- Posts
I’m looking for some schematics for final asymmetrical compliance limiters. I’m looking to make one that can be built from standard parts and be smaller than a pack of cigarettes.
I’m looking for some schematics for final asymmetrical compliance limiters. I’m looking to make one that can be built from standard parts and be smaller than a pack of cigarettes.
We are thinking of building self contained remote transmitter sites powered via solar cells and small batteries.
Total posts : 45366I have no idea what this “compliance limiter” is but perhaps if you would describe it in different words or elaborate on its function I and others may be able to help.
Neil
Total posts : 45366He’s talking about hardware like an Inovonics 222. I doubt he’ll be able to find a schematic and I also doubt that an equivilent circuit could be built easily and cheaply. I wouldn’t mind being proven wrong, however.
Total posts : 45366Basicly i want to shrink the size of inovonics 222 down to the size of a pack of cigarettes and have it run off 12Vdc
Thank You,
Rev. Robert P. Chrysafis
Universal Life Ministries
http://www.ulc.orgModerator Hunterdonfree
http://groups.yahoo.com/group/hunterdonfree
Total posts : 45366Some background…
The NRSC standard was intended to improve the frequency response of AM broadcasts by pre-emphasizing the higher audio frequencies at the transmitter to complement the shape of the r-f bandwidth of receivers meeting the assumptions of the standard.
But almost no commercial receivers were ever built to meet that standard; they have considerably narrower r-f bandwidth — which make them cheaper, and less likely to receive adjacent channel interference. Using NRSC at the transmitter is practically useless for these narrowband receivers, because they “hear” almost nothing in the spectrum where NRSC pre-emphasis occurs.
Asymmetric modulation of an AM transmitter can be useful to limit negative-going modulation peaks to ~99% while letting positive peaks go to ~125%, which gives a marginal improvement in the amplitude of the audio recovered in the receiver (while adding some distortion). Some AM transmitters are not capable of doing that, though. It depends on their design.
An AM transmitter cannot be modulated more than 100% on negative-going audio peaks without creating high distortion in the receiver, and interference on adjacent channels.
Commercial AM stations use audio processing that raises the average value of the audio waveform with respect to the peak value. The average value determines how loud that audio sounds at the receiver, other things equal. Such processing also adds distortion.
//
Total posts : 45366a hamilton rangemaster when set for 100mW input to finals is capable of -99% +130% modulation. Yes i know about most consumer receivers bandwidths, but it’s still a good idea to use NRSC processing.
Anyway i want to design a remote Tx site that is compact, self contained and solar powered.
a 19 inch 1RU inovonics 222 does not qualify as compact. basicly i want a 222 in a small compact format that runs on 12 volts DC
Thank You,
Rev. Robert P. Chrysafis
Universal Life Ministries
http://www.ulc.orgModerator Hunterdonfree
http://groups.yahoo.com/group/hunterdonfree
Total posts : 45366You can look at the circuit description and schematic of the model 222 here:
It’s an interesting read and a very clever circuit built with discrete parts.
Gerry
Total posts : 45366Thanks to the good posts here I now know what we are discussing. Bear with me as I learn.
There are some circuits I have encountered using op amps which may accomplish the assymetrical modulation. One is an “antilog” circuit which is intended to recover the original audio after processing through a “log” amplifier. I don’t know if this fits the criterion but I offer this as a research topic. This would boost the higher amplitude audio which seems to be the objective. The circuit uses a transistor in the feedback loop and is not very complicated.
Another circuit that I have used is called a “diode function generator”. This is fairly hard to design and set up but it provides a means to achieve almost any funtion of output vs. input.
If I understand what Rev. Robert is proposing it would seem that the audio processing could be done at the studio and there would be no need for processing at the transmitters. That way, an investment could be made at the studio and not at each transmitter, but perhaps this is also naive.
Neil
Total posts : 45366It’s also good to read the manual words. See http://www.inovon.com/download/222%20Manual.pdf
Look on page 10 under the heading “Assymetrical Modulation”. It says it will PASS program material that is assymetrical and further, the 222 WILL NOT CREATE ASSYMETRY. This is a lot more honest than the typical 222 ads that lead you to believe it will create assymetry.As others here have concluded, creating assymetry is difficult. Completely digital, DSP based processors could easily do it to any absurd extent with the proper algorithm programmed in. I think an analog version could do it by shifting the the level upward whenever the negative limit is touched. But, the hard part is getting the response time down to a fraction of a millisecond to properly follow the negative peaks at all audio frequencies up to 10kHz in real time.
PhilB
Total posts : 45366This may show my ignorance of the topic, but what would happen if you changed the DC offset of a waveform in the positive direction? Would that effectively force a positive asymmetry that would give you the desired modulation? This sounds too simple, so there must be something wrong with the idea.
Total posts : 45366I guess that’s what I was driving at with my comment about shifting the audio level (actually the whole waveform) up when the negative limit is hit. This could be done by detecting the amplitude of the audio and using the DC voltage level to shift the bias in an amplifier stage on peaks. I think the whole thing would requre a half dozen op amp stages and maybe some transistors. Not real easy.
I’m sure this, or something like it, has been done before, but I’ve never seen a circuit. There are a couple obstacles making it less than trivial to design. First, the detected amplitude DC voltage would need to be adjusted to produce the required DC bias shift level. Second, the response time of the detected voltage would need to be tailored carefully. Probably, like a good compressor, it would need to have a very fast attack time and a longer decay time. If the attack time isn’t fast enough, there will be negative clipping on sudden peaks until the bias adjusts. Fast attack usually means you have to dump a huge amount of current into a capacitor on the peaks. Slow decay means you need a capacitor big enough for the decay. These requirements are at odds with each other. Although, in this case, unlike a compressor, the decay time probably doesn’t need to be very long.
A DSP signal processor would do the job very well because it can “look ahead” to see where the signal is going and then adjust the digitized audio wave mathematically to shift the level upward. This could be almost seamless if done well. There would be some delay through the DSP which should’t be too noticeable.
These are just some random thoughts. I don’t have any experience with any of this, but it’s fun to think about!
PhilB
Total posts : 45366Quote.. From Rich..
“Asymmetric modulation of an AM transmitter can be useful to limit negative-going modulation peaks to ~99% while letting positive peaks go to ~125%, which gives a *marginal improvement* in the amplitude of the audio recovered in the receiver {{(while adding some distortion)}}.”I understand the quest of getting all of the legal available power possible but if it’s going to add distortion, I don’t see this as a good thing. And honestly, how much power will be gained by adding 25% to the positive peak @ 100 mw? Is it really worth the concern?
If you stay within the boundries of bandwidth of the receiver and operate as near to 100% as you can, wouldn’t it sound better?
Don
Total posts : 45366It depends on the definition of better. Some think that “louder” is better, even if distortion increases. Unfortunately that includes almost all commercial broadcast stations.
Total posts : 45366I just did a quick calculation to assess the benefit of 125% “asymetrical” modulation. My conclusion is don’t bother. Details upon request, but based on what I calculated there will be no significant range gain and even if I am wrong about a supposed range difference, the distortion price paid appears to not be worth any effort.
The biggest gains in range under the contraints of 15.219 are to be had by improving on the efficiency of the transmitters, and not by distortion of modulation to achieve artificial modulation percentages. How much is gained by an increase of range of a few hundred feet?
Just my informed opinion, open for debate, for your consideration.
Neil
Total posts : 45366Neil,
Actually, I don’t think power is the issue. The modulation level determines the audio output level of a radio. For any power level, the loudness will depend on the modulation level. So, 125% is better than 100%.
Supposedly, the human voice is naturally assymetrical. I’ve never seen any tech data that says how much so we could relate it to some modulation level over 100%. Theoretically, if the human voice is assymetrical enough and the 222 allows assymetry to pass, then 125% pos peak modulation won’t distort the waveform very much.
It seems to me that assymetry would be good only for voice. Music would suffer. But, since the 222 allows assymetry and doesn’t create it, then it would be good for both voice and music as long as you set the level to 100% for music. Then it may go higher then 100% for assymetrical voice. The result would be no distortion. But any processor that CREATES assymetry would distort music and wouldn’t sound good. I guess that’s why the 222 is designed to pass assymetry rather than create it.
As an interesting experiment, a while back I hooked a scope probe to the RF output of the second IF in a tube radio. I could see a very clean modulation envelope. I tuned in my favorite AM station, KYW, (all news, all the time). Interestingly there was a large amount of NEGATIVE peak clipping going on. There was no “upward” shifting of the waveform to prevent negative clipping. Of course, you can’t go below 0% modulation, so the waveform looked just like somewhat overmodulated unprocessed audio. I didn’t look at the positive peaks in details, but I do know there was no positive modulation clipping. I should go back and do this again sometime to see the modulation percentag on the positive peaks. KYW has the most “punch” of any station I have heard. The sound isn’t real clean, but since it’s all voice except for jingles and ads, whatever processing they are using is effective.
Phil B
- You must be logged in to reply to this topic.
