Re: [AMRadio] Frequency Response
By 'tilt' I mean the observable tilting of the top and bottom of a square wave (or resulting modulated envelope) caused by low frequency roll off and non constant group delay through the transmitter in question. That's why the transmitter frequency response, including transformers, should be essentially flat at least one octave above and one octave below the actual range of frequencies you plan to transmit. For example, if you intended to limit your audio response to 150-4000~, the audio chain should test flat from 75 to 8000~. Don k4kyv
Re: [AMRadio] Frequency Response
Exactly. However, if you intend to employ significant audio clipping as a means of peak control and loudness enhancement, the low freq response should ideally extend well below that to better accommodate the flat topped waveforms without substantial elevation of their peak level caused by the tilt. But again, this is only ham radio and your rule of thumb is a great one. If everyone's rig passed that test we'd be talking about something else. g Donald Chester wrote: By 'tilt' I mean the observable tilting of the top and bottom of a square wave (or resulting modulated envelope) caused by low frequency roll off and non constant group delay through the transmitter in question. That's why the transmitter frequency response, including transformers, should be essentially flat at least one octave above and one octave below the actual range of frequencies you plan to transmit. For example, if you intended to limit your audio response to 150-4000~, the audio chain should test flat from 75 to 8000~. Don k4kyv
RE: [AMRadio] Frequency Response
I have always admired what Don has done with his modulation system. That 3400 hertz brick wall filter is a rarity unless one uses high technology active circuitry such as active op-amps, switched capacitor filters, dsp, etc. I believe Don has a passive module that does a great job. I wonder where he got it? It is a passive module. I picked it up in Washington, DC back in 1972 when I had stopped over there to visit a friend. We went to a little hole-in-the-wall military surplus outlet. I paid $5 for it. It is potted, about 1 1/2 X 1 3/4 X 2 . It has an octal plug on the bottom. The impedance in/out is 10,000 ohms. I wasn't sure about it when I purchased it, but when I tried it out with a signal generator after returning home, I wished I had bought every one they had. It is flat up to about 3300~, and at 3500~ the signal is so far down as to be undetectable on the scope. You can simulate the envelope pattern of a cw signal by swishing the test oscillator about 100~! The only thing like that I have ever seen at a hamfest was physically and electrically identical but had a useless cutoff frequency of 2000~. Don K4KYV
Re: [AMRadio] Frequency Response
What about bandwidth? +/- 5kc would be a 10 kc band width. I thought we were supposed to limit our band width to 6kc. Please correct me if I am wrong. That is a popular urban myth. There is NOTHING in the US regulations that specifically limits bandwidth. The regulations specify good engineering and amateur practice, and deliberately leaves the specific bandwidth vague. In Canada, there is a rule on the books limiting bandwidth to 6 kHz, but I have never heard of them enforcing it against AM signals that may exceed that figure. Besides, accurate measurement of bandwidth would require a visit to the station with test instruments. Over-the-air measurement leaves too many possibilities for error due to propagation, QRM, QSB, etc. The US regulations could be interpreted to mean a reasonable bandwidth for the mode being used, considering band occupancy . If you had a cw signal with so much noise, hum or FM on the carrier that it was 3 kHz wide, the FCC probably could interpret that as a violation of good engineering practice. If the band is empty, as for example, 10 m. most of the time nowadays, or 160m in the middle of the day, you could run hi-fi AM with audio all the way up to 15 kHz and that would probably be ok as long as you made sure you were not causing any harmful interference to anyone. On the othre hand if you were limiting the audio response to 3000~ and generating the same wide bandwidth due to splatter (overmodulation or distortion), that would be considered not to be good engineering practice. If you operated the full hi-fi audio at high power on 75m at night when the band was crowded, that could be interpreted as violation of good amateur practice. The bottom line seems to be, use common sense and adjust bandwidth according to conditions, and make sure your transmitter's spurious distortion products fall within the FCC's specifications, which are listed in the rules. There was a flare-up regarding bandwidth a year or so ago, with hi-fi SSB. This resulted in petitions to specifically limit bandwidth. The FCC apparently turned them down. Now the ARRL is proposing to change the definitions of subbands to be defined by bandwidth instead of emission mode, to promote digital experimentation. The proposed bandwidth limit for AM is 9 khz. The League has received so much mail questioning the wisdom of such a change, that the League seems to be rethinking the idea. They still have an open invitiation to the amateur community to send them comments and opinions on this subject. Don K4KYV
Re: [AMRadio] Frequency Response
Thank you Don... this needs to be repeated every so often because these 'Urban Myths' seem to never die in spite of the facts... The way I've got it figured is the FCC doesn't want to do much of anything to limit bandwidth... not so much for the sake of AM but to leave the doors open for other yet unknown modes... I mean, and this is a silly comparison but..., Just how 'wide' is a spread spectrum signal??? and for giggles run an check on some of the DAT modes... intermodulation distortion is a bug in even some of the best equipment and that brings us back to 'good enginneering practices'... Bad example I admit but when you think about it... tnx agn Don... 73 vince ka1iic On Sunday 26 September 2004 06:02 pm, Donald Chester wrote: That is a popular urban myth. There is NOTHING in the US regulations that specifically limits bandwidth. The regulations specify good engineering
Re: [AMRadio] Frequency Response
IMHO, the best approach is to terminate the mod xfmr (or any audio xfmr) for best transient (square ware) response of the modulated envelope, not just an arbitrary upper frequency. Adjust for best stability, least overshoot and ringing on leading edges, and the least tilt. Check it at mid freq's, say 1khz, as well as low (50hz) and high (5khz). It's easier to see the tilt on the low end, and the overshoot on the midrange. All of these disturbances rob modulation power from you, as you have to reduce average modulation level to keep the peaks from overmodulating. Cleaning it up will help maximize modulation control, sometimes as much as several db. This in turn increases potential average loudness which is the name of the game. The best place for audio bandwidth control/limitation is in the low level audio chain. As to where to roll it off, I'd suggest 5khz, but I like to make it variable depending on conditions. With only 3KHz you might as well be on SSB. g Edward B Richards wrote: Tom; I believe the standard for communications is 300 to 3000 CPS +/- 3 dB. On Fri, 24 Sep 2004 14:54:49 -0800 Tom Elmore [EMAIL PROTECTED] writes: I am in the process of restoring a Stancor 60-P transmitter to service. I have replaced all the capacitors in the power supply and in the audio amplifier sections. I would like to terminate the modulation transformer with the correct resistance and check for frequency response on my scope. What is a good ballpark response that I should use for operating AM in the amateur bands. Thank You Tom Elmore KA1NVZ Anchorage Alaska
Re: [AMRadio] Frequency Response
The phone company is responsible for that illogical 3 kHz upper limit. The consonant sounds are missing unless you go above 4.4 kHz( 5 kHz is a good compromise). Joe W4AAB - Original Message - From: Gary Blau [EMAIL PROTECTED] To: Discussion of AM Radio amradio@mailman.qth.net Sent: Friday, September 24, 2004 11:20 PM Subject: Re: [AMRadio] Frequency Response IMHO, the best approach is to terminate the mod xfmr (or any audio xfmr) for best transient (square ware) response of the modulated envelope, not just an arbitrary upper frequency. Adjust for best stability, least overshoot and ringing on leading edges, and the least tilt. Check it at mid freq's, say 1khz, as well as low (50hz) and high (5khz). It's easier to see the tilt on the low end, and the overshoot on the midrange. All of these disturbances rob modulation power from you, as you have to reduce average modulation level to keep the peaks from overmodulating. Cleaning it up will help maximize modulation control, sometimes as much as several db. This in turn increases potential average loudness which is the name of the game. The best place for audio bandwidth control/limitation is in the low level audio chain. As to where to roll it off, I'd suggest 5khz, but I like to make it variable depending on conditions. With only 3KHz you might as well be on SSB. g Edward B Richards wrote: Tom; I believe the standard for communications is 300 to 3000 CPS +/- 3 dB. On Fri, 24 Sep 2004 14:54:49 -0800 Tom Elmore [EMAIL PROTECTED] writes: I am in the process of restoring a Stancor 60-P transmitter to service. I have replaced all the capacitors in the power supply and in the audio amplifier sections. I would like to terminate the modulation transformer with the correct resistance and check for frequency response on my scope. What is a good ballpark response that I should use for operating AM in the amateur bands. Thank You Tom Elmore KA1NVZ Anchorage Alaska __ AMRadio mailing list Home: http://mailman.qth.net/mailman/listinfo/amradio Help: http://mailman.qth.net/mmfaq.html Post: mailto:AMRadio@mailman.qth.net
Re: [AMRadio] Frequency Response
Gary I have read several of your writings that refers to tilt. What is it, or where can I go to learn more about it? Byron W3WKR
RE: [AMRadio] Frequency Response
I have always admired what Don has done with his modulation system. That 3400 hertz brick wall filter is a rarity unless one uses high technology active circuitry such as active op-amps, switched capacitor filters, dsp, etc. I believe Don has a passive module that does a great job. I wonder where he got it? Also consider your individual voice characteristics, microphone, and speech polarity when building a modulation system. Jim WD5JKO Donald Chester [EMAIL PROTECTED] wrote: What is a good ballpark response that I should use for operating AM in the amateur bands. I would shoot for minimum of 100~ to 5000~. According to the UTC catalogue (United Transformer Co.), to keep phase shift distortion to a minimum, the response capability of the transmitter should run one octave above and one octave below the actual response used, so a transmitter with 50~ to 10,000~ capability would sound better even with restricted response of the audio actually used. Of course, that would require broadcast quality components. I would suggest running response tests on the transformers in the rig, and then make sure the coupling components in the rest of the audio chain allow at least the response of the most restricted transformer. If you feel you need to further restrict the response in any way, it should be done right at the microphone preamp. If you limit your overall response to 300-3000~ you might just as well run SSB. I have been able to get satisfactory results with a very sharp, brickwall lowpass audio filter that cuts off at 3400~, in combination with a rising response boost in the mic preamp beginning at 800~ and peaking at about 9 dB from 2000~ to the cutoff frequency of the filter, while the low frequency response remains flat down to about 40~. Don k4kyv __ AMRadio mailing list Home: http://mailman.qth.net/mailman/listinfo/amradio Help: http://mailman.qth.net/mmfaq.html Post: mailto:AMRadio@mailman.qth.net
Re: [AMRadio] Frequency Response
What about bandwidth? +/- 5kc would be a 10 kc band width. I thought we were supposed to limit our band width to 6kc. Please correct me if I am wrong. 73, Ed Richards K6UUZ On Fri, 24 Sep 2004 23:37:25 -0500 Crawfish [EMAIL PROTECTED] writes: The phone company is responsible for that illogical 3 kHz upper limit. The consonant sounds are missing unless you go above 4.4 kHz( 5 kHz is a good compromise). Joe W4AAB
Re: [AMRadio] Frequency Response
If you listen around on AM much, there are very few 6 kHz signals, just like SSB. I hear lots of the SSB boys wider than 6 KHz.I don't believe that is according to the rules, but then, unless someone is being a real pain in the neck, the FCC isn't going to start measuring bandwidth.They have more important things to do, like not rocking the boat and pushing the agendas of the Electronic Industries Association. Just my 2 cents, adjusted for lack of inflation. Joe W4AAB - Original Message - From: Edward B Richards [EMAIL PROTECTED] To: amradio@mailman.qth.net Sent: Saturday, September 25, 2004 10:28 AM Subject: Re: [AMRadio] Frequency Response What about bandwidth? +/- 5kc would be a 10 kc band width. I thought we were supposed to limit our band width to 6kc. Please correct me if I am wrong. 73, Ed Richards K6UUZ On Fri, 24 Sep 2004 23:37:25 -0500 Crawfish [EMAIL PROTECTED] writes: The phone company is responsible for that illogical 3 kHz upper limit. The consonant sounds are missing unless you go above 4.4 kHz( 5 kHz is a good compromise). Joe W4AAB __ AMRadio mailing list Home: http://mailman.qth.net/mailman/listinfo/amradio Help: http://mailman.qth.net/mmfaq.html Post: mailto:AMRadio@mailman.qth.net
Re: [AMRadio] Frequency Response
Ed, You make a good point. I do however need to bring up that bandwidth needs to be specified with a certain roll off in DB. For example a mechanical filter in a receiver is often specified at -6, and -60 db and with a certain shape factor. A clean AM ham rig may be 30 Khz wide at the - 60 db points +/- 15 Khz away from the carrier. That same rig may be 6 Khz wide at the -6 db points, and 8 Khz wide at the -20 db points. So what bandwidth is this rig? I contend that a clean well modulated AM rig that has low distortion in the audio, and RF modulated stage (and any linear amplifiers) will generally not cause a problem in the ham bands when running the audio unrestricted. That is because the voice spectrum of the average ham's voice is such that the energy content is way down by 3-4 Khz anyway. A little roll off kicking in at 3 khz of about 6 db / octive or more is a good thing so long as the lowpass rolloff is preceeded with a high pass as Don has done. This needs to be taylored to the rig, microphone, and your voice. Look at the scenerio where the AM rig has a steep low pass filter after a speech clipper, but the high level modulated stage has high distortion, and the RF plate modulated stage is not linear. This rig will be wider in bandwidth than the clean AM rig with wide open audio. So you might say that my rigs are 30 Khz wide. Maybe so, but using the same criteria of not specifying a roll off in DB away from the carrier level, so is most any ham rig on the market today. Jim Edward B Richards [EMAIL PROTECTED] wrote: What about bandwidth? +/- 5kc would be a 10 kc band width. I thought we were supposed to limit our band width to 6kc. Please correct me if I am wrong. 73, Ed Richards K6UUZ On Fri, 24 Sep 2004 23:37:25 -0500 Crawfish writes: The phone company is responsible for that illogical 3 kHz upper limit. The consonant sounds are missing unless you go above 4.4 kHz( 5 kHz is a good compromise). Joe W4AAB __ AMRadio mailing list Home: http://mailman.qth.net/mailman/listinfo/amradio Help: http://mailman.qth.net/mmfaq.html Post: mailto:AMRadio@mailman.qth.net
Re: [AMRadio] Frequency Response
Hi Jim; For the 55 years I have been in electronics, bandwidth has always been defined as the half voltage points (-6dB). We all know that band passes don't have vertical sides. Therefore it will be narrower above the -6 dB points and wider below. Thanks for your input. 73, Ed Richards K6UUZ On Sat, 25 Sep 2004 10:29:14 -0700 (PDT) Jim Candela [EMAIL PROTECTED] writes: Ed, You make a good point. I do however need to bring up that bandwidth needs to be specified with a certain roll off in DB. For example a mechanical filter in a receiver is often specified at -6, and -60 db and with a certain shape factor. A clean AM ham rig may be 30 Khz wide at the - 60 db points +/- 15 Khz away from the carrier. That same rig may be 6 Khz wide at the -6 db points, and 8 Khz wide at the -20 db points. So what bandwidth is this rig?
Re: [AMRadio] Frequency Response
How you interpret the rules or who you believe on the air is always confusing, so here's the FCC's present rules on emission and bandwidth: http://frwebgate.access.gpo.gov/cgi-bin/get-cfr.cgi?TITLE=47PART=2SECTI ON=202YEAR=2001TYPE=TEXT Here's FCC's General Rules and Regulations: http://www.access.gpo.gov/nara/cfr/waisidx_01/47cfr2_01.html And here's the current Part 97: http://www.arrl.org/FandES/field/regulations/news/part97/part97.pdf Pete, wa2cwa On Sat, 25 Sep 2004 08:28:51 -0700 Edward B Richards [EMAIL PROTECTED] writes: What about bandwidth? +/- 5kc would be a 10 kc band width. I thought we were supposed to limit our band width to 6kc. Please correct me if I am wrong. 73, Ed Richards K6UUZ On Fri, 24 Sep 2004 23:37:25 -0500 Crawfish [EMAIL PROTECTED] writes: The phone company is responsible for that illogical 3 kHz upper limit. The consonant sounds are missing unless you go above 4.4 kHz( 5 kHz is a good compromise). Joe W4AAB Get your name as your email address. Includes spam protection, 1GB storage, no ads and more Only $1.99/ month - visit http://www.mysite.com/name today!
