Maker Pro
Maker Pro

Using Detuned AM/FM Radio as Spectrum Analyzer Input

epsolutions

Sep 7, 2019
106
Joined
Sep 7, 2019
Messages
106
I would like to use an inexpensive battery powered radio to capture and boost the entire AM/FM frequency band to display on a spectrum analyzer.

Is this as simple as shorting out the variable tuning cap, and bypassing the demodulation stages, or would other component changes be required?

Assuming the above would work, what amplification factor might I expect?

If the modulation stages are left in-circuit, what would the audio output look like?

am_fm_radio.jpg

PS: I know it is possible to build an amplifier from scratch, but I already have the radio and this is an experiment.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
A radio has many tuned LC circuits to select only one radio frequency but you want to select very many radio frequencies.
Most radios have AGC (automatic gain control) which adjusts the gain for each radio frequency.
The radio signal level from an FM radio station is constant. But the radio signal level from an AM station varies with the audio level.
Therefore your idea will not work.

A spectrum analyser produces a display of levels at many radio frequencies, not audio.

An audio spectrum analyser will display demodulated audio levels at all audio frequencies. At work I used an audio spectrum analyser with a pink noise generator feeding an audio amplifier for measuring speaker frequency response.
 

epsolutions

Sep 7, 2019
106
Joined
Sep 7, 2019
Messages
106
Is it not possible to bypass the LC circuits? Even without a diagram, those coils are easy to spot on the PCB. As in my OP, I would try simply shorting out the variable cap first for both bands.

Yes, I know about AM. All I want is the raw single of everything the antenna is picking up. If there are modulations, that is fine. But I am not interested in demodulating them, or audio per se. My spectrum analyzer is IF/RF.

What I am doing is checking signals in very fringe areas.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
The parallel LC parts are a certain fairly high impedance at resonance.allowing amplification. Shorting the LCs will produce no amplification and the transistors will produce lots of smoke.
Please learn about the basics of electronics.
 

epsolutions

Sep 7, 2019
106
Joined
Sep 7, 2019
Messages
106
Thank you for your advice. I may try anyway just to learn more about radio circuits. Maybe replace the LC tuner with series resistance and see what happens.
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,632
Joined
Nov 17, 2011
Messages
13,632
Maybe replace the LC tuner with series resistance and see what happens.
As @Audioguru explained that will not work.
It sounds as if you are not familiar with the operation of a spectrum analyzer. You seem to be trying to display the energy content fo all frequencies within a band at the same time. This is not possible with reasonable effort. To do this, you'd have to split the whole band into a number of sub bands - the finer the expected resolution, the more sub bands you will need. Next you will have to provide the same number of receivers as you have sub bands, one receiver for each band, tuned to the center of the band. Next you will have to collect the outputs of all receivers and display their amlitudes versus the sb bands to sho a spectrum. As I mentioned at the beginning: impractical.

Or you could use a single wide-band receiver, sample the input signal and do a fast fourier transform to compute the energy contents of the different sub bands. But then you'd have to sample at a multiple of the input frequency to get a faithful representation of the input signal for computing the fourier transform. As I mentioned at the beginning: impractical.

Instead a typical spectrum analyzer would use a tuned receiver, scan the input signal by tuning from the lowest input frequency to the highest input frequency and register the amplitude of the signal at each tuning point. The number of tuning stpes depends on the bandwidth of the tuned receiver. Essentially you would split the full input bandwidth into small chunks with a bandwidth equal to the tuner's bandwidth.
Example:
input bandwidth 80 MHz ... 100 MHz --> 20 MHz
tuner bandwidth 100 kHz
-> 200 sub bands centered at 80 MHz, 80.1 MHz, 80.2 MHz, ...
After each scan you can update the display to show the energy content of the frequencies in the different sub bands - a spectrum.

You will also have to disable the automatic gain control (AGC) of the tuner. AGC will try to equalize the amplitude of the received signal for each frequency and if the AGC works well, the result will be a flat spectrum without any hint of the energy distribution.
Of course, you could keep the AGC and use the AGC output (the control signal that turns gain up or down) as the indicator of energy content:
high gain = low energy,
low gain = high energy in the respective frequency band.
You will still have to scan the tuner across the whole band.

If you do have a spectrum analyzer (it is not clear to me from your post #1 whether you do or not), then an amplifier should not be necessary. HF spectrum analyzers are usually rather sensitive. Connecting an antenna to the analyzer's input may be all you need. If the signal of the passive antenna is too weak, try an active antenna or a simple antenna amplifier. These are wide-band and better suited to the task than a radio receiver's input circuit.
 

epsolutions

Sep 7, 2019
106
Joined
Sep 7, 2019
Messages
106
I am confused by your statements regarding spectrum analyzers. I have two R&S FHS6's. They can display at once once the entire spectrum of 100KHz to 6GHz, or any window in between.

In using the AM/FM radio as an input, I am only interested in monitoring the frequency spectrum and not amplitude, so the AGC may not be a problem.

It comes back to the effect of de-tuning the radio to acquire the widest possible spectrum, and implementing whatever other circuit mods this might require.

I guess this is another case of trying it and seeing firsthand what happens. As I understand the audio out will be similar to white noise. I would tap the signal before the demodulation stages.

Thank you for the advice so far. Unfortunately, active wide band antennas and amps are not within my budget.
 

bertus

Moderator
Nov 8, 2019
3,259
Joined
Nov 8, 2019
Messages
3,259
Hello,

What is your intention with the spectrum analyser?
Do you want see signals from the surroundings?
Perhaps a simple amplifier with a short wire antenna will do the trick.
This page shows you several amplifiers:
https://www.qsl.net/ok1cfp/ok1cfp-vhf-uhf-zesilovac-amplifier-mar-3.htm
The schematics show the MAR-3, but there are many more.
See the attached datasheet for more info.

Bertus
 

Attachments

  • mmics.pdf
    316.8 KB · Views: 0

epsolutions

Sep 7, 2019
106
Joined
Sep 7, 2019
Messages
106
Yes, my aim is to monitor the AM/FM frequency spectrum in poor reception areas.

The amp circuit is interesting but I am not skilled in RF construction and was looking for a ready-made solution. Even if there are compromises, at least it will be a starting point.
 

bertus

Moderator
Nov 8, 2019
3,259
Joined
Nov 8, 2019
Messages
3,259
Hello,

Did you see the construction pictures at the given site?
They seem to be rather simple.

Bertus
 

epsolutions

Sep 7, 2019
106
Joined
Sep 7, 2019
Messages
106
Yes. Just that it's an unfamiliar area for me. I already have a few "junk" radios around looking for excitement. Maybe it's a silly idea, but that's how I got started in electronics.
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,632
Joined
Nov 17, 2011
Messages
13,632
They can display at once once
They can't. It only looks so to you. When you look at the input circuit (see here, page 27) you will see that the input circuit is essentially a tuner as I described above in the section starting with
Instead a typical spectrum analyzer would use a tuned receiver,
What these devices do to make it look as if they were displaying the whole spectrum at once is: they continuously scan the frequency range and update the display with each pass.
I am only interested in monitoring the frequency spectrum and not amplitude
That makes no sense: A spectrum is the graphical display of amplitude (or energy) versus frequency. IF you do not care for the amplitude you'll have a flat line with no information whatsoever.
It comes back to the effect of de-tuning the radio to acquire the widest possible spectrum
As described above: this doesn't work.
active wide band antennas and amps are not within my budget.
< $ 10 ?
 

epsolutions

Sep 7, 2019
106
Joined
Sep 7, 2019
Messages
106
I think my wording has been misunderstood. The idea of the radio, valid or not, is simply an amp in front of a SA. The fact that the latter scans does not seem relevant.

As previously explained, I would be satisfied to see the frequency spectrum, and live with whatever effect the AGC may have on the amplitude. I am still curious to check the outcome, whether it "works" or not.

I have never seen a wide band RF amp for less than $10. Radios I can buy at garage sales for a few dollars.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,875
Joined
Jun 21, 2012
Messages
4,875
I have two R&S FHS6's.
Really? Here is a comprehensive "datasheet" describing the handheld Rhode & Schwarz FSH6 Spectrum Analyzer. Searching with Google, I couldn't find any information on "R&S FHS6's" that you claim to have. But, congratulations on having two of whatever it is that you DO have.

If I were looking to measure spectrum occupancy or signal density in both the AM radio band (550kHz to 1200kHz) and the commercial FM band (88MHz to 108MHz) in fringe reception areas, then I would connect an antenna to the spectrum analyzer, with or without a suitable pre-amplifier. You could use a ferrite loop-stick (un-tuned) antenna to sample the AM spectrum... or perhaps a largish flat coil of wire to create a small magnetic loop antenna. Either approach will be somewhat directional however. Perhaps a mono-pole vertical antenna will suffice for the FM spectrum, but these are sensitive to polarization. I would not attempt to measure both the AM and FM bands simultaneously, and certainly not with the same antenna..

It is not immediately clear to me that you actually know how a Rohde & Schwarz spectrum analyzer works. Maybe this datasheet (in English) will help. Perhaps that isn't necessary in your line of work.

Most spectrum analyzer input(s) are quite sensitive to damage from excessively strong signals. The low-noise mixer, to which is applied the input signal(s), is easily damaged. Be careful to measure (and/or attenuate) unknown signals before connecting them to the spectrum analyzer input(s).

And it is also not obvious that you have any idea how an AM/FM radio works, but feel free to correct me if am wrong about any of the above.

Virtually ALL AM/FM radios (and any other serious radio beyond crystal sets and ancient TRF models) operate on the superheterodyne principal. This means they all mix (heterodyne) the antenna signal (possibly amplified with a tuned RF amplifier) with a local oscillator (LO) signal to obtain a single range (or spectrum) of sum and difference frequencies. Both the sum and difference frequency spectrum equally represent the frequency spectrum of the received signal, although typically only the difference frequency spectrum is amplified and later demodulated to an audio frequency spectrum. Usually, the difference frequency spectral range is amplified by an intermediate frequency or IF RF amplifier, tuned to accept whatever modulation bandwidth is expected from the signal source. Note that the spectrum band of frequencies is different for each transmitting station, but by changing the local oscillator (LO) frequency, each station's spectrum can be made to fall within the IF amplifier bandwidth. Thus the IF amplifier is tuned to amplify a particular portion of the radio spectrum whose information content, regardless of modulation, is a duplicate of the original transmitted spectrum.

For the sake of completeness, I should mention that most communications receivers of newer design use direct conversion instead of superheterodyne conversion, and digital sampling of the antenna signal. With direct conversion, the local oscillator is tuned to the carrier frequency (or a replication of the original carrier frequency) and the difference spectrum is processed as a pseudo-audio signal. With in-phase and quadruture-phase demodulation, a pair of signals is created that can be used for a spectral display, much like a spectrum analyzer but with limited frequency range and functionality.

It is very efficient to design an RF amplifier that need respond to only a fixed and relatively narrow band of frequencies. Those little "cans" you see scattered about the innards of your AM/FM radio are shields for the LC tuned transformers that are connected to the inputs and outputs of the IF amplifiers. Because of the huge difference in frequencies used between the AM and FM bands, the IF RF amplifiers for each of these bands are also different and the intermediate frequency provided by the local oscillator is also different. Bear in mind that there is also a LOT of very active spectrum space between the largest AM broadcast frequency and the lowest FM broadcast frequency. Your AM/FM receiver is designed to ignore this spectrum.

Much of this "ignored" spectrum space, up to about 30MHz or so, is usually described as the the high-frequency (HF) "shortwave band" and is accessible by superheterodyne communication receivers. Some communication receivers have more than one intermediate frequency and are called dual-conversion receivers. The higher received frequencies, as selected by a local oscillator, are amplified in the first IF amplifier and then that output is mixed with a second local oscillator to create a second IF band of frequencies at a lower intermediate frequency.

The Rohde & Schwarz spectrum analyzer operates much like a communications radio, except the tunable local oscillator (LO) In a spectrum analyzer, the LO frequency, is not used to select a particular band of frequencies. The LO frequency is swept from a lower frequency to a higher frequency so the output at any instant represents a "snapshot" of the radio frequency energy present in a narrow, usually operator-adjustable, frequency band surrounding the local oscillator frequency.

my aim is to monitor the AM/FM frequency spectrum in poor reception areas.
Gee, us po' folk do this all the time when we are out traveling in the boondocks. We tune the car radio from one end of the band to the other hoping to hear a broadcast radio station. Hmmmm. Nothing but static noise on AM. FM is quieter, but still no stations are heard. Maybe we are in a radio "quiet zone". The nearest one to me is near Greenbank, WVA. My wife and I have never left the broad interstate trail to visit the National Radio Telescope, but I have, on my own, found some relatively quiet places while traveling in the great American West. Such "radio quiet" places are difficult to find, especially for AM signals at night: AM covers a LOT more distance at night than FM.
 

epsolutions

Sep 7, 2019
106
Joined
Sep 7, 2019
Messages
106
The assumption was that the radio I use will, as many do, have an RF amp prior to the mixer. If the RF filter can be disabled, my hope was that this would result in a "wide band" output. I would check the level before connecting the SA.

On the other hand, if the tuning is left intact, and the SA is connected after the IF amp, the numerical RF can be interpolated by subtracting the measured IF from the LO frequency.

For a persistent display on the SA of general reception using the IF, I could apply a fast ramp to the mixer. Maybe even white noise. But that is more than what I originally had in mind.

Anyway, this is just fiddling around to see what happens. Because of this, and my age, I do not always remember my test equipment by name. To your credit, you were not confused.
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,632
Joined
Nov 17, 2011
Messages
13,632
the numerical RF can be interpolated by subtracting the measured IF from the LO frequency.
I have no clue how this should work. After the mixer there are filters to strongly limit the bandwidth of the signal. Applying the Sa after the mixer makes no sense unless you want to analyze the mixed and filtered signal. This signal will give you no clue as to how the signal spectrum at the input (antenna) looks like.
I would connect an antenna to the spectrum analyzer, with or without a suitable pre-amplifier.
That's what I suggested in post #6. and this is how my colleagues at work use the SA.
 

bertus

Moderator
Nov 8, 2019
3,259
Joined
Nov 8, 2019
Messages
3,259
Hello,

I also mentioned the antenna with an amplifier in post #8.
The radio will have a to small bandwidth to see the signals.

Bertus
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,875
Joined
Jun 21, 2012
Messages
4,875
Because of this, and my age, I do not always remember my test equipment by name.
This sounds familiar... IIRC. :rolleyes:

Spectrum analyzers (SAs) or, more generally, vector network analyzers (VNAs), are fascinating pieces of equipment. One of these days I hope to own a nice VNA to supplement the RigExpert antenna analyzer I use in my amateur radio (AC8NS) hobby. Gotta stay under 1GHz and under $US 500 though since I am retired and have to carefully budget how much of the amateur radio spectrum I can afford to use.

At my age, I do not always remember what day of the week it is. Test equipment... not so much. All I currently own is a digital multi-meter and a digital storage oscilloscope plus amateur radio stuff... well, maybe a logic probe or two or three. I used to own an HP logic pulser for TTL, guaranteed to pulse a high node to low for a few microseconds or a low node to high, also for a few microseconds. Haven't found a need for that puppy during this century, but it will probably turn up sooner or later anyway.
 
Top