• Get involved.
    We want your input!
    Apply for Membership and join the conversations about everything related to broadcasting.

    After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
    If you use a disposable or false email address, your registration will be rejected.

    After your membership is approved, please take a minute to tell us a little bit about yourself.
    https://www.radiodiscussions.com/forums/introduce-yourself.1088/

    Thanks in advance and have fun!
    RadioDiscussions Administrators

NEW FM PROCESSOR

If I understand correctly, the last clip is actually more correct -- and still very much rectified. If it wasn't rectified, I believe it would look like butterfly (opposite of a diamond i suppose).
When the audio signal is at its highest negative voltage, the carrier should drop to ALMOST nothing, but must never completely turn off. When there's no audio signal at all, the carrier should be idling at half voltage (quarter power).

However, even so, there's one critical detail: The lines that make up your trapezoid aren't static. They're moving around a lot -- indicating nonlinearities. The carrier you're yielding does not precisely follow the signal BBP is generating. I also don't see the any peak control at all! BBP has razor-sharp peak control -- you should be hitting -95% almost constantly, unless you're using VERY light processing.

The question is -- what can be done to linearize your transmitter? The first thing I would try, if it was my transmitter, would be to bypass any mixer / reverb / filtering stages, and connect BBP's output as late in the chain as possible. You'd be able to be louder, cleaner, fuller AND go further if you can preserve the signal integrity, as it means tighter modulation control, higher average modulation, without having to clip any more than BBP is already doing!

Best,
///Leif
 
Tom Wells said:
konbaasiang said:
Tom, I still can't make heads or tails of your video!

Isn't it supposed to look like this?

///Leif
That would be the pattern I get when I put unrectified audio into the x input.
I'll make another video of that. It is not as easy to compare the +/- when they are at opposite ends of the display.
When both pos and neg audio run the trace the same way, it is easy to see that the positve peaks go way beyond the negative.

I'm trying to figure out now what a SHARC is. http://www.youtube.com/watch?v=A1aOwvM7l0s

What are we looking at here?

That's supposed to have audio but YouTube pulled it despite it being covered under fair use for education. Anyways, Leif... could you post that on your own site if you want to? It's a neat video.

As to what a SHARC is, I don't suppose you've heard of google or wikipedia? </smartass> http://www.analog.com/en/embedded-processing-dsp/sharc/content/index.html
 
<straight_answer>SHARC is a DSP chip.</straight_answer>. ;)

I ported the Breakaway Core (equivalent to Breakaway Live) to run on one a couple of years ago.


BTS, I see your point regarding measuring at the modulator -- you're absolutely right. However -- isn't the phase shift between the audio processor and the modulator part of the problem?

Phase shift ruins peak control. The path between the audio processor and the modulator needs to be as short and clean as possible to retain peak control. So, in an ideal setup, wouldn't you have a perfect trapezoid even if you connected X to the audio processor and Y to the RF sample output?

///Leif
 
I am also convinced that the non linearity seen is a result of not tapping the audio at the very grid where the 6SN7 mixes
audio and rf from the oscillator. If I get brave tonight I'll open this mess up. I haven't pulled real maintenance on this thing in a couple years.
In order to make the videos, the best audio sample I could find was off the side-chain amp driving the reverb tank. All others showed
phase distortion rotating the side view of a "slice" of conical section into partial rear view into a cone.

............
Some 2-3 hours later..
I did tap the audio at the grid itself and also had excessive circular presentation. Returning to former audio from sidechain amp with BBP audio, using tones, sweeps, proved best overall phase correction at 1000 was the very sweet spot I run reverb, which is also causing the wavy sides.
The tones allowed me to balance all tone controls, gains, etc to the point where the non-linear distortion crosses over at the midpoint of the sides. Clearly not as good as removing them as you point out.
I tried to make up the gain on the eq ( set flat otherwise) but could not make it sing.

Then I left all tone (phase distorters) alone and put the line in and verb gains back to "6".
It turns out the settings for tone were actually about what I had been using, but totally setting them flat as possible proved you
are right in that ANY tone level changes after the processor are all loss.
I wont go into how the reverb corrects my phase distortion at 1000 hz.

The audio sounds about 3db louder now. And even more huge.
Then I got to thinking about the fact that reverb is phase distortion in audio by defintion.
See if you add reverb to the audio in your clip whether the edge wavers the sides. Maybe it's because all this is happening in analog?

Anyway, if reverb in this case is causing the wavy sides, the last you-tube pattern is still about what I have.
I have to wonder how much the reactance of the antenna reflects into the width of the trace or overall linearity.
Tuning the load and couplng are a breeze with this pattern.
My intention in this tube modulator was to use the best hi-fi ideas from the 1930s to the 1960's, using receiving tubes and still sound
like WCFL and WLS from my youth. Every machine and circuit has its own "mass" and behavior. In the 1960s and 1970s these two stations
weren't just loud, they had insane punch. Someone can surely tell us what kind of AGC these used in those days.
It was a very definite type of distortion but sounded fantastic on AM.
I was pretty close before, but really needed the help of the multiband and asym.
I do run it really gently, Dance 0db drive, range and power 15, speed 0, both bass @-23, but no filters and 150%.
Just how far can you effectively skew the A/D range for asymmetry?

Breakaway was the only thing so far that has truly been able to make my 100mw make sounds like a stomping dinosaur in 5.1 theatre
sound on AM.

In continued listening I am still very happy in every way.

I'm afraid of the new plug ins. They would probably dislodge speakers from whatever they're mounted to.
 
A friend of mine used them the day after we tested them, for a live stream with a pretty active chat room. And within the first 5 minutes, someone listening with a pretty decent system was able to shake ice off of their roof with the sound.
 
BBP has a phase rotator built in, yet there are phase rotator plug ins available...correct? Is there any reason to add these plug-ins to BBP?
 
BBP does not have a phase rotator built in -- it actually has a phase scrambler built in. The two are actually not at all the same.

The phase scrambler is very effective at straightening out high-end asymmetry and preventing clipping distortion, in a very transparent, nonobtrusive way. It does not have much of a "sound" of its own.

As such, the plug-in phase rotators are preferrable as an effect. A lot of people seem to like the sound of the phase rotators.. Some even like the Phase Tornado -- although I cannot fathom why! :)

///Leif
 
konbaasiang said:
BBP does not have a phase rotator built in -- it actually has a phase scrambler built in. The two are actually not at all the same.
<snip>
///Leif

Rotating certainly scrambles. So, what's the semantic difference?

Kind Regards,
David
 
It is not at all certain that rotating scrambles.. In fact, it isn't so.

Phase rotation a pretty controlled concept -- it's done with allpass filters, and is essentially a frequency-dependent delay. Starting from the top (high frequencies), delay increases progressively as the frequency goes down, but never decreases. Phase rotation can be implemented in analog technology.

Phase scrambling (my own name) is a completely different technique, and it's one of those "can only be done in DSP" algorithms. I can't go into it in any further detail -- it's one of the many things that separates Breakaway from the competition, and makes Breakaway the only processor that can be competitively loud without imposing a sound of its own.

You may wish to test it for yourself. Try playing something with bass + saxophone with an agressive preset -- turn off the phase scrambling, make sure there's plenty of IM distortion. Then, substituting the phase scrambler for a traditional phase rotator, compare how much the timbre of the sound changes to with one or the other, for an equal amount of distortion prevention.

Some might ask, why bother with plug-ins then?
The primary goal of Breakaway was transparency at competitive loudness levels -- something that was never before possible on FM. Now that this has been achieved already, we can bring out the salt and pepper shakers, and the hot sauce!

Best regards,
///Leif
 
I lost the link to the plug ins...help!
 
Is there a description list of all of the presets available? For example...Plutonium...catchy name, but what are it's characteristics/number of bands/etc? I know it's possible to listen to each & take notes, but if it's already available, it'll allow me to cut to the chase more efficiently.
 
Can't answer for all the presets...but to my knowledge all presets are but I think virtually all the main presets use all 7 bands. If you let your mouse hover over the name of the presets, there's a bit of an explanation, but nothing real technical. These are my favorite flavors:

Regulator is an urban/chrban/chr preset that features thunderous bass characteristics
Plutonium and Plutonium NR are arguably the loudest presets in the bunch, besides New York...great for rock...very aggressive, nice mids
Oldies is a preset good for just about anything except urban. It's good for Oldies...but good on other stuff, too.
Helix started out as my favorite preset, but it's lacking in mids, IMHO...but it's a nice curvy-EQ preset...bumped on hi's and lo's
Rock is a great preset...for just that Rock.
CGsmooth is a preset by Cornelius of Telos that emulates the CGSmooth preset in the Omnia

Until a full control version of BB is released, it's hard to know precisely what is going on under the hood with the crossovers attacks or release settings.
 
Sgeirk said:
CGsmooth is a preset by Cornelius of Telos that emulates the CGSmooth preset in the Omnia

From what CG told me of his opinion on that preset, it seemed that he was saying CGSmooth is more of an idea of a sound that is in his head. So I personally wouldn't call it an emulation of the two Omnia presets he made, unless he says it is. Although he did say that the Breakaway version was doing exactly what he would want his own processing (that he's been working on) to do. Not a bad compliment at all. ;D

_________________________________________________

Jill FM is my answer to the original Jack FM sound... essentially a modernized (aka louder) "rocky mountain" classic rock sound. I tried to take the "masculine sounding" aspects from that, and make them more "feminine". I think you'll find you get more TSL from Jill FM with both sexes.

New York is balls-to-the-wall loudness, within the "acceptable limits" of distortion for the ideals of Breakaway (or it wouldn't have been included). In fact it probably is the limit. ;) It was created specifically for New York City, and really should not be used if you're not in such a backwards-thinking market in terms of audio processing. It's just there if you need it.

Zenith is my reference setting. It is designed to only sweeten the spectral balance and damper extremely loud "out of place" transients. Other than the very slow AGC, there is not much gain happening. It's totally one of those presets that you have to press bypass on or you would never realize that it's there... but when you do bypass it, you know why you're using it.

Rusticity '73 was built initially only to be this '73 version. It's inspired by the hysteresis and vibe of a "stacked" 1970s setup, in which one might be able to get by chaining a DBX comp, two Dorrough DAP 310s, and a CBS Volumax. Rusticity is a modernized version of that vibe, so you get the control of a 7-band and some of the "problems" handling some material are no longer problems, except for where I thought the hysteresis was needed and part of the vibe I was going for. The word "Rusticity" also has several puns that I liked and thought related to what this preset is doing pretty well.

Rusticity '83 was based on '73 and designed to have a bit more density and better control and less of the hysteresis of '73. Ironically on lots of material '73 ends up sounding more consistent on average - compare the two ABBA cuts on Leif's "Torture Test", imho '73 is the most consistent of any preset. But '83 controls "surprises" much more elegantly, so it is better for current popular music.

Rock was based on Plutonium by the way.

Rustonium is a new preset me and Leif are collaborating on, and you may have guessed that it's a combination of Rusticity and Plutonium. So far we like it a LOT. It's not released yet, which is why you can't find it in the preset menu yet.
 
I tested the CPU requirement for BB on my 3GHz Core 2 Duo machine and found the CPU stayed around 15%. This was with XP Pro. I tried running a web audio encoder program at the same time and the CPU stayed under 20%. I still need to run the web server program also and I think the one computer might be able to act as a complete web streaming machine. Unfortunately the web server program only runs on Windows Server.

I wonder if Breakaway has been tested on Windows Server 2003 or later?
 
Anyone using an EQ plug in with Live! or BBP? It seems most that look promising open in a .exe format & I believe what's need for Breakway use is a DSP (not VST) .dll. I'd prefer a graphic eq rather than one that you have to "draw" the curve...any ideas?
 
Funny you should ask. I've very recently been enjoying these:
http://www.anwida.com/product.asp?pid=10
http://www.anwida.com/product.asp?pid=11
but haven't used them in anything, yet. (I already have extensive EQ for precision and "touch")

They are VST plugins however, so you'll need a loader for Winamp DSP. There are a bunch about, and for someone who only runs a few plugins at once (if that) most are free.

However... myself I use a ton of plugins for mastering, so I use:
DirectiXer, turns any VST into a DirectX plugin
Adapt-X, can chain as many DirectX plugins you want and is a Winamp DSP :)
 
Jesse Graffam said:
Funny you should ask. I've very recently been enjoying these:
http://www.anwida.com/product.asp?pid=10
http://www.anwida.com/product.asp?pid=11
but haven't used them in anything, yet. (I already have extensive EQ for precision and "touch")

They are VST plugins however, so you'll need a loader for Winamp DSP. There are a bunch about, and for someone who only runs a few plugins at once (if that) most are free.

However... myself I use a ton of plugins for mastering, so I use:
DirectiXer, turns any VST into a DirectX plugin
Adapt-X, can chain as many DirectX plugins you want and is a Winamp DSP :)
Thanks Jesse...I'm not well versed in loaders & how they work. Do you simply install the loader in the same folder at the VST plug-in and then the VST's will show up in the BBP Browse window?
 
Status
This thread has been closed due to inactivity. You can create a new thread to discuss this topic.


Back
Top Bottom