Title: Symmetra Peak for W3AM Post by: lu8dam on September 20, 2022, 08:16:27 AM My name is Diego LU8DAM, from Argentina.
Im testing a 8 pole phase rotating for W3AM, based on Symmetra Peak. Here are a video an audio: https://www.youtube.com/watch?v=f4qgX1R_qSI (https://www.youtube.com/watch?v=f4qgX1R_qSI) All stages are homemade construction. Im used a 100 watts class E transmitter with discrete modulator and PWM generator. Greetings from Pergamino, Argentina. Title: Re: Symmetra Peak for W3AM Post by: KD6VXI on September 20, 2022, 05:26:31 PM Hello from the USVI!
I used the w3am phase rotator. I put swit he's for 4 poles 6 poles or 8 poles. Followed that with a compressor configured more as a limiter. Then I fed that into a positive peak stretcher that allowed me to dial my pos peaks wherever I wanted them. That fed my PWM and a 100 watt carrier 125 pct mod rf deck. It was a great setup. Hopefully March 2023 I will have my gear down here and we can QSO... What frequencies do you operate AM? --Shane WP2ASS / ex KD6VXI Title: Re: Symmetra Peak for W3AM Post by: lu8dam on September 20, 2022, 05:44:08 PM Hello, Shane.
Excellent work you have done. I left it with 8 poles. When I mount it in a cabinet I will add switches and a bypass. I transmit on 7150 kHz to crystal. Maybe you can listen through some remote receiver on the web. Best regards. Diego LU8DAM Title: Re: Symmetra Peak for W3AM Post by: Opcom on September 20, 2022, 11:37:41 PM Welcome! It's always good to find another ham using a phase rotator. More people should use them on AM.
Title: Re: Symmetra Peak for W3AM Post by: K9MB on September 21, 2022, 01:06:18 AM This circuit is also called an All Pass Filter, I believe.
Jim Tonne has done a huge amount of work on speech processing which is available on his web site. http://tonnesoftware.com/ Jim Tonne article on the All Pass Filter. http://tonnesoftware.com/downloads/AllpassNetworksInASpeechChain.pdf Jim Tonne article on speech processing. http://tonnesoftware.com/downloads/SpeechProcessor-As-Submitted-To-QEX.pdf These circuits may be the ones that Shane uses in his excellent lineup that he just described. This technology has been used in broadcasting for decades and can be easily built by Amateurs who follow the designs that Jim published. Jim goves full credit for the All Pass Filter design to W3AM but he more fully analyzes it’s operation. Jim also has software on designing active low pass filters and I have worked out versions that do not over shoot when high rise time signals are applied. I strongly recommend Jim’s articles for a complete understanding of good audio processing to get the most audio energy in sidebands without distortion or splatter. Excellent analog designs. 73, Mike Title: Re: Symmetra Peak for W3AM Post by: lu8dam on September 21, 2022, 07:02:57 AM Thanks for sharing Jim's docs. I have visited your site many times.
Personally, I have never used the LED and LDR technique to work on radio compression, but I will probably put into practice some of your other projects published in the articles. Thank you very much! Diego LU8DAM Title: Re: Symmetra Peak for W3AM Post by: K9MB on September 21, 2022, 10:40:52 AM Hi Diego,
I was privileged to work by correspondence with Jim Tonne last Summer while he updated his PWM paper and ELSIE plus his Active Low Pass software. The active low pass software update includes the ability to introduce square waves- a worst case for clipped audio and develop active filters that can eliminate overshoot and in fact, provide “undershoot”. Since a complete speech processing system includes a clipper, this undershoot anti-aliasing filter will prevent high frequency noise from getting to the modulator. I have been planning to publish on AM Fone all my results, but time constraints has prevented it so far. Therefore, I will share a little of my results here on the speech processing portion. The images below are of the undershooting active filter I worked out using Jim’s software and his guidance. Below is an email copied from 08/21/2021: In my portion, you will find the resistor values for various cutoff t I worked out that are appropriate for SSB to various AM bandwidths. You will notice the very steep skirts on the 7 pole filter designs above cutoff frequency while having no overshoot at all and in fact undershooting on transients. These responses are very important when placed right after the compressor circuit that I will post next. 73, Mike On 8/12/2021 7:13 PM, Michael Burke wrote: Hi Jim, I got into the active LP filter program this afternoon for an hour and it was amazing how easy it was to change the bandwidth with the filter response shape and transient response staying the same. 10k gave the 4.0 khz response as you know 13k gave a nice corner at 3.0khz that was down 50dB at 6.1kHz 11k gave 3.5khz and 50dB down at 7.4 khz 8.2k gave 4.5khz and 50dB down at 9.8khz 6.8 k gave 5.0khz and 36dB down at 10khz My goal is to keep my sidebands down significantly above +- 5khz of my operating frequency. The 4.0 filter is 13dB down at 5khz The 4.5 filter is only 4 dB down at 5khz For straight communications efficiency and minimal spectral spread, the 4.0 filter looks great to me, even if is not high fidelity sound. For SSB audio, the 3.0 filter is pretty good. It is 16 dB down at 4.0khz and 35dB down at 5khz I could squeeze the SSB a bit more, but figure that the crystal or mechanical filter will serve for that and I want to be able to boost the 2.0-3.0 khz speech frequencies for better readability. Thank you for the latest brick in this new house (or shack 😉) of knowledge in the design of filters for communications speech amps and modulators. I feel like I have scratched just the surface so far, but I am having fun. 73, MB Sent from my iPad On Aug 12, 2021, at 10:58 AM, JimTonne <tonne@comcast.net> wrote: Mike: Much, much easier way to move the bandwidth around would be for you to first develop a nice response. You did that. Then simply scale the resistor values, making them all the same. In your case the "nice" response was using 10k resistors. To move it up simply scale all of the resistor values, perhaps to 8200 or 7500 ohms. Would take about a minute and it would work. Would be better to use the 1% resistor values, whose values come in quite small increments, Been working fuiously on the active filter program, adding the "marker" display. - JimT Title: Re: Symmetra Peak for W3AM Post by: K9MB on September 21, 2022, 11:05:34 AM Hi Diego,
In this post, I will include an entire speech filterSystem except the clipper and the final undershooting final low pass anti aliasing filter. In this correspondence, Jim makes recommendations about the lineup necessary for an optimally designed speech system using analog design. Note in Jim’s list of the order of circuits, he lists his 4khz undershooting filter. Jim developed a 5 pole filter that has the same response as my pole filter, bit I prefer the 7 pole because it has much steeper skirts. You can run both filters on his software for comparison. MB Again, I will copy directly from an email to Jim Tonne. MB Email :August 08/2021 Is This a complete Speech Processing system? Hi Jim, I answer in context: Sent from my iPhone On Aug 8, 2021, at 12:55 PM, JimTonne <tonne@comcast.net> wrote: Mike: Put the allpass stuff prior to the first AGC. This will reduce the AGC activity and so enhance loudness. That AGC should have an attack time of the order of 2 (absolute minimum) to 8 or so (maximum). Shorter attack times will reduce the loudness. Longer will make the sound "pop" and is notgood-sounding. Ok, but I do not know the term “order of 2” so what is that? But notice I said make the attack time finite. That will allow transients to go on through that block. Those transients will cause overmodulation. So after the AGC you should install a clipper, even if it only works on transients. Ok- got it. After the clipper you should install the lowpass filter. That filter must not have ringing in its transient response. Your filter, the 7-poler, does a very very nice job. I ran the complete system (trapezoid generator followed by your lowpass followed by a PWM generator followed by the output lowpass). As I varied the frequency of the trapezoid (i.e., the clipped sinusoid) I noted the level of modulation. Your filter has full modulation without any peaks going into overmodulation, right up to 4 kHz. Then the modulation level rolls off. See the plot below. Good news! We have a saying- even a blind sow finds an acorn (or truffle ) from time to time… My hat size is not changing just yet. Questions abound but grateful to get anything right because I define that as progress…😉 <000743.jpg> My design uses 2 instead of 3 opamps and also has full modulation on trazezoidal waves up to 3500 Hz instead of your 4000 Hz. For SSB, your filter is better- I was thinking “good” quality AM here. Might build a separate speech processor for my SSB rigs later. I have a little circuit I use to make lopsided speech waveforms. I fired it up and ran its lopsided waveform into the W3AM allpass. Here is the schematic of that entire system: <000751.jpg> On the left is the speech simulator. On the right is the W3AM allpass. Here is the waveform out of the speech simulator, along with entire schematic: <000749.jpg> The red line down the middle is zero volts so you can see how lopsided that waveform is. Now the output of the allpass: Yes- I am familiar with this- having read your All Pass filter paper and also your new approach to speech processing paper. If you still have my previous email showin blocks for the speech system, please confirm that I got it right. 1. Your mic amp with treble boost snd lpf 2. W3AM All Filter 3. Your AGC-Compressor circuit 4. Your clipper 5 My version of your negative transient 4khxz lpf driver 6. My PWM modulator <000750.jpg> See how that allpass made the signal have equal peak values positive and negative. Put the treble boost prior to the AGC. The allpass can be before or after the treble boost but prior to the AGC. - JimT I feel like I am beginning to get a little clue about this and understand better how little I know on the subject. I am very grateful for your wise counsel and patience. 73, MB Title: Re: Symmetra Peak for W3AM Post by: K9MB on September 21, 2022, 11:27:00 AM Hi Diego,
In this post, I add Jim’s Clipper and the undershooting post clipper active low pass filter to complete the system. In the first image, I have copied Jim’s original clipper with the 5 pole post filter set for 4khz and in the second image, my 5khz filter is shown. In the final image I copy Jim’s original clipper circuit with 5 pole filter for reference. If you use the first image circuit then add the second 7 pole filter, you have completed the system. My 5 khz filter can be changed incutoff using the data from one of my previous posts today wherein I give a number of bandwidths which are all worked out on Active Low Pass Filter software and the bandwidth changes need only changing the 1% resistor values. The fact that the resistors are all the same value allows one to buy 100 of the high precision resistors and hand select them. Capacitor values must be carefully measured and accurately determined for best performance, but you only need to do this one time for many possible bandwidths in Jim Tonne’s system. 73, Mike K9MB Edit: I have called the filter in the image a 5khz filter using 10k resistors, but in fact the curve put the corner at about 4.5khz. I recommend that you run the filter in Jim’s Active Low Pass Software by keying in all the values and then trying other resistors. I actually like 6.8k better to give a full 5khz bandwidth, though the 10k filters will keep the width of the AM signal under 10kHz for crowded bamd conditions where people are sensitive to QRM from adjoining Transmitters. On 15 or 10 meters, wider audio is no problem and maybe ok on 40 meters where conditions vary widely. MB Title: Re: Symmetra Peak for W3AM Post by: lu8dam on September 22, 2022, 07:41:46 PM Thank you for all, Mike.
What type of low pass filter are you using, calculated with Jim's software? I use a Bessel-type filter with a cutoff frequency of 5kHz, as the last stage before entering the modulator. Title: Re: Symmetra Peak for W3AM Post by: lu8dam on September 22, 2022, 07:44:45 PM This is my discrete components PWM modulator.
https://youtube.com/shorts/-zzJTxQu4K4 (https://youtube.com/shorts/-zzJTxQu4K4) Title: Re: Symmetra Peak for W3AM Post by: K9MB on September 23, 2022, 01:05:55 AM Thank you for all, Mike. What type of low pass filter are you using, calculated with Jim's software? I use a Bessel-type filter with a cutoff frequency of 5kHz, as the last stage before entering the modulator. You really need to read Jim’s paper on PWM, but the thing you must have to properly design the pwm filter is to get the newest version of ELSIE on Tonne Software. Jim added a feature last year that he calls single termination routine. It uses Butterworth filters and calculates values so that the input impedance can be different than the output load impedance. You can also calculate the values and plot them at the RF-Tools site here: https://rf-tools.com/lc-filter/ The thing about ELSIE is that you can run transient response curves just like in the Active Filter program. This method of calculation gives very little overshoot response. Jim says that there is no reason to make the cutoff at a very low frequency if you have designed the speech system properly because the high frequency products will not be there. However, if the filter rings, it will not matter that you set it at 5khz. The images below show 28khz cutoff and 15khz cutoff. I chose to use the 18kHz cutoff filter. The most important spec is that the filter is more than 80dB down at 160khz, the clock speed for my modulator. Even the 28khz filter is over 60dB down, so either will work and not ring on transients. The crucial thing is to design the entire system, not do a slipshod job in the speech amp and then try to clean it up in the pwm filter. It would not work anyway because the narrow filters will likely ring and generate even more distortion and distortion in the sidebands where we do not want them. If we do the Speech chain right and make the pwm filter be a fast settling response that does ring. Get Elsie to start with and also run some numbers on the rf-tools site. 73, Mike Title: Re: Symmetra Peak for W3AM Post by: lu8dam on September 23, 2022, 06:56:49 AM Thanks for everything. The PWM filter is finely calculated for 75kHz, the same frequency as the PWM generator.
Yes, I have seen Jim's doc, I am aware of ELSIE and rf-tools as they are part of the tools I use to check filters. I have even programmed code to calculate active filters based on university documents. I leave you cordial greetings. Title: Re: Symmetra Peak for W3AM Post by: K9MB on September 23, 2022, 10:37:08 AM Thanks for everything. The PWM filter is finely calculated for 75kHz, the same frequency as the PWM generator. Yes, I have seen Jim's doc, I am aware of ELSIE and rf-tools as they are part of the tools I use to check filters. I have even programmed code to calculate active filters based on university documents. I leave you cordial greetings. Ok, so your clock speed is 75khz? Interesting… I am curious why you chose such a low frequency? The values of inductance must be very high and it does necessitate a much lower cutoff for the pwm filter. I like the idea of higher frequencies because the audio can be made flat out to 10khz for low distortion and good fidelity and still reject the clock while using only 4 poles? You must have a reason and I am interested. Thank you in advance. 73, MB Title: Re: Symmetra Peak for W3AM Post by: lu8dam on September 23, 2022, 12:47:26 PM Here are the reason:
https://steampoweredradio.com/manuals/gatesm.html (https://steampoweredradio.com/manuals/gatesm.html) And i based on this documents: https://steampoweredradio.com/pdf/gates%20harris/pdm%20manual.pdf (https://steampoweredradio.com/pdf/gates%20harris/pdm%20manual.pdf) Most of the stations in Argentina use sampling frequencies between 75 and 100 kHz. Surely, by tradition, it comes from all this time. Title: Re: Symmetra Peak for W3AM Post by: K9MB on September 25, 2022, 11:13:46 AM Here are the reason: https://steampoweredradio.com/manuals/gatesm.html (https://steampoweredradio.com/manuals/gatesm.html) And i based on this documents: https://steampoweredradio.com/pdf/gates%20harris/pdm%20manual.pdf (https://steampoweredradio.com/pdf/gates%20harris/pdm%20manual.pdf) Most of the stations in Argentina use sampling frequencies between 75 and 100 kHz. Surely, by tradition, it comes from all this time. I see. Well, those are older designs and for commercial stations where space is not a large consideration. There certainly is a sweet spot for frequency used that does not cause RF interference on broadcast, amateur and service bands. There was a discussion recently about linear vs switching power supplies. I use surplus 60hz transformers in linear supplies for driving my pwm and Class E transmitters, but a 3 kw linear supply will weigh 80lbs and a 3kw switcher will weigh less than 5 lbs and be much smaller. Higher clock frequencies fir pwm have the same advantage, with very high power 160khz converters being very compact because the inductors are inherently smaller and lighter. Also, the LPF can roll off higher in frequency, so high fidelity audio is easily achieved with simple 4 pole filters. Of course, if ome is just restoring vintage equipment, it makes no sense to change the 75khz clock speed. Harris obviously worked out all the details for great performance. Title: Re: Symmetra Peak for W3AM Post by: K9MB on September 25, 2022, 11:17:17 AM For those who are working on speech amplifier systems, it is a great idea to review the circuits published by Jim Tonne and consider the low pass undershoot filter designs as a part of an overall system that can drive any piece of equipment. AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
73, MB |