EEVblog Electronics Community Forum

General => General Technical Chat => Topic started by: EEVblog on August 12, 2018, 11:51:14 pm

Title: Why do megaphones still sound terrible?
Post by: EEVblog on August 12, 2018, 11:51:14 pm
You know what I mean, that horrible distorted scratchy narrow-band sound, to the point of being almost unintelligible?

Speech isn't hugely wide band, and there have been massive advancements in acoustics and electromagnetic tech in the last few decades. Yet megaphones still sound like the 1970's
I can understand limitations in the horn type arrangement, but why still so bad?

Title: Re: Why do megaphones still sound terrible?
Post by: james_s on August 13, 2018, 12:10:38 am
Has the design really changed much? I don't know how many companies make megaphones but it wouldn't surprise me in the least if they were still the same 1970s designs being produced. Part of it also might be the fault of the user, people yell into the things and/or turn the volume up all the way trying to make it as loud as possible when it would probably work better to speak into one using a normal voice.
Title: Re: Why do megaphones still sound terrible?
Post by: PlainName on August 13, 2018, 12:17:05 am
It wouldn't be due to filtering to prevent feedback, would it?
Title: Re: Why do megaphones still sound terrible?
Post by: JS on August 13, 2018, 12:18:42 am
Is this a trick question?

The size and weight constrains makes the coice of driver pretty narrow, you need loud, light and small or the thing would be unusable, try to find a driver with those properties able to reproduce 200Hz rasonably well. Then add a flatish response up to a few kHz, but small and light will dp fine in higher freqs. I think this is the limitong factor, running a small driver as far as it goes and you get distortion and poor LF response (LF in speech range, let's say 200Hz depending on the speaker) we are likely talking of a few 100W, 8" speaker to get nice quality, even a neodymium one would weight too much for a lady to hold for long.

Also, acoustic feedback in a small and lightweight enclosure will start becoming another problem, wich is the one modern technology could improve on (DSP anti feedback or feedback cancellation) up to a certain point.

Then, in the acoustic side of things, LF directivity is only driven by size, not much you can do about it, to keep a reasonable directivity down to 200Hz you want a thing closer to 1m wide, way too big for one of those. There are some advanced experiments on that respect but probably way to expensive and inefficent for a cheap battery powered device.

Oh, battery power, LF needs more power compared to HF, so you cant have lightweight battery life and high LF SPL. This also improved by new battery technology.

Those would be my firat guesses, mainly driver restrictions and to a lesser extent directivity. Constrains mainly impossed by physical package of the product.

JS

JS

Title: Re: Why do megaphones still sound terrible?
Post by: Nusa on August 13, 2018, 12:27:05 am
Horn speakers have poor high frequency spread, so the treble sucks. If it's pointed directly at you, it's a bit better, but it still sucks.
Horn speakers have to be large to get good bass, so for hand-carried sizes the bass sucks.
Horn speakers are better than cone speakers at producing volume.

A good experiment would be hooking a PA output to a studio speaker and see what you get.
Title: Re: Why do megaphones still sound terrible?
Post by: BrianHG on August 13, 2018, 12:38:33 am
I've seen the inside of one at summer camp back in the early 80s.  It was a huge ceramic tweeter inside the center of that cone.  I also guess the amp was poorly biased class B to save on battery life.  This way, if the mic was left on, only when sound went through was there any current draw.  It had only 4 x D or C battery cells above the handle and they lasted too long for a higher current drawing amp with tight biasing.  It was all discrete components.  The sound wasn't scratchy like you describe, or, like what I see in protest movies where event organizers have a cheap small megaphone where it sounds like the batteries are about dead, but, still, it's frequency range began above around 200hz to the upper 5Khz.  Oddly enough, the outdoor speakers atop the telephone polls on the camp's PA system, which looked like 5x-10x larger versions of a megaphone's speaker, actually sounded really good, though, there was a 500 watt plugin stationary amp driving all of them in parallel in the office with one of those tall silver CB style mikes on a stand.

As for today, I guess people just don't expect better, so, even if parts improved, the design hasn't.

My guess, if you want a good sounding one, get a top-tier, top size one from the 80s, if you can find one in mint condition.  However, they are much heavier than the ones today.
Title: Re: Why do megaphones still sound terrible?
Post by: Alex Eisenhut on August 13, 2018, 12:55:59 am
Planning a protest, Dave?
Title: Re: Why do megaphones still sound terrible?
Post by: TheSteve on August 13, 2018, 01:03:45 am
Hmm, a new EEVblog product? Only problem is the name, using standard convention it would be the "micro"phone which doesn't really work for this product. Might have to just go with the "MicroMegaphone".
Title: Re: Why do megaphones still sound terrible?
Post by: Jr460 on August 13, 2018, 01:13:50 am
Isn't the idea to get the message as far as you can in the direction it is pointed?


If so, then reduce the bandwidth to what you really need so you can get more power out.  Second, make it as loud as you can, so compress, smash the &^&$ out of the signal so the average is much closer to the peaks.

So if with newer tech you can take an old design and make it sound better, you can take the old design and make it carry further with the same old limited bandwidth and compressed signal.

Let see what the police will buy.   Model A which sounds great and carries 100 yards, or model B with makes no claims abut quality but carries 150+ yards.

Someone else said it, and I know from building one over 30 years ago, feedback is a big problem.
Title: Re: Why do megaphones still sound terrible?
Post by: BrianHG on August 13, 2018, 01:29:57 am
I'm sure this 150 year old design sounds fine:
Title: Re: Why do megaphones still sound terrible?
Post by: BrianHG on August 13, 2018, 01:32:19 am
Pick your poison: Megaphone images (https://www.google.com/search?client=firefox-b&biw=1568&bih=869&tbm=isch&sa=1&ei=eN5wW47rIqu0jwTznZXoCg&q=megaphones&oq=megaphones&gs_l=img.12...0.0.0.197670.0.0.0.0.0.0.0.0..0.0....0...1c..64.img..0.0.0....0.gFOcv8WK4dc)
Title: Re: Why do megaphones still sound terrible?
Post by: xrunner on August 13, 2018, 02:13:50 am
You know what I mean, that horrible distorted scratchy narrow-band sound, to the point of being almost unintelligible?

Because it's a legacy sort of experience. They are expected to sound that way.  ;D
Title: Re: Why do megaphones still sound terrible?
Post by: David Hess on August 13, 2018, 02:26:35 am
I can understand limitations in the horn type arrangement, but why still so bad?

If you think the horn arrangement sounds bad, you should hear one without a horn at the same power level in the same form factor.

The folded double exponential horn arrangement acts as an impedance transformer so that for a given power, the driver displacement is *lower* while the pressure is higher.  Driver displacement is closely tied to distortion so the double exponential horn allows much higher power levels at a given level of distortion.

Base reflex speaker enclosures play the same game by directing the wave off the back of the driver to reinforce the wave from the front.  The lower displacement lowers distortion at lower frequencies and especially intermodulation distortion which is what makes crummy speakers sound "boomy" when they should not be.
Title: Re: Why do megaphones still sound terrible?
Post by: calexanian on August 13, 2018, 02:59:23 am
All reasons mentioned are correct. One big factor not mentioned is market size. I would assume there are many small PA systems sold to every one megaphone. With modern high power portable PA systems on the market there is no need for better quality. There are good quality megaphones for ship to ship and military/police use but they are typically very expensive and have high quality drivers and are tuned specifically for the intelligibility range which is not necessarily full speech range.

the problem is that this is how they are used.
https://youtu.be/pAoB7ey0L70 (https://youtu.be/pAoB7ey0L70)
Title: Re: Why do megaphones still sound terrible?
Post by: rsjsouza on August 13, 2018, 03:35:02 am
I've seen the inside of one at summer camp back in the early 80s.  It was a huge ceramic tweeter inside the center of that cone.  I also guess the amp was poorly biased class B to save on battery life.  This way, if the mic was left on, only when sound went through was there any current draw. 
I read in an electronics book a long time ago that early megaphones used a rudimentary Class S amplifier (similar operation to class D) to conserve power at the expense of quality. However, I have never seen one to confirm that.

For the insides, check a modern version using a TEA2025B, a Class AB stereo integrated IC. The quiescent current is quite high for a battery powered device (35mA) and it does not have a shutdown pin. It can provide 4.7W bridged but the product claims 40W.  :-DD

https://www.youtube.com/watch?v=hO9Mdt-IwGA (https://www.youtube.com/watch?v=hO9Mdt-IwGA)
Title: Re: Why do megaphones still sound terrible?
Post by: GregDunn on August 13, 2018, 04:34:00 am
Horn speakers have poor high frequency spread, so the treble sucks. If it's pointed directly at you, it's a bit better, but it still sucks.
Horn speakers have to be large to get good bass, so for hand-carried sizes the bass sucks.
Horn speakers are better than cone speakers at producing volume.

Something else not often realized is that a horn causes something called "wavefront steepening" which, at higher volumes, causes serious distortion of the waveform.  The horn is intended to change a high pressure low amplitude signal into a low pressure high amplitude one.  Since air is increasingly nonlinear at higher pressures, the higher the initial pressure the worse the asymmetry and hence the distortion.  It can actually transform a sine wave into a sawtooth-shaped wave by the time it reaches the mouth of the horn.

https://www.klippel.de/fileadmin/_migrated/content_uploads/Loudspeaker_Nonlinearities%E2%80%93Causes_Parameters_Symptoms_01.pdf (https://www.klippel.de/fileadmin/_migrated/content_uploads/Loudspeaker_Nonlinearities%E2%80%93Causes_Parameters_Symptoms_01.pdf)
Title: Re: Why do megaphones still sound terrible?
Post by: vk6zgo on August 13, 2018, 06:13:39 am
I've seen the inside of one at summer camp back in the early 80s.  It was a huge ceramic tweeter inside the center of that cone.  I also guess the amp was poorly biased class B to save on battery life.  This way, if the mic was left on, only when sound went through was there any current draw. 
I read in an electronics book a long time ago that early megaphones used a rudimentary Class S amplifier (similar operation to class D) to conserve power at the expense of quality. However, I have never seen one to confirm that.

For the insides, check a modern version using a TEA2025B, a Class AB stereo integrated IC. The quiescent current is quite high for a battery powered device (35mA) and it does not have a shutdown pin. It can provide 4.7W bridged but the product claims 40W.  :-DD

https://www.youtube.com/watch?v=hO9Mdt-IwGA (https://www.youtube.com/watch?v=hO9Mdt-IwGA)

Really early megaphones used a carbon microphone, a 6volt battery, & a loudspeaker.
One of the loudest megaphones I've heard was of this type.
From memory it was called a Marconi (or AWA) "Jericho" after the biblical walls.
Title: Re: Why do megaphones still sound terrible?
Post by: Gyro on August 13, 2018, 08:17:54 am
Just to give an idea of the sort of horn size and flare rate that you need to achieve undistorted reproduction down to 1kHz - mouth diameter is 12-15". Getting down to 200Hz would be much bigger! [EDIT: You'd probably want to mold it out of something very light and transparent so you could still see who you were shouting at  :)]

Megaphones achieve very high acoustic efficiency and directionality but at the expense of poor bandwidth and high distortion. The re-entrant (folded) horn reduces the length, but doesn't help with the diameter.
Title: Re: Why do megaphones still sound terrible?
Post by: The Soulman on August 13, 2018, 10:52:34 am
Nice pic Gyro, but that isn't a folded horn nor megaphone quality.  ;D

If Dave is looking for a megaphone I don't think the band-with limitation of a smaller one could be a problem for his voice.  :-DD

But anyhow, they are all crap because: cost, size, power consumption. (in that order).
Title: Re: Why do megaphones still sound terrible?
Post by: ebclr on August 13, 2018, 12:11:35 pm
I guess is the same reason for the sound on planes when crew talking, who also sounds terrible, Legacy bandwidth limitation to voice band only, who nobody takes care to expand in nowadays easy bandwidth for audio on the cheap.
Title: Re: Why do megaphones still sound terrible?
Post by: Kjelt on August 13, 2018, 12:24:17 pm
It's the speaker cable, definitely:
https://www.eevblog.com/forum/chat/i-finally-saved-all-my-pennies-and-got-a- (https://www.eevblog.com/forum/chat/i-finally-saved-all-my-pennies-and-got-a-)$10k-hdmi-cable-and-then-this-happens/
Title: Re: Why do megaphones still sound terrible?
Post by: BrianHG on August 13, 2018, 04:44:33 pm
I guess is the same reason for the sound on planes when crew talking, who also sounds terrible, Legacy bandwidth limitation to voice band only, who nobody takes care to expand in nowadays easy bandwidth for audio on the cheap.
Obsolete...  I've been on a modern plane and the audio is fantastic now, very loud and a full bandwidth, at least 50hz to 10khz, even bass enhanced, and it's free of distortion.
Title: Re: Why do megaphones still sound terrible?
Post by: IanMacdonald on August 13, 2018, 06:33:53 pm
Basically, transducer efficiency is inversely proportional to bandwidth. That is true for speakers as well as aerials.  Thus, fidelity and efficiency seldom come in the same package unless it's a big package with multiple radiating elements for different frequencies. To create a small handheld unit with a loud output from a limited power source, well, go figure what has to be left out. 

The transducer used is coupled to a horn, same shape  as a trumpet, which gives excellent coupling to the surrounding air but has a natural resonance, such that it will only handle a narrow range of frequencies.

Title: Re: Why do megaphones still sound terrible?
Post by: Emo on August 13, 2018, 07:39:26 pm
If we connect such a transducer to a proper amplifier and audio source we can differentiate between horn-transducer and possible non-linearities/distortion in the amplifier. Nowadays high efficient class-D amplifiers could be an easy solution. DSP's might even be used to correct for non-linearities in the horn system. So basically with some testing we might be able to improve the design substantially
Title: Re: Why do megaphones still sound terrible?
Post by: LaserSteve on August 14, 2018, 03:42:08 am
Because the units linked below are expensive compared to the installed base of 150$ cheapies... many of which still use a complementary pair with a 2N3055 as the output stage.

https://www.wattre.com/ (https://www.wattre.com/)

I worked  for them for about a year, long ago.. Very loud, and very directive if needed.

Also it is difficult to get good sound out of a 1/4th meter diameter radiator.

If you looked at a plot of frequency vs pattern for a classic megaphone, , you'd find a strong dependence on angle from the emitter face  for how the response sounds.  Bass spreads a lot, while mids are semi-directional, and highs are very directional but die off quickly.


Steve
Title: Re: Why do megaphones still sound terrible?
Post by: PointyOintment on August 25, 2018, 05:17:19 pm
Isn't the idea to get the message as far as you can in the direction it is pointed?

If so, then reduce the bandwidth to what you really need so you can get more power out.  Second, make it as loud as you can, so compress, smash the &^&$ out of the signal so the average is much closer to the peaks.

How about discretizing the frequencies in the audio, so that you run a spectrum analysis on the input signal, with (say) 200 Hz bins, and then reconstruct the signal for output with only one central frequency peak per bin? That would, I think, reduce the total bandwidth of the output sound by a lot, allowing more power at each of the output frequencies. Plus, free autotune.

Basically, transducer efficiency is inversely proportional to bandwidth. That is true for speakers as well as aerials.  Thus, fidelity and efficiency seldom come in the same package unless it's a big package with multiple radiating elements for different frequencies. To create a small handheld unit with a loud output from a limited power source, well, go figure what has to be left out.

The transducer used is coupled to a horn, same shape  as a trumpet, which gives excellent coupling to the surrounding air but has a natural resonance, such that it will only handle a narrow range of frequencies.

Would it be possible to engineer some grooves/ridges or something into the horn to make it an acoustic metamaterial that presents a different effective horn to each of the above-described different discrete frequencies? I'm imagining it would work somewhat like the coupled tubes of different sizes in this thesis [PDF] (http://www.microflown.com/files/media/library/books/thesis_eerden.pdf), where the each size of grooves/ridges would interact with one of the discrete frequencies, so each frequency would see its own optimal horn shape.