#101
HTML-код
- Опубликовано: 6 сен 2024
- *More details here, please read*
This video shows how to measure FM Frequency Deviation using the Carrier or Bessel Null method. Ideally, no special equipment is needed besides a CW receiver capable of tuning to the FM transmit frequency, an audio generator, a method of properly adjusting the the audio into the transmitter, and a safe way of coupling the TX output into the receiver. This method takes advantage of the fact that the carrier of an FM signal modulated with a single audio tone will pass through zero (a null) at specific combinations of modulating frequency and deviation.
The setup I show is a little more complex because I don't have a CW receiver that operates on 2 meters, so I had to use a mixer to downconvert the 2m signal to an HF frequency. In this case, I used 10.7MHz because I had a ceramic resonator that could be used as an IF filter following the mixer. I also used a splitter in my setup so that I could also send the transmitted signal to a FM deviation meter to double check the deviation measurement.
Here is a link to the video on the mixer I used for the downconversion:
• #83: Basics of RF Mixe...
Here is the link to the video on RF Samplers, like the one I used to couple the transmitter output into the mixer and deviation meter:
• #61: Basics of RF Samp...
You will want to carefully adjust the CW receiver's passband to ensure that only the carrier is being measured, and not any of the sidebands. This is usually not a problem. Alternatively, you can use any one of a number of free programs that show you the audio spectrum, waterfall plots, etc., such as those used for PSK31, etc., or even free audio spectrum analysis software such as Spectrum Lab, to monitor the level of the carrier tone - even in the presence of the sideband tones. If your rig has a spectrum monitor function, you may even be able to use that. Just pick the method that works best for your CW receiver to monitor the level of the carrier tone as you vary the audio tone.
If your audio generator doesn't have an accurate frequency readout, remember that most decent DMMs can measure the frequency of audio tones, so use that.
You do have to take care in setting the audio level from the audio generator into the mic input of the rig. Note that some rigs have a DC bias on the audio line, so it's a good idea to couple the signal through a large value capacitor - preferably a non-polarized electrolytic. Typical mic levels are 10s of millivolts, so if your audio generator output can't be dialed down that low, you'll have to attenuate it with a voltage divider, pot, or attenuator. I used a variable attenuator in my setup. The rigs will generally have some audio compression circuitry, so if input signal is driven too hard, it will internally limit it to prevent over-deviation or audio clipping. Of course, these circuits aren't perfect, so be cautious. It is helpful to monitor/listen to the rig's output on an FM receiver as you adjust the audio level. Increase the level until the perceived volume is no longer increases. This means you've generally hit the max deviation of the transmitter.
Please let me know if I left out any critical details that you might need. And of course, take great care in your setup anytime you are connecting a transmitter directly to a receiver!!! This is why I have the transmit power turned all the way down, and I am using a RF tap that has at least 40-50dB of attenuation, and I have the attenuator enabled on the receiver. If you are unsure about it, then simply transmit into a dummy load and use an antenna on the receiver - there should be enough leakage for you to pick up the signal.
Hi Alan, another great video. After reading a response from one of your readers that said they did not have a function generator I came up with an idea to just try a free audio tone generator app on my smart phone (versus having to wire in my signal generator, etc.). My total setup was just a $20 RTL-SDR dongle receiver connected to my laptop with the RTL-SDR receiver software set for CW (but mode does not really matter since you are just looking at the carrier on the laptop display), my smart phone as the adjustable tone generator and my radio under test (no wiring needed between the cell phone and my radio under test, and no wiring between the radio under test and the RTL-SDR receiver). I just set the transmitter output power to 1 watt on the transmitter under test and did not use an antenna on the RTL-SDR receiver to make sure I would not damage it. This simple system appears to work just fine as long as the audio output form the cell phone is set to high (loud). Thanks for another very interesting measurement technique.
Awesome!! I knew the method from the books but never saw it in action. You are the man, Alan!
The frequency deviation for a given tone amplitude will vary with frequency in accordance with the nominal 6 dB per octave pre-emphasis on the microphone input to FM transmitters. For that reason, radios with separate data ports are preferred when using packet and other AFSK modes.
Hi Alan. Many thanks for sharing this video. Interestingly I discovered that this method works even better with a spectrum analyzer than the one described in video #159. Apparently, my spectrum analyzer, DSA815-TG, works a bit different than yours - I couldn't make it draw anything similar to the traces you got. However, the method shown in this video worked just fine. It's not necessary to use a mic input of the transceiver. Placing the microphone at a fixed distance from the speaker connected to the signal generator works just fine. If the transceiver doesn't have an antenna connector (e.g. LPD/PMR walkie-talkie), a small receiving antenna can be connected to the spectrum analyzer.
The problem with placing the mic at a fixed distance is that you don't have direct control over the deviation, since it will be a function of the distance, the speaker, audio reflections, etc.
Hi there, its all clicked into place now, I can see exactly what you doing now...and how your doing it...we might do this to calibrate a Heathikit deviation meter we have just repaired, I connected the power backwards and blew all the IC op amps..its working now, we upgraded them from the round metal type to the new type, they just plug straight in...cheers Fred in England.
Sadly I don't have a HF rig (it's on my shopping list, right after I get my amateur radio licence!), but I do have a spectrum analyser. I can report success using that as a receiver -- the procedure is similar. I did my experiments with a Marconi 2022E signal generator. An output level of -80dBm proved difficult to tune, though -60dBm proved to be plenty.
Set the TX to produce an unmodulated carrier. Set the SA centre frequency to this, with a reasonable span - say 10kHz. Get the signal as close to centre as you can.
Set the SA to its narrowest reasonable ResBW filter (mine is 30Hz), and span to minimum. If your SA has Windowed Sweep mode, set that too, to speed up the sweep as much as possible. You're aiming to just refresh the signal, not the noise at either side. Put a marker on the signal, then tune the audio oscillator until carrier power is at a minimum. Generally speaking, you'll go from a 'bump' to a flat line (no carrier) when you're absolutely on the Bessel null.
Alternatively, set Zero Span mode and drop the VBW filter to about 1Hz. Adjust until you get the line as low as possible, but I found this harder than using a narrow, windowed sweep.
Once you've got your modulating frequency (mine was 1.045kHz), multiply by 2.045 to get the FM deviation.
It's nowhere near as convenient as a modulation meter, but if you've got the equipment on hand, you can't beat the price!
If only this trick worked for off-air signals!
very helpfull video. I will try this at the next weekend to adjust the deviation of my old President Jackson 10/11m cb-radio.
thanks for showing.
Best 73,
DO5DKL
More evidence that amateur radio operators are awesome. Thanks!
Another great video Alan! I would be interested to know how to measure FM Deviation for off the air signals using a spectrum analyzer.
A quick and dirty method I have used in the past is to use the radios own tone burst (1750 hz in the UK) and adjust the deviation for the bessel null. It gives you a handy starting point...
Very Nice! exactly what I was looking for, worked perfect, subscribed.
Wonderful explanation my friend. Thanks so much for putting this one together. Time for me to build the mixer to keep around the shop.
Many thanks.
Amos
Thanks for sharing and very informative.. so am I understanding if the deviation is out of the 5 kHz range you would need to adjust the audio level in the settings of the radio it’s self.. ?
Depends on the radio. There is likely an adjustment for the deviation or the audio drive to the modulator. The service manual for the particular radio would indicate the proper adjustment point to set the deviation.
Thank you. Great video as always!
great explanation and demonstration!
What in the FM transmitter is setting the amount of FM deviation? its the amplitude voltage or power wattage of the FM amplifier that is determining the maximum FM deviation?
My very old Yaesu mobile radio's service manual deviation setting requires a 1kHz tone at a given amplitude to yield the nominal 5kHz deviation. The amplitude will be different on different radios. I did some Bessel null testing with my IC7100 data port input and found it needed a very low amplitude compared to other radios for a given deviation.
Very informative video, thanks for posting.
I do not quite understand how you connect the TX with RX and how much power arrives at the RX antenna jack. You said that you use a T-Adapter, terminated with a Dummy load and that you take you bring only a part of the TX power to the RX input. But how do you manage that reduction? Is it just a simple cable from the T-Adapter to the RX Antenne jack or is there an attenuator (like -30 dBm for 4 Watts output) in between somewhere.
I used an "RF Tap" - also called an "RF Sampler". The "tap" coming out of the side of the sampler is adjustable (like a trombone) to set the attenuation value.
@@w2aew Many thanks for this clarification
How about using a sdr? I will need to check but when I look closely at the waterfall of a FM station I can see the signal narrow during a quiet period and fatten up as more sound is created. I have wondered if it is the sound frequency or is it the volume level are being displayed by the software? Since we are looking at a frequency display I think it is the FM modulation I am seeing in the width change and the volume is in the vertical scale of the display?
Brilliant idea.
Isn't it just easier and simpler to use an FM deviation meter in the first place instead of all this faffing about. Good informative video all the same.
Yes, of course, assuming you *have* an FM Deviation Meter... The point of this video is to show how to make a this measurement if you don't have access to special equipment like a deviation meter.
Where did you get that nice graph of the Bessel Functions?
I believe I plotted it using Excel.
I want to learn all of this where can I start?
Great video, indeed! Would it be possible for you to indicate if modulation index is solely a function of modulating frequency and peak frequency deviation but not the amplitude of the modulating tone (in practical measurements). Thx.
Well, the peak deviation *is* a function of the modulating tone's amplitude - therefore it *does* change the modulation index.
Modulation index the amplitude of the modulating frequency in proportion to the amplitude of the carrier.
RUclips won't let me save the edit, the word "is" is missing of the word index. I'm sure I typed it, but underlined and maybe RUclips doesn't like underlines.
Sorry to be ignorant, but i have a 145.tranciever can i use that for the cw carrier reciever with out that mixer you have...? so do both radios have to be on the same freq? Fred in uk
Yes, if you've got an all mode 145 Rx with an S meter, you can use that, and you dispense with the mixer that I had to use.
I like most people dont have a signal generator, deviation meter, i do have a scope, but can you just use 2 x 144 mhz transievers and shout or whistle in the mic..is that possible? Fred in UK
No, you can't use a whistle or shout. In order to use this Bessel Null technique, you need to have precise control over the modulating frequency (in order to have precise control over the modulation index).
Hi, there's something I don't quite understand. Doesn't the amplitude of the modulating frequency affect the amount of deviation too? How did you decide where to set the amplitude of the modulating frequency in your example? Thanks.
The amplitude generally control the deviation. In most FM transmitters, and ALC circuit limits the maximum amplitude to the modulator, so you usually adjust the input amplitude up until the deviation doesn't change much anymore (assuming the service manual doesn't specify a particular value.)
Ah very good. I duplicated the method you used here using a USB SDR to see the spectrum @ 145 MHz, and check for a null, but I wasn't sure about the amplitude of the modulation signal. The method worked out perfectly. Thanks for the instructional video - N5INP
I was playing around with this on my computer yesterday and from the Wikipedia definition produced a Bessel null using the FM equation: y = Ac * cos(wFc * t + MI * sin(wFm * t)), where Ac is the overall signal amplitude, wFc is the carrier frequency (in radians/sec), t is time, MI is the modulation index, and wFm is the modulating frequency (single tone). The part that is relavant to your question is MI, which is the . The deviation is the (highest) modulating frequency multiplied by the modulating index, so the amplitude and frequency both affect deviation.
There are some missing words - I underlined them and RUclips stripped it from my comment, after "MI, which is the ", add "amplitude of the modulating frequency".
Hi, will the impeadance mismatch in your setup will cause significant reduction in sidebands amplitud that would affect the deviation? Significant sidebands are those that have a voltage of at least 1% of the unmodulated carrier.
No
Thanks, but what about the Splitter?
Yes I see how you do it now, can you tell me when you connect a Deviation meter to a radio antenna socket etc, do you have to be careful how much power you put into the meter, is say 4 watts too much, do you use an attenuator? Fred uk
Yes, you absolutely have to be careful about how much power you put into a deviation meter. The specifications of the meter will you the upper limit. Most often, you'll use a RF sampler, tap or directional coupler, or even loosely couple via an antenna. Rarely will you ever directly connect the transmitter to the meter.
***** Ok thank you for information, video is good Fred uk
Hi there I bought a Heathkit deviation meter IM 4180 like you have, came from Ohio to UK, doesn't really work, wondered if you had any ideas, the tuner works and the audio amp works, just the pointer hardly moves on tune on low dev setting, the dc power side seems ok, I suspect U 102 to U 104, the 8 volt regulator is set to 7.5 Fred in UK
I haven't studied the circuit on that unit yet, but the manual provides a pretty good description and adjustment/calibration adjustments. I'd suggest starting with doing the adjustments from the beginning. One thing I found with mine when I bought it was that one of the trimmer pots was intermittent - so examine those carefully.
***** Ok thanks, I think there are 3 of them that I've seen,,,,cheers Fred uk
We have this meter working now, I reversed the polarity by accident, we had to change all the IC;s which are plug in and the 8 volt regulator..Fred uk
Hi there, how exactly is the signal generator connected to the receive radio, is it direct onto your mixer board or through and attenuator..i presume its kind of in Parallel with the output carrier, how much attenuation do you have on the T piece, would a BNC with pin removed work..or is that too much....Fred in UK.
I only needed to use a RF signal generator in this case because I had to use the mixer to down-convert the 2m signal into the HF band. So, the signal generator is used as an LO for the mixer - it is not connected to the radio.
***** I think I said the wrong thing, the instrument you used that was set to 2.7khz and you wound down to 1.90khz, how was that connected to the radio, if it was..as you used that freq for the equasion sorry if i'm being stupid here hi Fred.
Tom Smith OK, got it. That generator was creating the audio signal and was being fed into the MIC input of the radio. I fed it through an attenuator to bring the level down to typical microphone signal levels. I adjusted this frequency until the carrier was nulled out, then multiply this value by 2.405 to arrive at the deviation level.
***** Ah right, I should have realized that, makes sense now...thanks, I have watched the video several times, it is good...cheers Fred.
Do you always use 2.405 kHz as the multiplier? Fred uk
Yes, because that corresponds to when the carrier is nulled out.
it'll take time to digest all this....but thanks.
Good video Alan. Subscribed. 73 de AA7YA.
Sneaky :-)