The AM Forum
March 28, 2024, 10:46:53 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Real World Experiences using All-Pass and Phase Rotators Wanted  (Read 7884 times)
0 Members and 1 Guest are viewing this topic.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« on: March 08, 2020, 01:28:30 PM »

Hello again.... JJ here,

I've decided to make some big audio changes in the shack and have been experimenting with a Chinese four-chip phase rotator board. They were on e-bay for cheap.  In the past I tried this board and didn't like much losing my positive peaks.  But after EQing my new audio chain, I find that it was mid-bassy and most of the positive peaks were made up of mud frequencies between 100 to 500 area. When I dialed the mud out on the EQ, I was left with mostly symmetrical looking audio.  So I said WTF and decided to try the all-pass board again and it actually made higher peaks but using a much better mud-freq cut.  IE, the all-pass took a drab looking rounded waveform and added some definition to it. I have it early, right after the 528e mic preamp.

I realize there is a split in thinking with many loving phase rotation and others never using it again.   I know AM broadcash stations use it a lot as well as foreign BC.  Some of my ham friends like Chuck /K1KW uses it as well as Clark/ BCG and a host of others on the air. The new Max processor will have an all-pass filter too.
 
Also, there are a group of some very smart AMers like Steve/QIX, Dave /VW and others who do not use phase rotation.  I don't remember TimTron's opinion, but I would guess he does not in ham radio, but may in BC service.

I wonder if we could start a dialogue of actual experiences based on phase rotation? What are some of the tricks, observations, good experiences and failures you've seen?  What if you were the only one using an audio chain - would you still do it? We have all read the patent and articles showing how well it works, but what have you seen and done with it?

I am in the middle of running new tests with my all-pass board and trying the more symetrical, denser audio route now - so would be very much interested in everything related to all-pass experiences.

Don't be shy...

T


Patent - Audio Signal Peak Energy Equalization:

http://www.w3am.com/SymmetraPeakPatent.pdf


Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #1 on: March 08, 2020, 02:09:18 PM »

** Another question:

An all-pass filter makes a signal more symmetrical.. so it gives us the opportunity to run "denser" audio.  So this automatically means more background room/fan/ blower noise, correct?   IE, is there any way other than a downward expander (noise gate) to run denser audio without an audio gain increase resulting in more background noise?


BTW, I am experimenting with my Aphex noise gate and have been able to reduce room noise  - by up to -10DB more by careful adjustments - without getting the unnerving on/off gating effect. A mix of compressor release times and the Aphex seems to mask these effects when using just a touch of downward expander.  More later...


T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
N1BCG
Member

Offline Offline

Posts: 836


« Reply #2 on: March 08, 2020, 07:23:10 PM »

Two thoughts...

1) Voice asymmetry is not the same across the vocal range in that lower frequencies often have the opposite polarity of higher frequencies. Because of this, determining “proper polarity” only accounts for most of the time, not uniformly for all of the time. This is one of several arguments for using an All Pass Filter.

2) If processing brings up background noise to objectionable levels then it’s set too high. Less compression and proper microphone technique go a long way toward having great audio in these conditions.
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #3 on: March 09, 2020, 01:26:04 AM »

My Homebrew processor is a phase rotator, same one you're playing with.

Then a compressor.

I use a Homebrew limiter comprised of a couple diodes in series with an led.  This limits positive and negative peaks.  I don't drive it into clipping however.  It's there as a stopgap.

That is then run into the homebrew peak stretcher that allows me to dial the positive peaks anywhere I want them. I get the best of both worlds, positive assymetry along with 'more torque under the curve'

This setup has garnered more than a couple good audio comments.

--Shane
KD6VXI

Logged
KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #4 on: March 09, 2020, 06:40:56 AM »

Shane,

What the heck is a "peak stretcher"? Can you give us a few words on how it works? Or, better yet, a schematic?


Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #5 on: March 09, 2020, 08:40:20 AM »

http://amfone.net/Amforum/index.php?topic=40320.0

I still haven't unpacked everything here, so the schematic is sitting in a box.  Somewhere.

But this is it, basically.

At the time of making the video I didn't have anything smaller than a 250 Watt slug for the pigeon meter.  But, the dummy load is also calibrated per a pigeon.

Oscope tells it all.

--Shane
KD6VXI
Logged
WB4AIO
WB4AIO
Contributing
Member
*
Offline Offline

Posts: 242


Better fidelity means better communication.


WWW
« Reply #6 on: March 09, 2020, 01:23:38 PM »

[...]

I am in the middle of running new tests with my all-pass board and trying the more symetrical, denser audio route now - so would be very much interested in everything related to all-pass experiences.

Don't be shy...

T


[...]



The phase relationships that cause asymmetry are special and critical. Almost anything that upsets them tends to remove the asymmetry. Thus, not only phase rotators, but audio transformers with lots of phase shift, audio filters that aren't phase linear, EQs at certain settings, and highly-colored non-flat microphones or transmitters can remove part or all of your natural voice asymmetry. Only on rare occasions will such such devices have _just the right_ characteristics to _introduce_ asymmetry; most of the time they will reduce or eliminate it.

And that makes sense if you think about it. It takes exact timing, exact coincidence of the upward-swinging wavefronts of two or more different frequencies to create asymmetry. Some voices and a few musical instruments (notably brass) do this naturally if they are captured with good fidelity, but not many other sounds do. Kick in some random phase delays, almost _any kind_ of phase or timing change, and that exact timing is upset.

Kahn's example of the toy soldiers all lined up perfectly on a table is a good analogy for asymmetrical audio. Bang on the table with a hammer or your hands, it doesn't matter, and very quickly the arrangement of the soldiers becomes random (like the waves in symmetrical audio). And no matter how short or how long you bang, or what you use for the banging, you'll never get them aligned again by random banging. (Well, almost never -- if you banged for longer than the universe has existed it's theoretically possible that you might cause them to be perfectly aligned again.) It's entropy at work, really. 52 cards tossed across the room never arrange themselves back in order either, nor do broken eggs reassemble themselves.

I've been using digital audio VST plugins for my amateur radio audio chain lately, instead of physical devices. I'll attach pictures of a couple of them that have an affect on asymmetry.

The first is Voxengo Phase, also called Voxengo PHA-979. It's made by the Russian studio software company Voxengo, and it lets you dial in any degree of phase shift to your audio that you desire. I find that a phase shift of minus 48 degrees is maximally effective in making my voice waveform symmetrical. (With careful adjustment, certain other phase shifts can actually _invert_ Nature's asymmetry!)

Most of the time, though, I don't need to run Voxengo Phase, because another plugin I use, TBProAudio's Impress compressor contains a bandpass filter which introduces (transparently, I cannot hear it) enough phase shift to make my audio look pretty symmetrical on its own. I have it set to pass 66 Hz to 10,000 Hz, using 192 dB per octave ( ! ) slopes (see screenshot).

73,

Kevin, WB4AIO.


* Voxengo Phase 2020-03-09 123321.png (133.83 KB, 533x284 - viewed 320 times.)

* Impress 2020-03-09 123902.png (249.13 KB, 664x518 - viewed 341 times.)
Logged

K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #7 on: March 09, 2020, 01:50:39 PM »

http://amfone.net/Amforum/index.php?topic=40320.0

Shane,

That's a very interesting thread. Thanks for pointing it out again.  I have tried a "peak stretcher" too - it's in that thread.


Kevin:  I gots to try out some of that software.  Among many other audio things, I have been experimenting with some DSP audio software lately and learning as much as I can.  Your DSP phase shifting approach is appealing.  Which program would you recommend mainly for changing around my polarity and phase shift?  Impress or Voxengo software?


I get the impression you are striving to obtain symmetrical audio rather than lining up the soldiers for big peaks. Yes, phase relationships are so fragile and even effected by our organic time of day and the state of our vocal chords.


Curious, but the other day I had the biggest positive peaks I ever had due to some changes I made, but the UGLY mud frequencies were dominant to achieve this. When I got rid of the mud, the audio was back to 115% positive or so - BUT it sounded great based on recordings.   So now I don't care anymore about big peaks, rather how natural the audio sounds with just a touch (5DB) of peak limiting.   It's the never-ending long winding road... Wink

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WB4AIO
WB4AIO
Contributing
Member
*
Offline Offline

Posts: 242


Better fidelity means better communication.


WWW
« Reply #8 on: March 09, 2020, 02:41:10 PM »

[...]


Kevin:  I gots to try out some of that software.  Among many other audio things, I have been experimenting with some DSP audio software lately and learning as much as I can.  Your DSP phase shifting approach is appealing.  Which program would you recommend mainly for changing around my polarity and phase shift?  Impress or Voxengo software?


I get the impression you are striving to obtain symmetrical audio rather than lining up the soldiers for big peaks. Yes, phase relationships are so fragile and even effected by our organic time of day and the state of our vocal chords.


Curious, but the other day I had the biggest positive peaks I ever had due to some changes I made, but the UGLY mud frequencies were dominant to achieve this. When I got rid of the mud, the audio was back to 115% positive or so - BUT it sounded great based on recordings.   So now I don't care anymore about big peaks, rather how natural the audio sounds with just a touch (5DB) of peak limiting.   It's the never-ending long winding road... ;)

T

Tom:

The Voxengo is the best plugin I have found for direct phase adjustment. It does the job well even though it's not a typical phase rotator. The Impress is a nice filter/compressor that _just happens_ to symmetricize my voice with its current settings, but who knows how many others could duplicate that result...

In recent years I have been favoring symmetrical or close-to-symmetrical audio; it's just as loud as naturally asymmetrical audio -- and imposes no need to reduce carrier level in linear AM. Actually I don't mind a _tiny_ amount of positive asymmetry. Then I can run my positive peaks at 100 per cent. and my negatives at around 90 per cent., staying well away from carrier cutoff or wavelet problems, even those induced by turbulence in the Heaviside layer.


73,

Kevin, WB4AIO.

I am really enjoying your (and Clark's and KHK's and others') discussions of audio here!
Logged

KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #9 on: March 09, 2020, 03:29:43 PM »

http://amfone.net/Amforum/index.php?topic=40320.0

I still haven't unpacked everything here, so the schematic is sitting in a box.  Somewhere.

But this is it, basically.

At the time of making the video I didn't have anything smaller than a 250 Watt slug for the pigeon meter.  But, the dummy load is also calibrated per a pigeon.

Oscope tells it all.

--Shane
KD6VXI
Shane,

Thanks for the link. The interesting nugget I pulled from that is how you first make the audio symmetrical (with a phase rotator), then dial-in the positive peak level with an asymmetrical amplifier. Perhaps, the best of both worlds? Very clever!


Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #10 on: March 09, 2020, 05:08:03 PM »


Tom:

The Voxengo is the best plugin I have found for direct phase adjustment. It does the job well even though it's not a typical phase rotator. The Impress is a nice filter/compressor that _just happens_ to symmetricize my voice with its current settings, but who knows how many others could duplicate that result...

In recent years I have been favoring symmetrical or close-to-symmetrical audio; it's just as loud as naturally asymmetrical audio -- and imposes no need to reduce carrier level in linear AM. Actually I don't mind a _tiny_ amount of positive symmetry. Then I can run my positive peaks at 100 per cent. and my negatives at around 90 per cent., staying well away from carrier cutoff or wavelet problems, even those induced by turbulence in the Heaviside layer.


73,

Kevin, WB4AIO.

I am really enjoying your (and Clark's and KHK's and others') discussions of audio here!



Wow. Amazing what is available for free. 

Website says:

"Voxengo offers you high quality professional audio plugins: VST plugins and AU plugins (AudioUnits), and sample rate converters, for Windows and Mac OS X computers. You can use our audio plugins with Steinberg Cubase, Logic Pro, Cakewalk Sonar, Ableton Live, Reaper, Digital Performer and other software with AU plugin and VST plugin support. Buy Voxengo Plugins here at pluginboutique.com with confidence, read reviews and customer feedback about Voxengo Software and feel free to download demo trial versions and Voxengo freeware. "

https://www.pluginboutique.com/manufacturers/31-Voxengo?s=price_asc

** So is there a free version of the "MotherSoftware" needed to plug in the modules?  What are you using?

What is the easist way to interface a mic to the plugins and then output to the rig? I would guess access to the computer soundcard in and out...

Interesting approach, OM.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #11 on: March 09, 2020, 06:31:38 PM »

I have been a proponent of the virtual rack and have been promoting it for years on here.  Glad to see someone else uses it as well.

I like Reaper.  It's free.  It's windows and Linux available.   I've used reaper, Adobe Audition and Live Professor before.  There are even more not listed.  But to me, Reaper has a professional feel and is free for non commercial use.

--Shane
KD6VXI
Logged
WB4AIO
WB4AIO
Contributing
Member
*
Offline Offline

Posts: 242


Better fidelity means better communication.


WWW
« Reply #12 on: March 11, 2020, 12:25:34 PM »

[...]
** So is there a free version of the "MotherSoftware" needed to plug in the modules?  What are you using?

What is the easist way to interface a mic to the plugins and then output to the rig? I would guess access to the computer soundcard in and out...

Interesting approach, OM.

T




The "mother software" I use is Cantabile Light, which is free. Reaper sounds interesting, though (Thanks, Shane!). I have Cantabile linked to the Voicemeeter Potato digital mixer (donationware), which provides all the routing options I need.

My main sound card (ESI Maya44ex) has an XLR mic input with phantom power built in. The digitized audio is routed to Voicemeeter Potato and thence to Cantabile, then after passing through the VST plugins chain it is digitally routed back into the output section of Voicemeeter Potato.

With my Anan, I just drive its digital (VAC) input with Potato and no actual audio cable has to go in to the transceiver. In fact, there are no audio cables at all, including speaker or line out cables, connected to the rig -- it's Ethernet only! But if I wanted a physical output, I could just route the output of Potato into one of my two sound cards' output jacks, no problem.

I like the digital approach because it's so nice to drive the rig with a test generator and see perfectly flat-topped square waves and ideally formed triangle waves on the RF output scope. That has to help with cleanliness and transparency.

73,

Kevin, WB4AIO
Logged

K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #13 on: March 11, 2020, 01:16:35 PM »


The "mother software" I use is Cantabile Light, which is free. Reaper sounds interesting, though (Thanks, Shane!). I have Cantabile linked to the Voicemeeter Potato digital mixer (donationware), which provides all the routing options I need.

My main sound card (ESI Maya44ex) has an XLR mic input with phantom power built in. The digitized audio is routed to Voicemeeter Potato and thence to Cantabile, then after passing through the VST plugins chain it is digitally routed back into the output section of Voicemeeter Potato.

With my Anan, I just drive its digital (VAC) input with Potato and no actual audio cable has to go in to the transceiver. In fact, there are no audio cables at all, including speaker or line out cables, connected to the rig -- it's Ethernet only! But if I wanted a physical output, I could just route the output of Potato into one of my two sound cards' output jacks, no problem.

I like the digital approach because it's so nice to drive the rig with a test generator and see perfectly flat-topped square waves and ideally formed triangle waves on the RF output scope. That has to help with cleanliness and transparency.

73,

Kevin, WB4AIO

"Voicemeeter Potato" ?  Only a cheap pimp would come up with a name like that!   Grin Grin Grin

I like your idea of using a sound card with an XLR jack and phantom supply. This is microphone ready and a great jump into the computer software audio whirl.  

OK on Cantabile as the MotherSoftware and is free.  I'm just playing around and trying a lot of things right now.  Learning a lot.

Several of my Flex and ANAN buddies run all of their audio chain thru computer based software like that.  And the ability to start with a perfect squarewave through the whole chain is amazing.  Does it remain a squarewave up into the higher frequencies like over 5K or does it round out like most every piece of audio hardware?

I've been playing around with some DSP software lately with very transparent and controllable results, but I'm still using my 528E as an audio preamp. The switch to an XLR audio card would be a logical step.

My main goal for this particular audio chain is to build a system that reproduces the most NATURAL voice possible. No fake bass. I have let the positive peaks go where they may and find when the mud frequencies are cut, most of the big peaks are reduced.  Right now I am using no compression, no phase rotation, just 5-7 DB of DSP peak limiting.  I have a DSP EQ that does an S curve to kill the mud and accent the highs.  And now playing with DSP 4.5, 6.5 and 8 KHz LP bandwidth filters.  That's it.  I want no pumping over-processed sounds or big low end "haze" anymore.  Just clean, natural audio that sounds BETTER on the air than I do in person. FM Radio, big carrier silence, "no static at all."  Based on some 4-1000A reports last night I am getting close to that goal.

At least that is the audio da jour for now until I get bored again.....   Wink

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W1AEX
Un-smug-a-licious
Contributing
Member
*
Offline Offline

Posts: 1488


Apache Labs SDR


WWW
« Reply #14 on: March 11, 2020, 01:17:05 PM »

Tom,

You could always go to the dark side and just run your old HPSDR board (or any ANAN hardware) with Thetis (or PSDR mRX PS) which is free and offers 10 band pre-eq, 10 band continuous frequency compression, 10 band post-eq, and an adjustable phase rotator. You also get negative peak soft-limiting in the deal!

Come to the dark side Luke...

Rob W1AEX


* phase rotator.jpg (60.74 KB, 619x511 - viewed 330 times.)
Logged

One thing I'm certain of is that there is too much certainty in the world.
WB4AIO
WB4AIO
Contributing
Member
*
Offline Offline

Posts: 242


Better fidelity means better communication.


WWW
« Reply #15 on: March 11, 2020, 02:08:59 PM »



[...]
And the ability to start with a perfect squarewave through the whole chain is amazing.  Does it remain a squarewave up into the higher frequencies like over 5K or does it round out like most every piece of audio hardware?

[...] My main goal for this particular audio chain is to build a system that reproduces the most NATURAL voice possible. No fake bass. I have let the positive peaks go where they may and find when the mud frequencies are cut, most of the big peaks are reduced.  Right now I am using no compression, no phase rotation, just 5-7 DB of DSP peak limiting.  I have a DSP EQ that does an S curve to kill the mud and accent the highs.  And of course the 4.5, 6.5 and 8 KHz LP bandwidth filters.  That's it.  I want no pumping over-processed sounds or big low end "haze" anymore.  Just clean, natural audio that sounds BETTER on the air than I do in person. Based on some 4-1000A reports last night I am getting close to that goal.

At least that is the audio da jour for now until I get bored again.....   ;)

T


Hope to hear your new audio chain soon. As a wise man once said of audio processing: less is more.

The higher frequency square waves round out a lot, starting around 800 Hz, because the Anan is brick wall filtered at 10 kHz max, so no chance for a true 5 kHz (or even 1 kHz) square wave!

I'll attach images of what the Anan-10 amplitude-modulated RF waveform looks like, using the VAC input, with a 35 Hz square wave and a 300 Hz triangle wave being fed through "Potato."

73,

Kevin, WB4AIO.


* Anan 35 Hz square wave 2020-03-11 135358.png (23.21 KB, 889x514 - viewed 318 times.)

* Anan 300 Hz triangle wave 2020-03-11 135843.png (25.22 KB, 886x507 - viewed 321 times.)
Logged

K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #16 on: March 11, 2020, 03:59:39 PM »

Tom,

You could always go to the dark side and just run your old HPSDR board (or any ANAN hardware) with Thetis (or PSDR mRX PS) which is free and offers 10 band pre-eq, 10 band continuous frequency compression, 10 band post-eq, and an adjustable phase rotator. You also get negative peak soft-limiting in the deal!

Come to the dark side Luke...

Rob W1AEX

Rob, you savage you!

Does it have LP filter DSP for bandwidth and positive peak limiting?  

I tried going to the dark side a few months ago when I got a new laptop.  I loaded the HPSDR software from the site and try as I may, I could not get it to recognize my HPSDR rig.  I had the same problem with the desk computer years ago until a W5 helped me via email. I never wuda figured it out myself.

So I went with Jeff/NBCs unused SDR receiver in the meantime.    I figgered I wud eventually get an ANAN or Flex anyway, so gave up on the old HPSDR rig.

But your way seems very easy when a Flex is involved.  

If I wanted to drive the 4-1000A GFZ audio driver, I would just route the software audio feed to the 4X1, bypassing the HPSDR/ANAN/ Flex right?   (And using the ANAN as an RF driver)


T




Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W1AEX
Un-smug-a-licious
Contributing
Member
*
Offline Offline

Posts: 1488


Apache Labs SDR


WWW
« Reply #17 on: March 12, 2020, 12:16:17 PM »


Does it have LP filter DSP for bandwidth and positive peak limiting?  

I tried going to the dark side a few months ago when I got a new laptop.  I loaded the HPSDR software from the site and try as I may, I could not get it to recognize my HPSDR rig.  I had the same problem with the desk computer years ago until a W5 helped me via email. I never wuda figured it out myself.

If I wanted to drive the 4-1000A GFZ audio driver, I would just route the software audio feed to the 4X1, bypassing the HPSDR/ANAN/ Flex right?   (And using the ANAN as an RF driver)

Tom,

The DSP will allow you to set any AM bandwidth between a very fat 40 kHz (20 kHz + 20 kHz) to a very skinny 200 cps (100 cps to 100 cps). The audio bandpass of the voice processing can also be set to extend as high as 20 kHz (by adjusting the top frequency setting in the pre-eq form to 20000) so that you can actually fill out that 40 kHz wide signal with audio. Although I don't know anyone who has actually operated that way it's possible and you can see it in the screenshot. If you enable the adaptive pre-distortion algorithm there will be no apparent energy outside the intended bandpass of the TX filter at any bandwidth setting.

The positive peak limiting is controlled by a soft-limiting algorithm that allows modulation to reach just below the clipping point of the DAC. No matter how hard you drive the TX audio chain you cannot reach the clipping point. There is a similar soft-limiting algorithm that will prevent you from reaching beyond 99% negative peaks while operating AM. At the end point of the TX audio chain you can adjust the ALC to allow up 120 dB of ALC compression, which is in the realm of insanity as a pin dropped from across the room will be as loud as the operator in front of the microphone. That being said, it will not splatter or go into hard-limiting and it will always be soft-limited below the clipping point of the DAC. If all of this is followed by the adaptive pre-distortion algorithm the IMD of the entire TX chain will be improved from the nominal 3rd order -32dB to up to -70dB with the newer Black Edition 7000DLE rigs no matter how hard you drive it. As an added bonus with adaptive pre-distortion enabled, your THD with AM will improve from a nominal 3% to 0.2%.

As far as the phase rotator goes, I have found what Kevin reported. With my voice it's not really useful for enhancing asymmetry but it does allow a user to create better symmetry, which is very useful with SSB to allow the maximum audio drive without letting random peaks reach a point where you might start flat-topping your linear amplifier, although the adaptive pre-distortion algorithm will fight that scenario from happening if it's enabled. Not sure about your last question though. You can pull out the processed audio via the MON function, but you lose all the end-stage soft-limiting protection and there are probably far better ways to go. The direct sampling transmitter in any ANAN is a very nice driver for a big rig and far cleaner than the leaky LO plagued Flex 5000's that I see on the band being used as RF drivers. If you have a panadapter you can see multiple carriers above and below the operating frequency every time one of those abominations is keyed!

Rob



* 40K.jpg (417.02 KB, 1282x722 - viewed 311 times.)
Logged

One thing I'm certain of is that there is too much certainty in the world.
WB4AIO
WB4AIO
Contributing
Member
*
Offline Offline

Posts: 242


Better fidelity means better communication.


WWW
« Reply #18 on: March 12, 2020, 12:25:29 PM »

[...]
The DSP will allow you to set any AM bandwidth between a very fat 40 kHz (20 kHz + 20 kHz) to a very skinny 200 cps (100 cps to 100 cps). The audio bandpass of the voice processing can also be set to extend as high as 20 kHz (by adjusting the top frequency setting in the pre-eq form to 20000) so that you can actually fill out that 40 kHz wide signal with audio. Although I don't know anyone who has actually operated that way it's possible and you can see it in the screenshot. If you enable the adaptive pre-distortion algorithm there will be no apparent energy outside the intended bandpass of the TX filter at any bandwidth setting.

[...]



Rob:

I haven't tried Thetis yet.

Does Thetis allow high fidelity transmission on FM? Or does it, like all versions of PowerSDR, limit you to 300-3000 space shuttle audio in that mode?

73,

Kevin, WB4AIO.
Logged

W1AEX
Un-smug-a-licious
Contributing
Member
*
Offline Offline

Posts: 1488


Apache Labs SDR


WWW
« Reply #19 on: March 12, 2020, 01:14:39 PM »

Rob:

I haven't tried Thetis yet.

Does Thetis allow high fidelity transmission on FM? Or does it, like all versions of PowerSDR, limit you to 300-3000 space shuttle audio in that mode?

73,

Kevin, WB4AIO.

Hi Kevin,

The FM transmission characteristics are just as you described it, a very space shuttle-like sound that is locked at 3k + 3k no matter what settings you enter into the DSP filter form. I have not heard any current chatter about altering that in any of the software platforms that can be run on the hardware.

73,

Rob W1AEX

Logged

One thing I'm certain of is that there is too much certainty in the world.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #20 on: March 12, 2020, 02:14:19 PM »

Hi Rob,


 You said:

"You could always go to the dark side and just run your old HPSDR board (or any ANAN hardware) with Thetis (or PSDR mRX PS) which is free and offers 10 band pre-eq, 10 band continuous frequency compression, 10 band post-eq, and an adjustable phase rotator. You also get negative peak soft-limiting in the deal!"


Add high level predistortion, LP bandwidth filtering and it sounds like quite THE setup.

Right now many of us are in the hybrid whirl with our feet in hardware and software at the same time.  Like the 1960s days with PC boards using transistors and tubes mixed in.

Probably the day I buy a used ANAN will be that day I go all audio chain software.

T
 
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W1AEX
Un-smug-a-licious
Contributing
Member
*
Offline Offline

Posts: 1488


Apache Labs SDR


WWW
« Reply #21 on: March 12, 2020, 03:28:18 PM »

Tom,

The amazing variety of equipment that AM operators are running is one of the greatest things about running AM. It really is the only mode where you can tune across the band and hear users conversing together with homebrew plate modulated rigs, homebrew class E rigs, a vintage Collins/Johnson/Heathkit/Knight Kit/etc, a restored military transmitter, an SDR, a multi-mode transceiver, a junkbox transmitter, a rescued broadcast transmitter, a 50C5 modulated by a pair of 6AQ5s and whatever else you can imagine. I think the fun factor is what has kept this mode around for so long. I'll look forward to hearing whatever hybrid design you come up with, and more importantly, what name it will get as it joins Fabio and all the other VU creations.

73, Rob W1AEX
Logged

One thing I'm certain of is that there is too much certainty in the world.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #22 on: March 12, 2020, 03:34:24 PM »

I've never thought of it that way, but true... we are able to use radios and general technology from the 1920's to the 2020's, about 100 years worth.

As someone recently paraphrased you, Rob,  "These are the best days of AM."

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #23 on: March 13, 2020, 01:36:08 AM »

Only a Symmetra-Peak, which is in use. It works well, but other than what's in the literature and patent, there is not much else to discuss about it other than that it delivers the promised result, with the caveat that it is not a panacea and can be randomly less effective if there are large frequency-related phase shifts occurring over narrow frequency ranges elsewhere in the audio path.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
KD1SH
Member

Offline Offline

Posts: 737



« Reply #24 on: September 04, 2020, 09:43:32 AM »

  I just built myself the phase rotator described by W3AM (http://www.w3am.com/8poleapf.html), and tried it briefly last night, with mixed results. More testing will be needed, but here's my initial take: My usual audio chain is slightly asymmetrical in the positive direction, which is fine with my screen modulated transmitter since it revels in the big positive peaks. But, my Clegg Zeus, the six-meter AM transmitter, is plate modulated, and I hate to abuse the mod-iron and would prefer a more tightly compressed audio for weaker signal work. My DBX-286S is great, but it doesn't have a limiter; I figured that keeping the audio symmetrical might aid the compressor in keeping things tight.
  So, I put the phase rotator into the audio chain using the DBX-286S's "insert" jack, which puts it between the initial preamp and the compressor. On the REA Mod Monitor, the asymmetry was less, but now in the negative direction. Of course I don't know what sort of asymmetry the compressor and all the other downstream DBX circuitry is introducing. That's the devil in all this, I guess; voices can be asymmetrical, microphones can be asymmetrical, compressors can be asymmetrical; it's a shell-game. Actually, the Zeus itself might be complicit; one of the 811's in the modulator might be a little softer than the other, or something.
  Tonight I'll do some more testing, using various compressor settings, and trying it with the compressor bypassed. And, even though phase rotators are customarily placed upstream of compressors and limiters, I'm going to try patching it in after the DBX, right before the radio.
  This weekend I'm going to build this: https://sound-au.com/project67.htm (the version in figure 2). Looks promising, and maybe placing my phase rotator right before it might get me what I want.
Logged

"Gosh, Batman, I never knew there were no punctuation marks in alphabet soup!"
—Robin, in the 1960's Batman TV series.
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.075 seconds with 19 queries.