Split Topic - FM Technical Argument

[kb2vxa]

Hi guys,

I see you dancing around the answer but not quite getting to it. The output voltage produced by the detector is directly proportional to the FM deviation. The unit or system in question has lower deviation than the others, you weren't clear if it's just one unit or all units on the frequency. There can be two reasons for this, like they said one person may not be using the mic correctly or the transmitter deviation is set lower than "normal". There can be two reasons for this as well, set too low by accident or the unit/system is using the new +/- 2.5 KHz deviation rather than the older +/- 5KHz standard. The volume on my scanner does the same thing, the new fire/EMS radios in town use the new standard while the railroad and Coast Guard still use the old.

BTW, the even older standard was +/- 15KHz with 30KHz channel spacing. I once had an old RCA base unit that was converted by changing the IF filter and lowering the deviation pot. The idea behind changing the filter is to bring back the receive volume to where it should be, the bandwidth has to match the deviation, if it's too wide the volume will be low.

That's why the new systems don't sound loud enough on a scanner with the IF wide enough to accomodate all the signals plus the frequency offset needed for proper tracking, the UHF band is offset from the VHF by 5KHz. Look at the frequency assignments and do the math (IF conversion) and you'll see why the last IF is 14KHz wide.

Yeah, I know, you asked the time and I told you how to build a clock. (;->)

 

[N_Jay]

The IF filter bandwith change when narrow banding a radio is to regain the adjacent channel rejection specification.

It has nothing to do with the recovered audio volume.


Along with deviation, most decent FM transmitters also have deviation limiters. The combination of the deviation setting (the gain setting from mic to modulator) and the deviation limiter together create a compressor to boost under-modulated signals while preventing over modulation.

 

[kb2vxa]

Hi NJay and all,

Please don't think I'm being argumentive but:

"The IF filter bandwith change when narrow banding a radio is to regain the adjacent channel rejection specification. "

That is distinctly a part of it, you don't want "bleed over" from the new splinter frequencies. Please note, adjacent frequencies are staggered from area to area like TV channels. No one service area has two channels side by side. This is because the IF skirt isn't exactly square, the bandwidth is specified at the -6dB points.

"It has nothing to do with the recovered audio volume. "

Then why did changing the 30KHz filter in my RCA to 15KHz bring the detector output voltage back to normal? The same is true with my ham rig which has three FM IF filters, wider equals less volume, narrower equals more. I'm not good at complex math so please explain in simple English.

"Along with deviation, most decent FM transmitters also have deviation limiters. The combination of the deviation setting (the gain setting from mic to modulator) and the deviation limiter together create a compressor to boost under-modulated signals while preventing over modulation."

True and Motorola uses digital audio processing, the trade name is "serisoidal modulation". The thing to remember for those unfamiliar is the mic gain pot is before the limiter and the deviation pot is the last in line before the reactance modulator so maximum peak "swing" is set after processing.

Now here's an interesting note, my Azden AZ-20A HT uses transmit volume compression. I have never seen this in any other ham HT, I can close talk it or hold it a foot away and it sounds the same on air. That gives me the advantage with the speaker/mic clipped to my lapel, I don't give it any special consideration, just press the switch and talk.

 

[N_Jay]

I would guess that when you change IF filters, you are also changing audio gain somewhere in the system.

If not, I would be perplexed as to how a filter change could effect volume.

Look at it this way.

A 2.5 kHz wide signal will pass either a 12.5 kHz or a 25 kHz filter without significant attenuation.

That same signal applied to the same demodulator and audio chain would have to "magically" come out at different volume levels, for your assertion to be correct.

As to why Ham gear does not have deviation limiters, ask those folks who keep insisting that there is no difference between a part 47 and a part 90 certified radio!

 

[DaveNF2G]

I have to disagree with N_Jay once again.

...

Now that you've all regained consciousness after fainting, I'll explain. :twisted:

Narrowing the receive bandwidth increases the signal-to-noise ratio, which results in a louder sounding signal, compared to the same signal recovered from within a wider passband.

The transmit volume (actually, deviation in an FM signal, as somebody already pointed out), does not change, but the receiver does not know that. It sees a modulated carrier that occupies a higher percentage of the channel, thus appearing (mathematically) to have more deviation, so it is able to transfer more desired audio and less noise energy into the audio stages. As if by magic, the signal being received by the radio with narrower bandwidth sounds louder.

Try changing your scanner's receive mode from "narrow" FM to "wide" FM while listening to a normal two-way radio transmission. The voice on the signal will sound "weaker" or "softer" (and slightly less loud), despite the fact that nothing has been changed beside the bandwidth of the receiver. (If your receiver has AGC, disable it if possible for this experiment. It tends to mask the difference on strong signals.)

 

[Al42]

"It has nothing to do with the recovered audio volume. "

Then why did changing the 30KHz filter in my RCA to 15KHz bring the detector output voltage back to normal? The same is true with my ham rig which has three FM IF filters, wider equals less volume, narrower equals more. I'm not good at complex math so please explain in simple English.

Discriminators (and ratio detectors) have a design specification called volts/frequency (it's just a frequency to voltage converter). If you want the same audio level for a signal with half the deviation you change the design to give twice the v/f. It's a simple resistor change if the circuit is designed for that, so switching between WFM, FM and NFM in a scanner is simple.

 

[N_Jay]

I have to disagree with N_Jay once again.

...

Now that you've all regained consciousness after fainting, I'll explain. :twisted:

Narrowing the receive bandwidth increases the signal-to-noise ratio, which results in a louder sounding signal, compared to the same signal recovered from within a wider passband.

The transmit volume (actually, deviation in an FM signal, as somebody already pointed out), does not change, but the receiver does not know that. It sees a modulated carrier that occupies a higher percentage of the channel, thus appearing (mathematically) to have more deviation, so it is able to transfer more desired audio and less noise energy into the audio stages. As if by magic, the signal being received by the radio with narrower bandwidth sounds louder.

Try changing your scanner's receive mode from "narrow" FM to "wide" FM while listening to a normal two-way radio transmission. The voice on the signal will sound "weaker" or "softer" (and slightly less loud), despite the fact that nothing has been changed beside the bandwidth of the receiver. (If your receiver has AGC, disable it if possible for this experiment. It tends to mask the difference on strong signals.)

I would like to agree wityh you but as HAL said:
"Sorry Dave, I can't do that!"

We will continue to disagree (probably because you are mistaken).

We are dealing with FM so S/N does not affect volume, it only affects quieting.

Lets go back to my analogy.

We have a signal well above 20 dBq. For arguments sake a dead carrier.
We send that signal through a 25 kHz filter and to a discriminator, and through a 12.5 kHz filter and to the same design discriminator.
Since it is a near zero bandwidth signal the filters are irrelevant (Assuming the center frequency loss is the same).

The receivers are now both fully quieted. Background noise is irrelevant.

We modulate that signal (FM) with a waveform that keeps it at less than 12.5 kHz bandwidth.

How does the discriminator know what filter it is going through????

I will bet you dollars to doughnuts that when you change bandwidths you also change the gain somewhere in the audio chain.

:twisted: :wink:

 

[DaveNF2G]

The fact remains that, the more closely the receiver bandwidth matches the b/w of the transmitted signal, the "louder" the signal will sound subjectively to the listener. Beyond that, the technical details are a distinction without a difference.

 

[Al42]

The fact remains that, the more closely the receiver bandwidth matches the b/w of the transmitted signal, the "louder" the signal will sound subjectively to the listener.

If you change bandwidth, and the signal is narrower than the narrowest bandwidth, you won't notice any apparent difference in audio level - unless there's considerable lack of quieting. Try it - get a receiver with a ceramic IF filter and wire in a switch so that you can switch between 2 filters - say a D and a G (or whatever would be appropriate for the signal you're using for a test). If the signal is dead full quieting you won't notice any difference. (and a VU meter won't show any difference on a steady tone).

 

[N_Jay]

The fact remains that, the more closely the receiver bandwidth matches the b/w of the transmitted signal, the "louder" the signal will sound subjectively to the listener. Beyond that, the technical details are a distinction without a difference.

LOL,

(Sounds like an "I give up, but won't admit I'm wrong" answer to me.) :roll:

I'll agree with you under the condition that there is significant background noise (or lack of quieting in the FM world).

Of course that answer is both subjective and self fulfilling.

Its like saying; You know if you turn off the TV the scanner gets louder.

Sure, try turning up the TV real loud, and you will see that the scanner SUBJECTIVLY gets softer! :wink:

 

[N_Jay]

The fact remains . . . . . subjectively to the listener. . . . the technical details are a distinction without a difference.

LOL, I had to read this again.

Let's see the FACT is SUBJECTIVE

and the DETAILS only indicate a DISTINCTION but without DIFFERENCE?

 

[DaveNF2G]

The fact remains that, the more closely the receiver bandwidth matches the b/w of the transmitted signal, the "louder" the signal will sound subjectively to the listener.

If you change bandwidth, and the signal is narrower than the narrowest bandwidth, you won't notice any apparent difference in audio level - unless there's considerable lack of quieting. Try it - get a receiver with a ceramic IF filter and wire in a switch so that you can switch between 2 filters - say a D and a G (or whatever would be appropriate for the signal you're using for a test). If the signal is dead full quieting you won't notice any difference. (and a VU meter won't show any difference on a steady tone).

I have to ask you guys who are being so all-fired technical about the whole thing - why the need to show off? How many scanner listeners know or care about VU meter measurements? How many scanner listeners enjoy dead full quieting on all signals? Please climb down out of your ivory towers and join the rest of us in the real world.

Life is subjective. We don't take measurements to validate all of our experiences.

 

[loumaag]

You guys may continue this discussion as a sperate topic; however, please keep it civil or it will be locked. :!:

 

[N_Jay]

I have to ask you guys who are being so all-fired technical about the whole thing - why the need to show off? How many scanner listeners know or care about VU meter measurements? How many scanner listeners enjoy dead full quieting on all signals? Please climb down out of your ivory towers and join the rest of us in the real world.

Life is subjective. We don't take measurements to validate all of our experiences.

Dave, You post a lot of real good stuff here.

I don't know why you are you are having such a hard time with this one.

Just say "Hmmm, You learn something new every day".

It realy does not hurt that bad.

 

[kb2vxa]

Hi Dave and all,

"The fact remains that, the more closely the receiver bandwidth matches the b/w of the transmitted signal, the "louder" the signal will sound subjectively to the listener. Beyond that, the technical details are a distinction without a difference."

Thanks for that summation for the less technical minded, your previous more detailed explanation did it for me. I had rather suspected that but held back comment because after some things happened on other threads I'm less prone to sticking my neck out now.

NJay, I'm familiar with S/N ratio but that's not applicable to FM, with that mode it's called "quieting". To put it simply, with AM and SSB (amplitude related modes) most of the noise accompanies the signal (noise floor) while with FM all that white noise you hear is generated by the receiver itself. Since it's amplitude related the limiters ahead of the discriminator take care of it when the IF signal gets strong enough to drive them into saturation. With a ratio detector limiters are not needed, it's immune to amplitude variations.

Back to S/N and amplitude related modes, considering the nature of HF in particular (VHF is inherently quiet) older receivers still "heard" rather well over the noise floor except for the "cheap" ones which were noisy in themselves. The tube type CB rigs were horrible. Along comes solid state and things improve somewhat but with the advent of the J-FET in particular internal noise dropped almost to zero. Today S/N ratio is hardly worth considering because communication receivers are all good all the way 'round. Those crappy portables "ain't worth the powder to blow them to hell" so let's just class then with the psychic TV shows, "for entertainment purposes only".

With today's spectrum pollution a quiet location is worth 10 receivers so unless you live out in the styx somewhere you'll never notice the difference for the most part.

 

[N_Jay]

Hi Dave and all,

"The fact remains that, the more closely the receiver bandwidth matches the b/w of the transmitted signal, the "louder" the signal will sound subjectively to the listener. Beyond that, the technical details are a distinction without a difference."

Thanks for that summation for the less technical minded, your previous more detailed explanation did it for me. I had rather suspected that but held back comment because after some things happened on other threads I'm less prone to sticking my neck out now.

NJay, I'm familiar with S/N ratio but that's not applicable to FM, with that mode it's called "quieting". To put it simply, with AM and SSB (amplitude related modes) most of the noise accompanies the signal (noise floor) while with FM all that white noise you hear is generated by the receiver itself. Since it's amplitude related the limiters ahead of the discriminator take care of it when the IF signal gets strong enough to drive them into saturation. With a ratio detector limiters are not needed, it's immune to amplitude variations.

Back to S/N and amplitude related modes, considering the nature of HF in particular (VHF is inherently quiet) older receivers still "heard" rather well over the noise floor except for the "cheap" ones which were noisy in themselves. The tube type CB rigs were horrible. Along comes solid state and things improve somewhat but with the advent of the J-FET in particular internal noise dropped almost to zero. Today S/N ratio is hardly worth considering because communication receivers are all good all the way 'round. Those crappy portables "ain't worth the powder to blow them to hell" so let's just class then with the psychic TV shows, "for entertainment purposes only".

With today's spectrum pollution a quiet location is worth 10 receivers so unless you live out in the styx somewhere you'll never notice the difference for the most part.

Warren,
Please try to use the quotation features of the forum system, as it makes your postes much eaiser to read.

 

[PhilJSmith67]

I've got to throw my 2 cents in here.

Changes in the IF bandwidth are directly proportional to the selectivity of the receiver. As long as the IF bandwidth remains above the point that allows the max deviation of the input signal to pass, narrowing the IF bandwidth should have minimal effect on the audio volume.

Any scanners I have encountered with separate settings for WFM/FM/NFM do not only switch the IF stages; they must also change the V/F ratio at the discriminator. This is precisely how the RF board in my Regency/AOR MX7000 handles the WFM/NFM output pin from the micro. I cannot see how a receiver would be able to function without switching both the IF and V/F in concert. If I tuned in a WFM signal in NFM mode, but the receiver changed only the IF while failing to increase the discriminator's V/F ratio, I would end up with a massively distorted signal, but at a very low volume level.

 

[PhilJSmith67]

cont'd

I have in front of me an old Electra/Bearcat DX1000 general coverage receiver.

It is tuned to a signal of varying strength on 29.620 in FM mode. Changing between the selectable IF bandwidths of 2.5 kHz, 6 kHz, and 12 kHz has some effect on the quality of the audio, but no effect whatsoever on the volume level. The discriminator is a fixed circuit in this receiver.

 

[N_Jay]

Re: cont'd

I have in front of me an old Electra/Bearcat DX1000 general coverage receiver.

It is tuned to a signal of varying strength on 29.620 in FM mode. Changing between the selectable IF bandwidths of 2.5 kHz, 6 kHz, and 12 kHz has some effect on the quality of the audio, but no effect whatsoever on the volume level. The discriminator is a fixed circuit in this receiver.

No way!!! :twisted: :twisted: :twisted: heheheehe :wink:

 

[kb2vxa]

Hi Phil and all,

"Any scanners I have encountered with separate settings for WFM/FM/NFM do not only switch the IF stages; they must also change the V/F ratio at the discriminator. This is precisely how the RF board in my Regency/AOR MX7000 handles the WFM/NFM output pin from the micro. I cannot see how a receiver would be able to function without switching both the IF and V/F in concert."

Well, you've never encountered an IC-706Mk2G or a National Radio NC-303 with converters and FM adapter. Never say "must" or you will sometimes be proven wrong. Just because you "cannot see" doesn't mean someone else hasn't seen it. (;->)

BTW, that 303 and an oscilloscope connected to the IF made an excellent tool for analyzing HF digital and analog multiplexed signals before the PC sound card software came along.

I got my answer a while back so from that point on thanks for the entertaining arguments. Like Bart said, "Well, I got my laugh, I'm outa here.". (;->)

BTW, don't lock the thread Lou, you'll spoil their fun. If you haven't seen a bunch of Rabbis around a table arguing law ya ain't seen nuttin yet. (;->)