The AM Forum
March 28, 2024, 05:37:54 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: power: Carrier and PEP  (Read 8008 times)
0 Members and 1 Guest are viewing this topic.
David, K3TUE
Per-spiring AM'er
Contributing
Member
*
Offline Offline

Posts: 394



« on: August 15, 2005, 10:47:34 PM »

I have read a number of places 2 different statements about the ratio of carrier to PEP at 100% modulation and I was hoping I would get it cleared up here.

Some say: Adjust the carrier to 25% of the PEP reading.

  This would imply to me that the carrier is 1/5 of the total power.

Some say: Your unmodulated carrier should be 25% of your total power.

  This would imply that the carrier is 1/4 of the total power.

So my question is, which is it?
Logged

David, K3TUE
Bacon, WA3WDR
Member

Offline Offline

Posts: 881



« Reply #1 on: August 16, 2005, 08:24:16 AM »

It's 1/4 for 100% modulation.

If you use more than 100% positive modulation, the carrier would be less than 1/4 of the peak.  So a carrier of 1/4 of your peak power allows you 100% positive modulation.  A 1/5 ratio would allow you about 124% positive modulation.  Here is a chart of carrier to peak ratio versus positive modulation capability.

carrier/peak - maximum positive modulation
-------------------------------------------------------
1/4 - 100%
1/5 - 124%
1/6 - 145%
1/7 - 165%
1/8 - 183%
1/9 - 200%

Figure you are going to have an audio sine wave modulating your carrier 100% on a classic tube-type class C plate-modulated transmitter.  Your B+ is +600 volts.  Your audio is varying from -600V to +600V, and you add that to the +600V B+.  So now your plate voltage is +600VDC, and it varies from 0 to +1200V and back to 0V because of the modulation.

Let's say at +600V you have 200mA of plate current, for a plate input power of 120 watts.

Now you modulate. your B+ varies from 0 to +1200V.  The properly operating class C stage acts like a resistive load, so your plate current varies along with the plate voltage from 0 to 400 mA.  Your input power will be 0V * 0ma = 0W at the negative peak, 600V * 200mA = 120W carrier, and 1200V * 400mA = 480W on the positive peak.

So your carrier is 120W, and your positive peak is 480W, and that's with 100% positive modulation.  The carrier is 1/4 of the power of the positive peak, or 1/2 of the voltage.

Power varies differently from voltage.  Here you doubled the voltage on the positive peak - but that did not double the power, it quadrupled it.  That's because doubling the voltage doubles the current, and power is voltage times current, so you doubled the power twice, which multiplied it by four.

But a voice waveform is not a sine wave, and you are not limited to 100% modulation.  Because of voice waveform asymmetry, you might see 2:1 ratio of positive to negative peaks.  This makes it possible to modulate 200% in the positive direction without distortion or splatter.   You could easily have a voice waveform that causes your modulator to produce variations from -600V to +1200V. Add that to the +600V B+, and your plate voltage will vary from 0 to +1800V with modulation.  We'll assume your transmitter can handle this.

Your plate current will vary from 0 to 600 mA.  Input power would be 0V * 0MA = 0w on the negative peak, 600v * 200Ma = 120W carrier, and 1800V * 600mA = 1080w on the positive peak.  That's a ratio of 1080/120, or 9:1 peak to carrier, at 200% modulation.  So in this case your carrier would be 1/9 of the power of the positive peak, or 1/3 of the voltage.

So in reality an AM station would typically be set up for a carrier of 1/9 to 1/4 of the peak output power, depending on modulation.  It is possible to exceed 200% using reduced carrier DSB with a balanced modulator, and the carrier could go all the way to zero, but that doesn't work well with receivers using a typical diode detector.
Logged

Truth can be stranger than fiction.  But fiction can be pretty strange, too!
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #2 on: August 16, 2005, 10:52:25 AM »

Good list to have, Bacon. For the longest time I thought that 200% modulation was 1/8. But I see it's 183%.

Back a couple of years ago, about five of us at WA1QIX's shack ran an audio test with his transmitter. We used a calibrated, [edumecated]modulation monitor with a meter - no guess work on the scope. Each one of us spoke into the mike and we measured the positive peaks. It turned out that the most asymetrical [Al,K1JCL] was close to 140% positive. The others were between 120-130%. My voice happened to be only 110%.  Everyone is different.

So it appears that to get a negative 100% modulation figure, most males must adjust their carriers allowing headroom for at least 130% + to be on the safe side.  I usually adjust mine for 1/5 carrier and find the positive peaks on the scope stay rounded. That's the way to tell - you want no flat topping under maximum modulation. ie, For low level systems, [linears] turn the carrier level down and increase audio until you have about 95% negative and positive peaks that are just before flat topping. That's the max use of your rig.

One could play around with loading on your linear amplifier to get better peaks - usually loading it more heavily. Do this by decreasing C2 and repeaking C1 for about 8% decrease in output carrier at max drive.

For plate modulated rigs, using lots of grid drive and tuning the final amplifier will help, but generally, you are stuck with what you have, based upon mod transformer ratios and general design. Some commercial ham plate modulated rigs are good for barely 100% positive. Others will do more. The scope is the only way to tell if you are flat topping. If the rig is limited, for example, you may have to run it at 80% negative and 100% positive to stay clean.

73,
Tom, K1JJ
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.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #3 on: August 16, 2005, 10:58:49 AM »

you have to be a chain smoker to really drag those BAs.
Logged
Bacon, WA3WDR
Member

Offline Offline

Posts: 881



« Reply #4 on: August 16, 2005, 11:06:56 AM »

With unprocessed audio, I got around 1.3 or 1.4:1 also.  But that wasn't enough to satisfy me, so I was doing all kinds of stuff to emphasize voice asymmetry.  I found out I could peak the asymmetry with a low frequency phase adjustment.  And my compressor had some square-law distortion, so I set up polarization so that its distortion also emphasized the positive peaks.  I unbalanced the peak detector in the compressor to allow the positive peaks to go up 2:1.  I typically got about 1.8 to 2.0:1, for modulation of 180 to 200% positive.

But KNX Dino beat me easily.  He had to have 2.2:1.  It looked like 220% positive.  It was amazing.  Some guys have the pipes...
Logged

Truth can be stranger than fiction.  But fiction can be pretty strange, too!
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #5 on: August 16, 2005, 11:58:22 AM »

Yep, big audio.

I also went through the curve.  It all "peaked" out when I built a 4-1000A  upside down tube plate modulated rig. Modulated by a pair of 3CX-2500F3's. Talk about big audio. It was really a high level balanced modulator.  But guess what... after a few months I tore it down and rebuilt it into a conventional final. Though it could do 200-300%+++ modulation, I just got complaints of peak distortion. It was clean on the scope and tone tests, but the detectors out there were getting over loaded. Just as you'd expect.

I later tried it with various highly modified riceboxes and even a custom built 1496 bal mod chip and using BIG linears. Again, 200-300% clean audio, but complaints.

So, these days, I've gone  the other way and usually run 600W of carrier and 50-60% modulation [or whatever makes 1500W pep]. Reports are good and splatter reports are a thing of the past.

Another thing I had no luck with is the 3 diode limiter, high level. I had them on all rigs at one time and could do 150%+, plate modulated. But it always came down to the same thing - no matter what method you use, running over 130% or so results in the average boat anchor RX to overload its detector. So why bother, except as a neg paek limiter safety net?

As an additonal attempt... a number of years ago, HUZman designed me up an op amp chip circuit that  had more gain on the positive audio side and less on the negative side. Kinda an artificial assymetrical voice maker. Worked as designed on the scoipe, but had reports of distortion like every other technique when cranked up.

A test for anyone to try: Try this test on your own boat anchor receiver. Take a ricebox that will do over 150% modulation on AM. Put an audio tone through it and verify on the scope that the peaks are clean and rounded. Now tune it in on your receiver and put the scope probe on the detector output [or audio output]. If it's like my own SP-600, you may find that anything over 110% or so results in a progressively, [though slight at first] distorted waveform. Try it. Maybe yours is better.

Bottom line is that  infinite modulation percentage is FB on ssb and with sync detector RX's on AM, but since most everyone runs envelope detectors, you're lucky to get by with your own ballsy natural asymetrical voice without overloading many of them.

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.
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #6 on: August 17, 2005, 07:38:59 AM »

Almost all receivers will start to distort the detected waveform approaching 90% mudulation.  On some receivers with a large capacitance load on the detector,  you can see distortion beginning at about 70% mudulation.

73 John
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


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


« Reply #7 on: August 17, 2005, 10:24:04 AM »

Almost all receivers will start to distort the detected waveform approaching 90% mudulation.  On some receivers with a large capacitance load on the detector,  you can see distortion beginning at about 70% mudulation.
73 John

Wow... even worst than I thought.  That might make us all think before turning up the TX audio to 150%++....

Another problem was when simply tapping off my SP-600 tube detector directly into the aux of a 1K? load solid state amplifier. It loaded down the detector. The waveform had a slanted  look to it. I could hear the distortion on the AM broadcash band quite clearly.  The problem was solved with a 1 meg input FET type op amp as a buffer. Still, anything over the magic 100% modulation area started to show up as a saturated waveform, with an ugly compression look to it....

Sync detectors, though expensive for the good ones, are really the solution to recovering BIG audio. Just like Afro wigs are the solution to getting BIG hair.  Or, the Charles Atlas course for  becoming a REEEAAL BIG man.  Or, buying the cheapest polyester suit you can find to becoming a BIG stud.  Same exact thing. But no BIG thang. Right on?

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.
Bacon, WA3WDR
Member

Offline Offline

Posts: 881



« Reply #8 on: August 18, 2005, 11:56:21 AM »

I remember W2ZM Bob talking about modulation density.  He made a good point: when we get really asymmetrical audio, we put little stringbeans up in the positive direction, which is not all that much RMS power, but it requires huge PEP and peak audio power.  Attention to modulation density might be a better approach.

A major issue I encountered, especially with newer solid state receivers, was the receive AGC.  In the older receivers, AGC responded to the average RF voltage, and that was perfect for AM.  But in the newer receivers, AGC is peak-oriented.  This means that modulation changes the receiver gain.  You have full gain with no modulation, and reduced gain with modulation, and even less gain when you have heavy positive modulation.  The result is a really crummy compressed sound on received audio.  This compressed quality goes away if you use manual AGC on the same receiver.

And as you point out, even an old-time receiver with average AGC may not have the headroom to pass much more than 100% positive modulation.  It may have been designed that way for noise-limiting reasons, or it could just be that the dynamic range of the IF amp was not designed to handle extreme positive modulation, because it was usually not encountered.

When I put scopes on my receiver IFs, I saw that the RF/IF dynamic range seemed to be able to deal with 200% modulation, but I didn't do that with many receivers.

Diode detector distortion can be very significant in negative modulation.  There, the diode causes the negative-going slew rate to be slower than the positive going slew rate, and capacitance to ground on the output side makes that a problem.  This causes the negative-going output to drop slower than the signal does, causing "diagonal clipping."

And the detector diode characteristic is usually not linear when it is supposed to be conducting.  I think thermionic diodes are better than solid state diodes, because the forward voltage dead zone basically does not exist.  However, with proper active circuitry, solid state rectifier distortion can be pretty much eliminated.  But even that is a moot point, because the worst distortion is atmospheric, bringing us back to synchronous detection.

Bottom line: heavy voice asymmetry supermodulation is probably best reserved for highly competitive conditions.
Logged

Truth can be stranger than fiction.  But fiction can be pretty strange, too!
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 0.056 seconds with 18 queries.