AM radio receiver block

peppe from sweden
peppe from sweden Member Posts: 59 Member
edited August 2022 in Reaktor

It's not made yet. But there are a AM Eurorack module spitting out radio signals and it would be a very cool module to sample from, crackles and swish swoch sounds. AM apps does also exist so it should be possible to make one right?

Not many controls, CV in for searching of course.

When I learn more i will begin making one.

Idea's are welcome.

Peppe

Tagged:
«134

Comments

  • Laureano Lopez
    Laureano Lopez Member Posts: 102 Advisor
    edited August 2022

    There's a thread in the old forum where we played a little with AM radio stuff, Colin also uploaded an ensemble. We were actually trying to make a Costas loop and got into radio for the fun -it got complicated and I don't remember much of it, so I'm not sure how useful it could be.

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor
    edited August 2022

    that thread was a riot, your contributions in particular were confusing (in a good way), like a lot of the permutations had very different behavior that warrants future study


    and @colB 's ensemble pretty well demonstrates a virtual AM airspace imo

  • peppe from sweden
    peppe from sweden Member Posts: 59 Member

    Thanks, I check I out!

    My idea with is to sample bits and pieces of it.

  • Laureano Lopez
    Laureano Lopez Member Posts: 102 Advisor

    It was wild indeed, I think we all put our little bit to the general confusion 😁

  • colB
    colB Member Posts: 762 Guru

    Heh, yes, like all the good Reaktor nerd threads.

    The ensemble I posted is really not an accurate model of what's happening in a radio, iirc, it's overly complicated with some redundant components because of my lack of understanding, but maybe it does give a reasonable impression in terms of the sound.

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor
    edited September 2022


    i mean, it tunes and demodulates. those are the basic concepts of AM, which it does


    had a good look around/ messed with it today. first went goblin mode and changed the rectifier stage to an envelope detector (laureano's decoupled one) but mistakenly added it later in the chain without first replacing the rectifier. however it bears mention because i obvserved the filtering and double rectification of envelope detector in this orientation could actually added in to sound a bit more closer to how i think of AM sounding on the receiving end, a bit stepped on


    then i messed around with the demodulator section with the sine frequency delta or offset, cause its theory of operation vexed me. in turning it off i found out it works fine without it (only difference i noted was a couple less of those troublesome ringing frequencies in baseband/at audible range. i've never heard this expressed before but beat frequencies seem like a special case where adding waves together works non linearly and i've always wondered what the boundary of that is


    this outcome is even more confusing than if the extra oscs were helping to demodulate, because if it's not now where is the demodulated baseband signal coming from?


    like, i'd expect this if an impulse wave was used as carrier cause sidebands would be above and below every harmonic... but as i understand every 'station' centers on a single frequency of at least 50khz. so necessarily what this means, unless i've profoundly misunderstood something, rectification can kick islands of supersonic signal back to baseband no questions asked?


    like, i'd assume if that was true it would at least be dependent on carrier frequency relative to sample rate


    one thing i can say with certainty is i'd recommend fewer stations for more headroom in bandwidth for each indiv station, , as i see currently their range is bandlimited at 2k, well below around 3.5k generally considered necessary for solid speech transmission/recognition

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor

    if ya got more or less specific situation going on and every person has a different way to be confused about it, that is probably the perfect distribution for an ideal amount of confusion, which is never zero

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor

    oh yeah forgot to add the file lol lmao


  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor
    edited September 2022

    also a bit confused about this picture spectrally (it connects all waves+their carrier sines recorded at 96k. like, if reaktor is oversampling all carrier they should appear at their correct frequencies to the program. but unlike unwanted sonic content you ig can't filter off this stuff has modulator's information in it?


    maybe procedurally turning off oscillators that fold back in the range of the currently tuned-to carrier frequency??


  • KoaN
    KoaN Member Posts: 103 Advisor

    Cool! Lots of interesting textures with the stepped on and env demod filtering,sounds very authentic too...i feel i have heard this type of distortion often in AM...squared-rectifier type of.

  • ANDREW221231
    ANDREW221231 Member Posts: 295 Advisor

    its funny everything i was trying was a dead end and it was literally a last ditch attempt to see if throwing it all together might do something


    like they compliment each other well but there was no good reason to expect they'd should combine to sound more AM like. envelope followers are used in AM but im sure not like that


    i guess all the rectification would give even order harmonics, maybe thats part of transistor radio sound, never thought about that. one other thing i was messing with (after the version posted here) was having less bands that were closer together or even just overlapping . its possible to get a classic sweeped tuning sound/behavior especially if there's close to interference at the boundaries between stations

  • bolabo
    bolabo Member Posts: 97 Advisor
    edited September 2022

    This is brilliant!!! Sounds very authentic, you should definitely upload it (with embedded samples) to the user library! :)

  • Laureano Lopez
    Laureano Lopez Member Posts: 102 Advisor

    Heck, you got me nostalgic 😁 I can't get into this sinkhole now, but I'll add it to my sinkholing backlog (?)

  • colB
    colB Member Posts: 762 Guru

    then i messed around with the demodulator section with the sine frequency delta or offset, cause its theory of operation vexed me. in turning it off i found out it works fine without it...


    this outcome is even more confusing than if the extra oscs were helping to demodulate, because if it's not now where is the demodulated baseband signal coming from?

    This is what I meant by redundant components in my original model. There's some stuff there that may or may not be correct, but it's difficult to be sure with my level of understanding, because it's not actually modulating, then transmitting through the airwaves, then receiving and demodulating... it's just a hacked Reaktor model :-)

    one thing i can say with certainty is i'd recommend fewer stations for more headroom in bandwidth for each indiv station, , as i see currently their range is bandlimited at 2k, well below around 3.5k generally considered necessary for solid speech transmission/recognition

    I wasn't going for headroom, I wanted there to be interference between stations etc. and less distance to tune on the dial to go from one to the other... in order for this to be more interesting/playable as a tool for sound design. Almost like a cartoon caricature of what a good radio might have sounded like. Like some of those exaggerated tape effect plugins that produce a sounds that would have caused any real tape machine to be binned or sent for repair :-D

    I like the sound of the distortion you added. Definitely adds something when used at a subtle setting.

    Along with the distortion, I would like more ideas for adding noise and 'authentic' sounding interference. So characterful crackle and hum at different points on the dial... I did experiment with various types of broad band noise, but it would be interesting to get some stuff at narrower bands that might interfere with channels... The sort of thing that might have been caused by electrical storms etc. Not entirely sure how to go about that though.

  • Paule
    Paule Member Posts: 1,328 Expert
    edited September 2022

    Here are my 2 cents with 15 stations in radio blah blah.

    Important is the folder Samples just beneath the ens. So the snaps works fine.

    Inside the folder Samples you can create much more subfolders.

    one station for mambo stereo additional = 16 stations

Back To Top