaoloni-1
ad: Subscribe
  1. K5UJ

    K5UJ Ham Member QRZ Page

  2. N2DTS

    N2DTS Ham Member QRZ Page

    I had that mod monitor and have that scope!
    KG2IR always said the same, run 85% mod and more carrier.

    As a hold over from the past, I always tend to run lots of audio on the carrier.
    That works for poor copy/weak signal work....

    Also, many use SDR's with sync detection which eliminates a lot of the problem.
    I wonder why AM broadcast receivers never really got sync detection...
     
    KB1QHI likes this.
  3. K5UJ

    K5UJ Ham Member QRZ Page

    I hope the fellows who have 150% or more positive asymmetry learn from this. Just because the technology makes it possible doesn't mean it is good.

    High asymmetry exacerbates selective fading distortion on skywave.
    It is hell to copy using envelope detectors.

    Those who brush off operators using old receivers are simply rude.

    Agree with KG2IR--if a ham wants to be heard, then run more power not more audio. However, too little audio isn't good either. As Mr. Persons wrote, find the sweet spot.

    The downside to synchronous detection is that it masks distorted audio. I've heard operators who have SDRs and sync. detect. on all the time tell others their audio was great when in reality is was way too hot and distorted.

    Probably cost. The general public knew nothing of s.d. and manufacturers would only make what could be sold for as low production expense as possible. Additionally, s.d. isn't needed much with ground wave. It's really only good for QSB on skywave. Most medium wave listeners aren't dxers. The higher end swl sets had s.d.
     
  4. K4KYV

    K4KYV Subscriber QRZ Page

    That depends on whether the asymmetry of the waveform is natural to the voice characteristics of the operator doing the talking, or artificially the result of "processing", negative cycle loading, ultramodulation, etc. My homebrew xmtrs are capable of about 130% positive, and the Gates-Raytheon hybrid, close to 140% with the 845s driving the 833As. I phase the audio so that the half-cycle with the greater excursion is in the positive direction. That allows greater percentage of modulation in the negative direction without flat-topping on the positive and thus less splatter. The naturally extended positive peaks are of short duration, while the average peaks of longer duration hover around 100% or less. If the spikes at >100% cause distortion, they are infrequent enough and of such short duration that no-one has ever mentioned hearing it, and I don't hear anything monitoring on my own receiver.

    I have a passive Kahn Symmetra-Peak all-pass filter that I can instantly switch into the audio line. It makes the waveform look more symmetrical on the envelope pattern on the scope, but I have never experienced any advantage to using it.

    Natural asymmetry does not increase sideband power beyond allowing more complete modulation of the carrier by allowing negative peaks to approach closer to 100% negative without breaking base line nor flat-topping on the positive. The area under the curve of the positive peaks is still same as that of the negative peaks; the modulation transformer cannot pass DC. This works only if the modulator is capable of the necessary undistorted peak power, and the RF final has good modulation linearity out to the capability of the modulator to produce positive peaks.

    I am speaking with experience limited to analogue technology; this may not be strictly true with hi-tech digital technology currently used in state-of-the-art AM broadcast processors, combined with 100% DC coupling to the modulator stage such as with a pulse-width modulator.
     
    W2VW and WZ5Q like this.
  5. WB2GCR

    WB2GCR Ham Member QRZ Page

    The writer speaks of commercial broadcast - not amateur radio.
    Entirely different conditions, requirements, and equipment.

    Imho, does not apply to ham radio at all.
     
    W2VW likes this.
  6. WA3VJB

    WA3VJB Ham Member QRZ Page

    I disagree.

    The thrust of the article speaks to what many of us have tried to address at home, with our stations, on AM.


    AM radio isn’t supposed to sound bad.
    It can be a clean and pleasurable listening experience,
    even when there is only 3 kHz of audio bandwidth.
    On the other hand, AM can be ugly to the ear when there are maladjustments.



     
  7. KL7AJ

    KL7AJ Ham Member QRZ Page

    One of the most effective "processors" an AM operator can use is a voice polarity detector and "flipper over." Typical male voices are highly asymmetrical, the difference between the positive and negative peaks can be dramatic....commonly a ratio of 6:1 or more. You can always benefit by assuring that the larger peaks create positive modulation, while the smaller peaks downward modulate. This can do more for you than a lot of compression. The circuits for achieving this are fairly simple.
     
  8. K5UJ

    K5UJ Ham Member QRZ Page

    Hi Bear, please list what is different and why it matters. I mean, what are the conditions, requirements and conditions specifically?
    I have one of those things, or maybe it forces symmetry, I can't remember, but either way I've never even bothered to try it. All that is needed is a look at one's base band audio with an oscilloscope. If the voice produces asymmetry, and the higher peaks are negative going, then a simple swap at some point in the audio chain is all that is needed. By that I mean flipping the + and - lines in a balanced line. Let's say you don't have balanced line. I'll use a D104 mic element as an example. Flip the element leads or (this is probably easier) swap the center conductor and cable shield around.

    By the way, the most important "processor" is your voice. I've been preaching this to deaf ears for a few years: Develop your voice. Practice reading copy and listening to yourself. Work on diction, annunciation and projection. Speaking clearly is the phone equivalent of well developed characters on CW. I work not a few fellows on phone who don't know how to work the mic, have slipshod pronunciation, slurring words together with a kind of whispery voice etc. I get about 50% of what they're saying. All the processing gear in the world won't fix bad speech.
     
  9. WB2GCR

    WB2GCR Ham Member QRZ Page

    Well, the article seems to apply to transmitters that have no protection from "base lining", driving the output into
    cutoff. Yes, there is in theory more "distortion" if you modulate more than 100% positive, and clip before
    going say ~98% negative. Irrelevant to 99% of all ham radio receivers that I know of.

    The argument that highly positively modulated AM signals will distort the detector on ham gear
    is something that I have never personally witnessed on my gear. That spans a fairly broad range
    of tube and solid state gear - not everything, but a lot of stuff. Certainly never noticed it on tube
    gear.

    Yes, the All Pass Filter aka "phase rotator" will make the positive and negative essentially identical.
    Apparently the ANAN's linear amplifier predistortion scheme turns out to do the same thing? At least
    that's what casual observation of the received waveform with it on and off shows. Makes sense
    as it tries to correct non-linearities.

    Broadcast AM's aim is to have maximum audio "density" within a broadcaster's direct market, above
    some number of microvolts, and just about zero interest in how they sound near the noise, or when the
    band is long, short, or broken. That's not ham radio.

    Apparently, or so it seems, outside the NE USA, not many hams are running >100% positive
    modulation? Here in the NE it's fairly common. You can now listen in using one of the many
    WebSDR sources and hear for yourself. Just keep in mind that the web based sites that use
    a certain software ALL have a known audio artifact. Both the WA1QIX and the K3FEF sites use
    this software. So don't fully judge the audio quality that way!

    As far as I can see, it increases the effective transmitted power, as the positive peaks are higher,
    and that is more peak power. It's also easier to hear under marginal conditions. And it sounds
    GOOD! All things being equal, stations with already good audio, good mic control, and with ops
    who know how to use the mic sound great with >100% positive peak modulation! :D

    Ymmv...
     
  10. K5UJ

    K5UJ Ham Member QRZ Page

    What in the article gave you the impression the author is against > 100% positive modulation?
     

Share This Page

ad: wmr-1