: : I'm hoping that Don can come in here, and remind me how it is/was : >he was peaking the audio at around 3kc, and then rolling the whole thing : >off at 3.5kc (or slightly higher). And, what's the bottom end? : > : : I do it by using a rising pre-emphasis between curve in the mic preamp, : starting at 800 cps and peaking 8-10 dB at 3 kc/s and above. Also, I use : a D-104 mic (for highs) and a dynamic mic (for lows) mixed together in : phase. The low-pass cutoff is accomplished using passive L-C filters. With : the flip of a switch I can choose between a gradual hf cutoff filter at 5-6 : kc/s, a very sharp cutoff filtor at 3400 cps, or no lowpass filter at all. : The bottom end goes down to 30 cps or so, but the scope shows visible : distortion on a sinewave below about 40 cps.
and at those low frequencies, it's rather difficult to hear the bass, anyway. So, would you say that decent (someone define "hi-fi" Ham AM Audio, please) audio could be well received from 50Hz to 4kHz, or less (on the High end)? : With the 3400 cps filter in the circuit, the response curve is flat from 40 : cps to 800 cps, then rises steadily up to about 2 kc/s and begins to : flatten off beyond 3 kc/s, but hits a brick wall at 3400 cps. I bought the : filter at a military surplus store in Washington DC back in 1973 ( I wish I : had purchased half a dozen of them). There is virtually no attenuation at : 3300 cps; at 3400 it is over 25 dB down, and at 3500 cps you can't detect : the signal on the oscilloscope. : The only other filter I have ever seen that has that sharp a cutoff is one : designed by W2WLR that he sent me to evaluate. On the air, when tuned onto your signal, you've got some good, loud CLEAN audio, Don. Your signal isn't as wide as some of the others who are trying to put 10kHz of signal into a 6kHz bandpass. Please, everyone - excuse the comparison, but it does indeed sound a lot like the 'rack audio' SSB guys who are adding pre-emphasis, tone-tailored audio into their SSB rig, AFTER the Balanced modulator, to get a 'wider' spectrum to pass their audio through, taking up sometimes MORE than the conventional AM signal space, and while doing that, complain that "that ol' AM mode takes up too my bandwidth". Just some thinking, while sitting here at 02:18 (insomnia is a bitch!), does anyone think that the rack-audio guys are working on their radios as they are because they hear the AM'ers with decent audio, and are trying to mimick them (of course, without admitting that the AM mode has superior audio -smirk-) or is it because they see that AM is still allowed on the bands, they know damn well how wide an AM signal is, and figure (in their infinite wisdom) that "well, if the AM'ers can do it, than WE CAN TOO!" They just don't remember the AM/SSB wars from years ago. Back then, the fight was over not which MODE was better, but which mode was better for HAM RADIO. The SSB'ers argued that their signal took up 1/3 of the space, of a conventional AM signal. They're right. The AM'ers said that the resulting audio of a SSB signal, at 2.1kHz, was like trying to push an elephant through a keyhole. We were right about that, too. Now, the SSB'ers are using up to 10kHz of audio, AFTER the balanced modulator and when I hear a comment from someone who's fairly connected with Riley and the FCC that says that between Bob Heil and Riley Holingsworth, they're talking about limiting signals to 3kHz (that would be for SSB *only*, I hope!) and I share this informatoin, only to get told "you don't know what you're talking about". Am I the only one that see's the handwriting on the wall? I'm not trying to be a handwringer guys, but hey - something could be happening and party, it might be the blame of the wide-band, "hi-fi" AM audio that 'some' of the AM community runs. Yes run decent audio, yes make air-checks, yes homebrew it all! LIMIT the total bandwidth, so that ALL amatuers can get along on the same band(s). : One thing that is often overlooked is that the audio system capability needs : to extend one octave above and below the intended frequency response of the : program material, to minimise phase distortion effects. This is per : information in the United Transformer Corp. catalogue. For example, if you : intend to transmit high quality voice audio from 100 to 3500 cps, your audio : chain from the mic through the modulator stage needs to be as flat and : distortion free as possible from 50 to 7000 cps. Good "communications" : quality audio 300-3000 cps would require the transmitter to be capable of : 150-6000 cps (that is almost exactly the audio specification of the BC-610). Interesting about the '610... when I fed the output of the Solid State Audio Driver straight to the grids of the 100TH's in the W5MEU "E" model, the difference in audio was (according to others) tremendous, compared to the BC-614 Speech-amp, and that's one that John himself had modified for audio. : Any limiting of frequency response should be done using shaping circuitry : at low levels, not by limiting response at high levels by using transformers : with poor frequency response, or putting capacitors across interstage or : modulation transformers, or similar gimmickry often seen in ham rigs. Meaning, running the best mod iron possible, shunt the DC off the secondary of the modulation transformer, and run a big enough modulator with big enough tubes to 'loaf' along. Currently, I'm in the process of putting my pp 250TH rig back on the air, with a pair of 450TL's as modulators. I was concerend that the ol' buzzard RCA 5,500 ohms Z 1:1 mod transformer wouldn't hold that much audio, but then I remembered - Otis/K5SWK uses the same exact mod transformer for his pair of 833A's. It'll be fine. : My system seems to work well with my own voice. However, when N3DRB came by : for a visit, he operated my station, and everyone told him his voice didn't : sound nearly as good as it does on his rig at home. Each operator should : make sure the rig is taylored individually to his or her voice. Just as any good asymmetrical voice would well modulate (4) 813's in push-pull parallel :-) 73 = Best Regards, -=Jeff/W5OMR=-
