Has anyone attempted to build these circuits and mods?
http://www.bruhns.us/APL/APL.html
In answer to your question.. I haven't built any circuits in my life (unless you want to include a little dc motor in jr high)
But concerning the topic of asymmetric processing.... And I'm not pretending to comprehend this whole subject, but apparently, there is no processor, not even the 222 that creates asymmetry. And I re-qoute this comment of John Wood at Inovonics..
The 222 does not create asymmetry, nor should any processor. If it does, it is, by definition, introducing 2nd harmonic distortion. Any sound, save a pure sine-wave tone, is generally asymmetrical... a trumpet note, the human voice, even a 'blended' orchestra. The 222 simply allows naturally-occurring positive-going program peaks to assume a value up to 125% of the negative ones, as allowed by the FCC in AM broadcasting practice.
I wish I understood his statement better.. but it implies to me that if we can use other processing to "assume 125% peaks" with lower priced gear, then we accomplish the advantage of the 222.
Is my assuming comprehension is correct?
More research ... I came up with this from SoundForge:
http://www.studiotoolz.net/sonicbirth
... makes it easier to design your own AU plugin circuits. There is a limiter amongst the ready-made plugins ... perhaps I can modify it to attenuate negative peaks, while using my other de-clipper brickwall control to kill distortion at max modulation.
Anyone else? ....
Hoping ...
"I wish I understood his statement better.. but it implies to me that if we can use other processing to "assume 125% peaks" with lower priced gear, then we accomplish the advantage of the 222. Is my assuming comprehension is correct?"
I think yes and no, from my limited knowledge ... he's right from a purely technical standpoint, because if you clip the negative phase, you will introduce distortion, no doubt. If you pre-limit the signal, you can somewhat avoid the problem, but then you've lost a percentage of modulation and you're kind of back where you started.
Again with my limited view, the 'magic' of the Inovonics 222 is that it doesn't do anything as a form of attenuation, like, say, an AGC does. Instead it somehow allows the positive signal peaks to come through stronger. I'm sure they wouldn't be selling these things if that were easy to do. I've looked at the circuitry in its manual ... not something I'd want to try and build. Besides, it's probably patented.
I just thought: Really down and dirty basics (dunno if this is how a 222 actually works) ... If you split up the signal into several different paths, one set 180 degrees out of phase so you end up with one positive-only and the other negative-only, then you can cut or boost modulation separately, and finally set them back in phase. If in addition, you have another clean signal brought through the same processing but with nothing done to it, then you can blend that with your differential signals and have a lot of control..
I wonder how much latency the 222 output has ... hmm ...
So, the consequences in any case is that while you do have modulation level differences between the negative and positive portions of the signal, which could be construed, in my view at least, as asymmetrical, you do not have any clipping, compression, limiting, or other coloring of the signal. Depends what you mean by "asymmetrical".
See? He basically said that messing with peaks alone in half the signal would cause distortion except in the case of a pure sine wave, which makes perfect sense. IOW, you have to maintain full signal integrity, all the sound, throughout the process.
Anyway, if I can toy with some AU plugin processing and get a bit more positive phase modulation without affecting the negative phase, that might be worth some effort.
There is a technique which is used in my amateur radio transmitter for AM on the "low" bands, whereby it is possible to produce a higher power AM signal compared to a conventional AM transmitter with the same power rating and without negative clipping. It is used to provide a more powerful AM signal with modulation while keeping the no audio DC power input to the final low so as not to overheat the finals.
With no audio the DC power input is "a few" (from the manual) watts. With audio, the DC power is increased in proportion to the audio amplitude up to a DC power (the carrier) of around 50 watts with a PEP of 260 watts. The effect is to produce a modulated signal which is nearly equivalent to that of a conventional AM transmitter by boosting the carrier power with the modulation. (This is not double sideband suppressed carrier since the carrier power is increased with the audio.) Since human speech has a high peak to average power ratio, especially with pauses, the average power into the finals is low and they don't overheat. Keep in mind that in AM the carrier power is constant and the audio is contained in both sidebands but, unlike standard AM, this system increases both the carrier power and the sideband power with audio. The effect is the "few watt" transmitter produces a signal equivalent to that of a 50 watt transmitter.
I recall but can't find the article where this was described for a part 15 transmitter. Though it appears to be legal under part 15.219 it can be argued that it is not. In the case of my transmitter it is done to prevent overheating the finals rather than because of a legal requirement.
This technique is a function of the transmitter and is not produced by audio processing.
Neil
The technique Neil described is called Progressive AM Modulation and is fascinating because it both increases the overall efficiency of the transmitter and allows peak power to be many times higher than achievable by standard AM modulation. If you are inclined to wade through patent-speak, Harris Corp. has a patent that describes the workings, patent #3898590. http://www.freepatentsonline.com/3898590.pdf
Power input to the final RF stage is defined as the DC voltage times the DC current. In normal AM modulation, the power is constant. It is the same through the modulation range of 0% to 100%. At 100% modulation, the average power is the same as it is at 0% modulation.
Things are different with progressive AM modulation because the power supply voltage to the final RF stage is increased in accordance with the peak level of the input audio signal, so the peak power can increase above the normally accepted 100% modulation limit of 4 times the 0% modulation level, and the resulting average power level increases.
There is no theoretical limit to the maximum power attainable with this scheme. A 100mW transmitter, measured with no audio input, could produce an average power of thousands of watts at maximum audio input level.
In a viable part 15 transmitter design, progressive AM modulation could be achieved by detecting the peak audio level, buffering it as required, and feeding the result to one of the many available voltage regulator ICs. The resulting voltage to the final RF stage would then increase proportionally to the peak audio input.
Legality is not likely in the end. When a manufacturer is going through the certification process, there is a chance that the certification lab will just measure the power input at 0% modulation and it will be certified, or maybe the lab will notice the trick. If the FCC happens by with their equipment, they may judge you as legal if you happen to have a dead carrier, or more likely, they will measure your field strength over the limit when you have program material playing.
As Neil said, I also recall a part 15 transmitter circuit that used this technique and produced an average power of 300 mW (if I remember correctly) at high modulation levels.
Phil
Phil, thanks for putting a name on the technique I described. It is interesting that the date on the Harris patent is 1973 yet my Drake TR-3 which uses this was designed in the early 60s.
Your statement: "In normal AM modulation, the power is constant. It is the same through the modulation range of 0% to 100%. At 100% modulation, the average power is the same as it is at 0% modulation." may be a bit confusing. Did you mean that the carrier power is constant rather than the average power?
The equations for AM are:
P(total) = P(carrier) + P (Upper Side Band) + P (Lower Side Band)
and restated to account for the modulation
P(total) = Pc (1 + m^2/2) where m is the modulation percentage divided by 100 and the term m^2/2 represents the power in both sidebands.
What this means is that the total power in an AM signal consists of the carrier power plus the power in both sidebands. At 100% modulation the total power is 1.5 times the carrier power. In the system we are describing the carrier power is not constant so the "1" in the equation could be replaced by a function of the audio amplitude.
Neil
Rhemaradio has operated with an Inovonics 222 into a Hamilton Rangemaster for several years. The functions of the processor are very simple. First, it is a single wide band mono audio processor to be used explicitly with AM broadcast transmitters.
Human voice and some musical content contains asymmetry that causes many AM transmitters to actually sound as though the peak modulation has been reduced during some asymmetrical content. To ameliorate this situation, the input circuit does a look ahead function and smoothly flips the audio 180 degrees (back and forth) to optimize the positive modulation. This function is common in most AM processors over the past 50+ years.
The 222 then runs the audio through a limiter to level out the average audio peaks (0 db to -12 db, adjustable). The audio is then pre-emphasized where the audio above 1 KHz is amplified on a special 75 micro-second R/C curve (NRSC standard). The audio pre-emphasis adds more clarity to the high frequency audio to improve intelligibility.
The entire audio package is then hard limited by a circuit that allows the user to control the negative peaks (<99%) with out controlling the positive peaks. This keeps the modulation circuit from over modulating and creating unpleasant distortion in receivers. By allowing the positive peaks to reach an average in access of 100 percent, the audio signal to noise ratio through the transmitter is improved. In the old days we used to accomplish this process by back biasing opposing diodes between negative and positive audio leads.
And finally a smoothing filter removes the square wave content from the audio then onto the 10 KHz low pass audio filter to comply with NRSC band pass specs.
This is an over simplified brief explanation of the working of the Inovonics 222 AM audio processor.
Higher priced audio processors also have multi-band compressor limiters. The 222 does not. However, with computer generated multi-band pre-processing or a stand alone multi-band compressor limiter at the input to the Inovonics unit, the 222 can be a very potent addition to any AM broadcast station.
With average positive modulation of 110+ percent, the output audio signal to noise improvement can be pretty impressive. This can make for more range and a more listenable signal.
Neil said "I recall but can't find the article where this was described for a part 15 transmitter. Though it appears to be legal under part 15.219 it can be argued that it is not. In the case of my transmitter it is done to prevent overheating the finals rather than because of a legal requirement.
This technique is a function of the transmitter and is not produced by audio processing."
Yeah ... that I got, it's have to be. Very cool idea. As it is, I run my playback software right on the edge, using a de-clipper. I use a software-based multi-band compressor, a brick wall limiter/de-clipper. The signal isn't too bad, and seems to hold up OK on AM radio as well as the internet.
Then Phil said "As Neil said, I also recall a part 15 transmitter circuit that used this technique and produced an average power of 300 mW (if I remember correctly) at high modulation levels."
Not to be on the air 24/7 with it, of course, but I'd certainly be willing to experiment, mod an existing transmitter or something, to see what happens ... or if y'all can find the circuit, I'll take a shot at building it from scratch ... 'specially if there's a photo template for a PC board.
I found the "progressive AM" part 15 transmitter I was thinking of. It is the Panaxis AM-100. I have a copy of their schematic in my disc library but I don't have the link for it. I just did an internet search but couldn't find it.
Maybe if you search for Panaxis AM-100 and spend some more time on this than I did you might find it.
Neil
check the library here. there is a bunch of panaxis stuff in the lprc library. think the am 100 and its psu is there too.
Thanks to an email I received, it is better to use the more common name "Controlled Carrier AM Modulation" rather than progressive modulation. A Google search for controlled carrier modulation returns many more hits. I noticed a number of ham transmitters have used this. I saw Drake, Knight and Heathkit. There are likely others. I also saw a Varian 250kW shortwave broadcast transmitter that uses controlled carrier. The impetus seems to be to cut the carrier power down when there is little or no audio so as to reduce stress on the output tubes.
Neal pointed out that I was misleading when I said “In normal AM modulation, the power is constant.” I was thinking of the input power to the final RF stage, not the total radiated power consisting of carrier power plus sideband power. The input to the final is indeed constant over all modulation levels from 0% to 100% modulation. The input power to the final determines the carrier power.
In another email to me, it was pointed out that the AVC in AM radio receivers may cause strange and potentially annoying results when aggressive or poorly designed controlled carrier modulation is used. As an example, he Knight T-60 apparently ran at low carrier power until the audio input level passed a certain threshold. Then is switched to high power. This could cause an annoying pop or click in the receiver when it switched back and forth unless the microphone level was set carefully.
Even if the power supply voltage is varied continuously to follow the audio, there still could be artifacts due to the receiver AVC. Receivers typically have a slow AVC time constant, on the order of maybe .5 to 1 second. AVC is intended to level the volume for strong stations vs. weak stations and to level out fading. You can imagine what would happen when audio suddenly goes from low to very high. As the AVC lags in catching up to the increased carrier power, the volume will increase and then fall back to normal for up to 1 second.
In Part 15 transmitters, we aren’t concerned about reducing stress on the output stage. In our case, controlled carrier modulation could be used to increase the carrier power above 100mW in such a way that it continuously follows the audio level, and ensures that the carrier power will always increase, as required, so that audio peaks will never exceed 100% modulation. As I said before, there is no theoretical limit to the maximum full modulation carrier power vs. the idle 100mW level.
A decent Part 15 design might limit the highest carrier power to several hundred mW to reduce chances of FCC actions and to minimize the AVC artifacts.
Phil
Phil,
If one monitors the average voltage and current into a directly modulated RF final amplifier both will remain constant with or without modulation. This is because both the current and voltage swing above and below the quiescent values (assuming linear operaton) but remember that power is not the product of average AC volts times average AC amps. Power is the product of the instantaneous voltage and the instantaneous current. This is, of course, instantaneous power and to get the average power this is integrated over a time interval and is divided by the time interval. When measured in this manner, the average input power will increase with modulation by the amount calculated in my post above. The sideband power comes from the power input to the final as does that of the carrier.
Hope this helps.
Neil
In order not to complicate my previous post I left this out but here it is in case anyone is still wondering about average power.
Let's keep it simple. Suppose we have a voltage which is 1 volt and it is on for 1 second and off for 1 second. The average voltage over time is (1+0)/(1+1) = 1/2 volt. Do the same for a 1 amp current waveform such as this and you get 1/2 amp. on average. Multiplying these together to get power will yield an average power of 1/4 watt.
But the correct way to get average power is to multiply volts and amps before averaging. If both volts and amps are on at the same time and off at the same time then the power will be 1 watt for 1 second and 0 watts for 1 second. The average is (1+0)/(1+1) which is 1/2 watt, not 1/4 watt as calculated by multiplying the average volts and amps.
So, to get average power it is necessary to multiply volts and amps on an instantaneous basis before dividing by the period to get the average.
This is why the power input to a final APPEARS constant with and without modulation but it is not.
Neil