The AM Forum
March 28, 2024, 09:38:29 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 2 [3]   Go Down
  Print  
Author Topic: 300% Positive Peaks AM Modulation  (Read 50356 times)
0 Members and 1 Guest are viewing this topic.
WA2SQQ
Contributing
Member
*
Offline Offline

Posts: 1090


« Reply #50 on: August 26, 2016, 11:31:22 AM »

I was following this a while back and was wondering if it could be used with today's SDR radios, like Flex or Anan? You mentioned that you released the circuit on FaceBook - would it be possible to post a link to it? If you prefer, perhaps to <my call> @ gmail.com ?
Definitely interesting ...

Regarding the CB comments, so many people here probably got their start on 27 mhz, as I did from the late 1960's. We all needed to start somewhere!
Thanks
Logged
Steve - K4HX
Administrator
Member

Offline Offline

Posts: 2722



« Reply #51 on: August 26, 2016, 12:32:50 PM »

U.S. broadcasters are limited to 125% positive peak modulation. None in my area approach this limit.

I have enough asymmetry in my voice that no processing is needed to get 130-140 percent positive peak modulation. For most ops and most transmitters, this sort of circuit is probably not needed or appropriate.
Logged
KA6BFB
Member

Offline Offline

Posts: 7


« Reply #52 on: August 26, 2016, 01:43:45 PM »

Hello WA2SQQ

I have a Flex 1500 and I looked at the output circuitry of that years ago, and if memory serves the amplifier stage was linear and this won't work well at all for this circuit. I seem to recall that there was the ability to DC couple the audio in to the radio, and if that's true you could use the asymmetry circuit as it is in this thread. Steve, K4HX brings up a good point about the natural asymmetry in most men's voices. You want to be able to have the ability to invert the audio going into the asymmetry circuit if you want to in order to pick what you want to do in reference to that natural asymmetry. The third generation of this circuit does that. Or you can just reverse both of the diodes and that will do it.

Dave  KA6BFB
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #53 on: August 26, 2016, 06:57:54 PM »

The point of this board,  the one I made (using a THAT processor chip)  and others is to be able to compress and then increase talk power.

Processing should reduce the peaks so as to increase your avg modulation.   Then run through an assymetry board to increase loudness.

In cars,  we call it 'torque under the curve'.   

I've built this one as well.   Mine does 420 pct pos peaks.   This does 500.

Dave,   you and I spoke about this board before, heartedly.   No matter,  people should choose to public domain something.   Not have it chosen for them. 

IOW,  you're welcome :-)

Tell John I said hi.

--Shane
KD6VXI

Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #54 on: August 27, 2016, 12:25:22 AM »

Shane, thanks for posting an explanation about its workings, and clearing up the origins of the board original design/intellectual property for all to see. This issue of someone claiming your work as theirs is also being discussed on a certain unique CB forum - in fact its topic there had a link to this topic. Some people can be dastardly and I hope you can prevail in asserting your rights if that is your intent.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Tim WA1HnyLR
Member

Offline Offline

Posts: 159


WWW
« Reply #55 on: August 29, 2016, 04:54:13 AM »

Ahhhhh. Interesting subject. The fact of the matter is the traditional AM system has it's limits. The ability of the transmitter to modulate to a level that is wanted and the ability of the RX detector to recover the modulation. As said many times a conventional diode detector will only handle so much modulation before it does an ax murder job on the signal. We have to look at why we do AM. AM is by no means an energy efficient mode compared to SSB but has many positive aspects of its' use. the big draw for me is the beauty in simplicity. The ease of achieving a natural "High Fidelity" type of sound. Yes the average human voice does have a certain amount of asymmetrical characteristics. A true high fidelity communications system requires that both ends of the circuit are up to the job. As we all know it is a compromise between loudness and overall naturalness of the signal. Most of us know the limits of what our transmitters will do for modulation. Those of us who use older diode detector receivers accept a certain amount of distortion. Forcing positive peak modulation is creating harmonic distortion of the operators voice. It is tolerable to a point. After a while it begins to be grating .The other factor is the system of modulation used and how badly does it fall apart when making"Phatt" modulation. I would rather run more carrier power instead. It's all about signal to noise ratio in the end. Good broadcash style audio processing to keep the average  modulation up as well as tailoring the audio response curve of your microphonium pre amp circuitry. IE: upper mid range pre-emphasis /reduction in the mud frequencies /having a real deep low end but not overemphasized. Obviously we all have our unique voices . Each individual operator requires a different set of curves to make it right. There is no"one size fits all" in the AM audio game. AM is what it is . If you have the ability to modulate 300% just turn up the carrier wick and strap.
Tim WA1HnyLR
Logged
N1BCG
Member

Offline Offline

Posts: 836


« Reply #56 on: September 13, 2016, 11:10:13 PM »

A lot also depends on what you're trying to accomplish. If achieving massive peaks is your thing and you don't care how it sounds, go for it, but you're still limited to 1500 watts PEP in most cases. That means you're going to have to drop your carrier level (for those who actually care).

Then there's the issue of what is seen vs what is heard. Peaks, by nature of their short duration, contain very little energy so they look impressive yet accomplish very little. If the objective is to be loud and clean on anyone's receiver while maximizing the S/N ratio then the focus should be on getting as much average modulation power into the 1500 watt envelope as possible. -100% is one limit and anything over +100% will cost you in carrier power. 375 Watts modulated at +100% IS legal limit as is 200 Watts at +167%.

There's a formula for this:  Max Carrier W = ((3873 / (MOD% + 100))^2

Some more examples of what creates 1500W PEP:

%Mod    Watts

100         375

125         296

150         240

200         167

300         94  (Yep. Full legal limit with just 94W @ +300%)
            
Loud and clean on all radios has been the goal of broadcasters for decades and broadcast processing manufacturers have thoroughly documented how it's done. That's a wheel that doesn't need to be reinvented.
Logged
N1BCG
Member

Offline Offline

Posts: 836


« Reply #57 on: September 14, 2016, 07:51:01 PM »

As you can tell, this is the topic I can be baited into most easily.  Argh!!

I can't not mention three critically important factors of asymmetry which are almost never discussed.

1) Natural asymmetry

Most voices, particularly male, create asymmetrical electrical waveforms where one polarity has a higher magnitude than the other when measured at the output of a transducer (microphone). After amplification and being added to a carrier, that waveform will appear as asymmetrical modulation on a scope or modulation monitor.

Caveat: Asymmetry is -not- always uniform over the total vocal range. This is a BIG deal because lower frequencies often have the -opposite- asymmetry to higher (presence) frequencies. Hold that thought...

2) Artificial asymmetry

The need to prevent negative peaks from cutting off the carrier at -100% is well documented and clippers are often used to accomplish this. Slowly increasing modulation to the point where negative clipping *just* begins will show you exactly how asymmetric your voice is naturally. This is the point where your signal is at its cleanest.

Caveat: Going beyond this point increases perceived volume, BUT, it also increases distortion on *all* receivers, even SDR, because the negative waveform is getting increasingly cut off. This can look amazing on a mod monitor but it comes at a cost, particularly if you don't have a low-pass filter to reduce the harmonics generated by the clipping action.

Oh yes, then there are those other frequencies mentioned earlier that have an opposite "polarity". Those are now generating even more distortion since they present even greater energy to the negative clipper.

3) Controlled asymmetry

It's possible to be both loud and clean on the majority of receivers whether boat anchors or SDRs, and taking a page from broadcasters will yield decades of refinement in this pursuit.

All broadcast processors sold in the past 30 years incorporate "phase rotation" circuitry to greatly reduce asymmetry at all frequencies.

"Great Zorks! What does that do to my manliness!?!"

Nothing. Hear me out... a phase rotator is also known as an all-pass filter that slowly changes the phase with a change in frequency. For example, at 50Hz the phase change may be 0 degrees, but it slowly increases with frequency so that by 3-4kHz the phase is shifted 180 degrees and so forth. By the way, complete phase rotator circuits are available for less than $20 on the net.

"I just lost my monster positive peaks. Now what?"

Simple. You now present that mostly symmetric and consistent waveform to your negative peak limiter and decide how much clipping/asymmetry you want. Keep in mind that peaks by nature of their short duration contain very little energy no matter how impressive they appear. Your "loudness" will come from an increase in average, not peak, modulation which is what compressors do.

Put all this together and you have a clean and loud station as heard by everyone.
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #58 on: September 15, 2016, 08:21:45 AM »

Every time I see the use of diodes to clip the negative I think that there ought to be a better way.

Seems to me that it would be more expensive, but one could pretty accurately split the audio signal, much like a class B
amplifier does, or even an AB, and then take each half wave stream/side and process it separately. One could apply a soft knee limiter to the negative going half and use whatever gain one wished for the positive. You'd need to take into account the "zero crossing" area to prevent "crossover distortion" but that would not be terribly difficult to accomplish electronically.

I'm guessing the reason this sort of method is not used much is the cost and complexity of that negative going soft knee limiter.

Anyone think of a reason that my proposed method would not work as described?
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
KA6BFB
Member

Offline Offline

Posts: 7


« Reply #59 on: September 15, 2016, 12:42:46 PM »


Seems to me that it would be more expensive, but one could pretty accurately split the audio signal, much like a class B
amplifier does, or even an AB, and then take each half wave stream/side and process it separately.

That is what this circuit does, except it uses one circuit to do it. The diodes don't clip the audio, they provide different paths for the positive and negative halves in the same amplifier stage. One direction has a gain of one, and the other direction has a gain set by a pot. The op amp gain makes up for the diode drop and there is no distortion. See the attached video for the audio signal with the asymmetry. The positive peaks are changed with a pot. After that, see it applied to a carrier. See the video at https://www.youtube.com/watch?v=x3ZnCy9TQxQ&feature=youtu.be
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #60 on: September 15, 2016, 07:39:01 PM »

This circuit has no clipper.

A LOT of the distortion heard,  past 125 pct,  is caused by negative cycle loafing,  negative peak limiting,  whatever you want to call it.

This circuit,  as well as the one I came up with,  does NOTHING to the negative peaks.   You set your audio input level to give you the amount of negative going modulation you want,  90 to 100 pct.

Then,  dial your positive peaks to whatever percentage you want.

It stretches the positive peaks.

In all actuality,  it's a precision peak rectumfrier.  Not much different than  what makes the needle swing in a bird.

Years ago a pro audio board discussed this method of creating assymetry without negative peak distortion.

--Shane
KD6VXI
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #61 on: September 15, 2016, 08:44:27 PM »

gotcha... so as long as you don't set the "reference" level (the negative) to high, ur good to go... assuming you have
the available power on the positive going side, and you have the requisite absolute polarity correct... VG!

simple enough.

                           _-_-
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #62 on: September 15, 2016, 09:09:40 PM »

Yes.

I have a video of mine in action, on a sinewave.

This is the rf modulated signal.  I had the baseline audio a little hot  and am over mod in the negative region by a couple pct.

https://youtu.be/cajJYlqSv6Q

Somewhere I have baseline audio as well.

https://youtu.be/HsJ1KK8HFgA

Since a pic is worth a thousand words,  maybe a video is worth more?   Lol

--Shane
KD6VXI
Logged
KA6BFB
Member

Offline Offline

Posts: 7


« Reply #63 on: September 15, 2016, 09:09:57 PM »



Years ago a pro audio board discussed this method of creating assymetry without negative peak distortion.

--Shane
KD6VXI

Shane, do you have a schematic available on this pro audio board? I would like to see how they did it.
Also, send me an email some time. I would like to chat.

Dave  KA6BFB
Logged
N1BCG
Member

Offline Offline

Posts: 836


« Reply #64 on: September 15, 2016, 09:10:16 PM »

...but in the end, it's average, not peak, modulation that affects perceived loudness.

Amplitude x Time = Energy
Logged
KA6BFB
Member

Offline Offline

Posts: 7


« Reply #65 on: September 15, 2016, 09:12:29 PM »

Shane

Your second video looks like it has a little crossover distortion

Dave KA6BFB
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #66 on: September 15, 2016, 09:31:29 PM »

Lol.   Wanna see distortion.   This is my revision one.

https://youtu.be/Br8WNE7hp5g

BCG:  I run phase rotation (from 2 to 8 poles),  multiband compression and a peak limiter.   My avg mod is way up there.   Once it has run through all processing,  I run it into an assymetry module,  allowing me to set avg mod where I want it,  since my peak to avg ratio is very low.

Dave,  the schematic was never presented,  a gentlemen proposed a precision peak rectifier for assy creation.

Your method was actually patent in 2011 I believe....  

http://www.patentsencyclopedia.com/app/20110255701

And here's the pro audio board (web board,  not pc board)  post.   About half way down.


https://www.gearslutz.com/board/geekslutz-forum/670584-modifying-peak-limiter-produce-asymmetrical-waveforms.html


Dave,  John has my number.   I don't always answer,  but I do call back.

--Shane
Logged
KA6BFB
Member

Offline Offline

Posts: 7


« Reply #67 on: September 15, 2016, 09:49:19 PM »

Shane

You got my hopes up. I was hoping that the patent you referenced was my circuit so that the Florida Counterfeiter would be thwarted in his alleged patent application. Unfortunately, it is not my circuit. Figure 18 on the patent is the closest. This circuit actually might have a slight advantage even though it is more complex. Since part of this circuit is always conducting, a slower op amp can be used because there is not a need to overcome the diode drop quickly. It would have been interesting if it was my circuit, since my circuit was sold a full two years before that patent was issued.

Dave  KA6BFB
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #68 on: September 15, 2016, 11:21:18 PM »

I hadn't looked at the patent doc,  other than a abstract available on my phone.   It came up looking for the second page I posted (and had seen a few years before).

--Shane
KD6VXI
Logged
IN3IEX
Member

Offline Offline

Posts: 128


« Reply #69 on: October 12, 2016, 07:52:32 AM »

If you don't want distortion just use "controlled carrier" !!
Keep the right amount of residual carrier to satisfy the receiver AGC, maybe use a linear amplifier designed for SSB, it will be just fine. (R. Drake said, I suppose.)

I do that.
Logged
w1vtp
Member

Offline Offline

Posts: 2638



« Reply #70 on: October 13, 2016, 11:05:11 AM »

As you can tell, this is the topic I can be baited into most easily.  Argh!!

<snip...........

All broadcast processors sold in the past 30 years incorporate "phase rotation" circuitry to greatly reduce asymmetry at all frequencies.

"Great Zorks! What does that do to my manliness!?!"

Nothing. Hear me out... a phase rotator is also known as an all-pass filter that slowly changes the phase with a change in frequency. For example, at 50Hz the phase change may be 0 degrees, but it slowly increases with frequency so that by 3-4kHz the phase is shifted 180 degrees and so forth. By the way, complete phase rotator circuits are available for less than $20 on the net.

"I just lost my monster positive peaks. Now what?"

Simple. You now present that mostly symmetric and consistent waveform to your negative peak limiter and decide how much clipping/asymmetry you want. Keep in mind that peaks by nature of their short duration contain very little energy no matter how impressive they appear. Your "loudness" will come from an increase in average, not peak, modulation which is what compressors do.

Put all this together and you have a clean and loud station as heard by everyone.

Well put Clark.  Most Boatanchor transmitters would fill well with this application.  An exception, perhaps, would be my 32V1 that HLR modified.  Dunno what "magic" he did but I easily hit 125% positive peaks with it so I probably would not be interested in putting a phase rotator in it. 

However, the Viking II    I have in the back room might do well with that circuit you refer to

Al
Logged
W8IXY
Contributing
Member
*
Offline Offline

Posts: 124


« Reply #71 on: October 13, 2016, 06:01:26 PM »

Regarding 300% positive modulation:

You need to consider just what objective that 300% positive modulation is going to do for you.  This is only going to consider what occurs in attempting to transmit extreme asymmetrical, and not mention receiving it right now.

*If you are aiming for loudness, the greatest point to consider is the peak to average ratio of the audio you intend to modulate the transmitter with.
*You need to have a transmitter capable of supplying that peak power without distortion to whatever audio waveform you intend to create.

If, for example, you are intending to use a transmitter capable of supplying a 100 watt carrier, for 300% positive modulation, (I haven't worked out the math here, so please accept the generalizations), that transmitter will need to provide nearly a kilowatt of undistorted output power at the desired % of modulation.  If you are trying to achieve extremely asymmetrical AM modulation in the transmitter circuitry, you will wind up with a severely distorted audio modulation envelope and most likely generate significant splatter. The place for creating asymmetrical audio is in the audio processor, external to the transmitter.  The place to create the capability to supply that peak power is in the transmitter.

In my over half century of high power AM broadcast experience, and with designing commercial AM and FM audio broadcast processors, here is a tip I offer to anyone wanting to try out 300% positive modulation.....

*Design your transmitter to handle any waveform of modulating audio with which you wish to send it, so that it can handle every form of asymmetry you throw at it.  It must handle that audio WITHOUT DISTORTION. Of course, the negative peaks will clip at -100%.  The positive peaks must be passed through undistorted up to 300% (or whatever limit you wish to set). You can generate a very asymmetrical sine, triangle, or sawtooth wave with many audio generator devices.  Once you have a transmitter that can handle those waveforms without distortion, then you buy/build an audio processor which will develop a very asymmetrical wave.

In broadcast, we first use a phase rotator/all pass filter to REMOVE any asymmetry (so we have a consistent waveform on which to process), then we use equalization, compression, limiting, clipping and filtering to achieve our desired peak to average ratio.  THEN we use special circuitry to create how much CONTROLLED asymmetry we want as our final product. 

Theoretically and/or just for fun, you can insert varying amounts of carrier into a DSB transmitter to find out for yourself just how much of a difference you can hear (and the resultant artifacts) as you vary the ratio of carrier to sideband power.

Read about what some broadcasters are doing with a technique called MDCL.  If properly applied, you can actually appear to run more power without exceeding the FCC's peak power output limitation.  https://en.wikipedia.org/wiki/Dynamic_carrier_control

Have fun!

73
Ted  W8IXY

Logged
KA6BFB
Member

Offline Offline

Posts: 7


« Reply #72 on: October 13, 2016, 06:49:26 PM »

Hi Ted

I designed the circuit this whole thread is revolving around. I had no intention at the time for that kind of operation and I think it is ridiculous. The circuit is flexible enough that it can easily do it but it is not very practical. I use around 120% for my transmitter on 80M if I use that at all. When listening on my Icom 706 with the noise blanker on, anything above that sounds awful. I think the 300% claim started out as a way to brag what it could do, not what it should do. I could be wrong on that though. I clicked on the original Youtube thread and it is no longer available.

73  Dave KA6BFB
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #73 on: October 13, 2016, 08:51:03 PM »

Dave,

John sent me your phone number.   Been pulling 60 to 80 hr weeks since at work.   Haven't forgotten about ya.

--Shane
KD6VXI
Logged
Pages: 1 2 [3]   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 0.051 seconds with 18 queries.