No RF gear here?

you can never be sure that an exact copy will work first time.

That's true with just about everything from audio amps to automobile racing engines......The probability of success does improve by sticking close to a proven design, especially if it has been built by others successfully.

I do agree that RF has its own degrees of difficulty and verifying completely legal performance of a transmitter requires at least a spectrum analyzer and a competent operator of said analyzer. I also built my own 40 meter ham transmitter with an 807 and a surplus crystal when I was about 14 years old. It worked, and put out enough power to light up a 40 watt bulb being used as a dummy load. Back then you had to visit the FCC office to take the Novice ham test. My father promised to take me, but that never happened, so the TX got robbed for parts and tossed.

With FM broadcast band you are restricted to around 10mw of power......rule is around 5-10mw power and you're safe from the FCC.

Not always the case. Back in the late 60's some enterprising students at the University of Miami (Florida) connected an FM wireless transmitter up to a small wire antenna in the top floor of a dorm room on campus and went on the air as "Radio Sunshine." The legal low powered transmitter covered most of the campus, but spilled over to the retirement community surrounding the school. It seems that their "hippie views" were not well received by the elderly listeners and complaints poured into the FCC. A visit ensued, in fact several, but Sunshine's operators were not breaking any laws. The school tried to shut down "Sunshine" so it started moving around, yet kept up random operations.

Things were about to get ugly since one of the Miami TV stations did a news segment on the "radio pirates" on campus. That news broadcast fixed the problem. Someone donated a 10 watt Gates FM broadcast exciter which was FCC type accepted for use as a stand alone low power mono FM station. The FCC fast tracked a broadcast license for the school, and WVUM 90.5 FM (Voice of the University of Miami) was born. With 10 watts, I could receive them from my house about 3 miles away. The school began offering radio broadcasting and other related courses. WVUM is still on the air, and a few years ago finally got a new TX that had far more power and stereo.

wvum | UM School of Communication
 
There is one area of RF that is (still) directly of interest to audio - wireless microphones. Commercial ones fall mainly into two categories - somewhat affordable and very unreliable, or prohibitively expensive and acceptably reliable. I've tried the former, and been burned (even my $150 AKG WMS 40 Mini failed within 2 years). I can't afford the latter (typically upwards of $500 for a single microphone).

I've also tried things like inexpensive Bluetooth transmitters and receivers of the sort you find on Amazon or Ebay; they turn out to be useless for live music, because latencies are in the tens to hundreds of milliseconds range. You sing or play now, and hear the sound emerge from the speakers a tenth of a second later. That makes it impossible to sing or play well.

I've looked for off-the-shelf RF transmitter/receiver modules to do what I want, but the ones I turn up all seem to be designed to transmit digital data, so they can't be simply interfaced with an analog microphone signal.

If anyone has any suggestions on a reasonably easy way to put together a legal, reasonably reliable, short-range wireless microphone RF link, of good enough quality for live vocals, I'd love to hear them. Plain old FM may be entirely adequate quality; if digital latency is involved, it needs to be under 10 mS, preferably under 5 mS.

When I was about twelve years old, I used to confuse my dog by talking to him via my dad's AM radio, and a one-transistor AM transmitter I threw together with a handful of junk-box parts. What I'm hoping for now is a better quality version of that.

-Gnobuddy
 
I turn up all seem to be designed to transmit digital data.........if digital latency is involved, it needs to be under 10 mS, preferably under 5 mS.

I have tried a couple of the cheap modules found on Amazon and also found the latency to be unacceptable. Not only is it audible, it is not always a constant time delay either. I am experimenting with wireless MIDI.

It should be relatively easy to do a decent FM transmitter with latency in the microsecond range. The main issues are interference and legality. The 915 MHz ISM frequency band (902 to 928 MHz) is allocated for almost anything in North America including Canada. There are some restrictions, but zero enforcement. There are tons of cheap Chinese transmitters in everything from cordless phones and wireless headphones to wireless video cameras (baby monitors). It is also legal for a licensed amateur radio operator to fire off 1.5 Killowatts in that band, and I have bounced a signal off the moon and off the ISS using about 600 watts. There are strong cellular and paging transmitters directly adjacent to that band.

For short ranges it may be possible to design a transmitter with 100 milliwatts or so, and a very good receiver to be capable of ignoring nearby cell towers. There would have to be several frequency choices in case there is a cordless phone on your first choice.

The uber expensive wireless microphones transmit on unused TV channels. The FCC allows this under certain restrictions, and only big companies with the required RF testing resources can play. Since unused TV channels theoretically have zero RF in them, range and interference issues should not exist. The frequency re-farming that occurred with the digital TV switchover in the US made TV frequencies more crowded, and lots of expensive microphones obsolete since they only had 4 frequency choices.

2.4 GHz is another possibility, but things get a bit more complicated as the frequency goes up.
 
The 915 MHz ISM frequency band (902 to 928 MHz) is allocated for almost anything in North America including Canada.
Thanks for the hints!

I found some Linx modules in that band, and they're designed to accept analog inputs. But they're "Not recommended for new designs", and Digikey doesn't stock them any more:
https://linxtechnologies.com/wp/product/es-series-rf-transmitter-receiver-modules/

So no joy there, probably.
...frequency re-farming... made... lots of expensive microphones obsolete
Judging from stuff I read all over the Internet at the time, that made a lot of musicians and sound guys very upset. Quite understandably so.

-Gnobuddy
 
Gnobuddy said:
If anyone has any suggestions on a reasonably easy way to put together a legal, reasonably reliable, short-range wireless microphone RF link, of good enough quality for live vocals, I'd love to hear them.
If it was possible, the internet would be full of designs. Making RF circuits which are small, low power and remain reliable when in the hands of a musician is a non-trivial problem. This is why the commercial offerings are expensive. In the UK you need type-approved equipment everywhere apart from amateur bands, and there you can't play music or do broadcasting, so a legal DIY route simply does not exist here.

I have used Sennheiser UHF wireless mikes in a church setting, but even in that fairly safe environment they still got broken within a few years. Design flaws were part of the problem, such as relying on solder joints to provide mechanical strength. However, when intact they worked very well - diversity reception was a big help.
 
My journey into electronics started with RF projects first when I was very young. In the beginning, I got into building transistor radio (receive only) while my father was building audio tube amps. Back then, there was no cell phones but only a rotary pulse dialing landline phone. Not too long after that, we (me and some other electronics friends) wonder if we can keep in touch by transmitting. We got into 27 Mc briefly. Eventually got my HAM license. I like homebrew RF stuffs putting most my time on antennas....

Fast forward to now residing in FL, found out this is not a very HAM friendly region. With communication more readily available, I faded out of RF into AF. That's when I jointed diyaudio.com.

Building RF stuffs is different than AF. Depending on frequency you working on, a short piece of wire can be an antenna, transmission line or something else that can throw you off.

I never like wireless mic. They are not very HiFi. You have to modulate the RF signal with the AF signal so it can be transmitted wirelessly. Then on the receiving end, you demodulate it back to AF. Delay and distortion is inevitable. I do involve in live sound and I always prefer wired connection although many dislike the clumsiness. I prefer sound quality over convenient.

AC
 
Making RF circuits which are small, low power and remain reliable when in the hands of a musician is a non-trivial problem.
Agreed. At the frequencies currently legally usable, I wouldn't dream of trying to build entirely from scratch. I was hoping to find RF modules that would accept an audio signal.

I did find a few, but all the current offerings are designed to accept a digital audio signal, and many of these digital systems will have unacceptable latency for my intended purposes.

Given the vast number of commercial products that cost much less than $500, and still manage to transmit and receive audio, it doesn't seem like an impossible problem to me. Radio has been around for well over a century now, surely we can put together an affordable short-range wireless audio link with 60 dB SNR and 100Hz - 15kHz bandwidth?

I suspect the $500 per microphone price has a lot to do with the fact that we're talking about a specialty item, with somewhat limited competition in the marketplace.

However, when intact they worked very well - diversity reception was a big help.
My AGK Mini worked quite well, just long enough to get past the warranty period, then died, with no visible indication as to why it died. I have no RF test gear, and didn't even attempt a post-mortem.

At the other end of the spectrum, I recently attended the Vancouver Mini Maker Faire. The large building contained a number of exhibits featuring Tesla coils, Jacob's Ladders, and random high-voltage fluorescent or neon lighting. In that hostile RF environment, not a single wireless microphone worked, as far as I could tell. (Several exhibitors had brought them.)

-Gnobuddy
 
I never like wireless mic.
You have plenty of company. The majority of musicians dislike, or distrust, them.

They are not very HiFi.
Not compared to CD quality digital audio, certainly. But the weakest link in the live audio chain - by far - is usually the utterly horrible-sounding Shure SM58 microphones that the majority of musicians insist on using!

Delay and distortion is inevitable.
Digital systems often have very noticeable latency, even tens or hundreds of milliseconds. This is not inherent to an RF link, however, but rather to the nature of the digital communication protocol being used. The speed of light (RF) is incredibly high, and at 2.4 GHz, there is no inherent latency longer than nanoseconds.

But when you start to send multiple audio channels over one audio link, and add digital data buffering, and then add two-way communication, some sort of data protocol that negotiates handshaking, data transmission and re-transmission, and so on, and so forth, then you can end up with these unacceptably long latencies.

Keep in mind that sound takes about a millisecond to travel a foot through air. Musicians routinely perform with guitar amps or vocal monitors that are five to ten feet from their ears. This means a delay - latency - of five to ten milliseconds is entirely acceptable to (and not even noticed by) most musicians.

I do involve in live sound and I always prefer wired connection although many dislike the clumsiness. I prefer sound quality over convenient.

AC
If I had a dollar for every microphone cable I've uncoiled, plugged in, then later unplugged and re-coiled, only to have to do it again, and again, and again, I'd be moderately wealthy. :D

-Gnobuddy
 
Wideband FM provides acceptably low noise and distortion. If it is good enough for broadcasting the whole show then it ought to be good enough for one mike channel. Some devices include companding, to make the best use of the dynamic range.

I was happy with the Sennheiser offerings, apart from the need for repairs to solder joints from time to time. This was speech PA, so intelligibility was the main aim. We used two tie mikes (with belt clip transmitters) and one handheld. The main problem was trying (but mainly failing) to teach people that a handheld mike is not supposed to be used in the 'ice cream cone' position - I told them that if they could lick it then it is too close to their mouth.
 
I suspect the $500 per microphone price has a lot to do with the fact that we're talking about a specialty item, with somewhat limited competition in the marketplace.

In terms of consumer electronics, you should see the cost of a mid range hearing aid. Makes a $500 price tag look very cheap.

So what do the TV broadcasters use for live panel discussions at venues that don't support booms? I see tie clip types, rarely a desk/stand mounted mic.
Perhaps wireless digital transmission, yet seemingly no latency issues with video sync or house PA.
Perhaps a heap of signal processing done to deal with this.... and top shelf equipment no band or pleb could afford.
 
I was just throwing my thoughts in a lounge settings LOL. Are we back to technical discussion or what ? Rather than talking about SM58, I'd have beer in my hands. LOL

You have plenty of company. The majority of musicians dislike, or distrust, them.


Not compared to CD quality digital audio, certainly. But the weakest link in the live audio chain - by far - is usually the utterly horrible-sounding Shure SM58 microphones that the majority of musicians insist on using!


Digital systems often have very noticeable latency, even tens or hundreds of milliseconds. This is not inherent to an RF link, however, but rather to the nature of the digital communication protocol being used. The speed of light (RF) is incredibly high, and at 2.4 GHz, there is no inherent latency longer than nanoseconds.

But when you start to send multiple audio channels over one audio link, and add digital data buffering, and then add two-way communication, some sort of data protocol that negotiates handshaking, data transmission and re-transmission, and so on, and so forth, then you can end up with these unacceptably long latencies.

Keep in mind that sound takes about a millisecond to travel a foot through air. Musicians routinely perform with guitar amps or vocal monitors that are five to ten feet from their ears. This means a delay - latency - of five to ten milliseconds is entirely acceptable to (and not even noticed by) most musicians.


If I had a dollar for every microphone cable I've uncoiled, plugged in, then later unplugged and re-coiled, only to have to do it again, and again, and again, I'd be moderately wealthy. :D

-Gnobuddy
 
Wideband FM provides acceptably low noise and distortion.
That's what was on my mind as well. Surely plain old analogue FM should be good enough for live music performance.

In the early 2000's I messed about with radio-control model aircraft for a while. At the budget end of the price spectrum, you could buy a 72 MHz FM RC receiver for $20 (USD), and those cheap little receivers worked well enough most of the time.

Latency and reliability are pretty important to RC flight - when you're flying your model upside down 15 feet off the ground at 50 mph, a short glitch or interruption in the RF link is all it takes to destroy the model.

(RC control is a very low-bandwidth affair, though, and that probably helped the receivers to have better noise immunity.)

I told them that if they could lick it then it is too close to their mouth.
This is another side-effect the SM58 is largely responsible for: those hoary old mics are so insensitive that singers and speakers learn to "get right up on it", and then yell loudly.

I recently converted a long-time SM58 user to a vastly better hand-held condenser microphone, after she heard how much better she sounded through it. But now she's having to learn to back off about six inches from the mic when she sings, and also that she no longer has to belt out everything. :)

-Gnobuddy
 
Gnobuddy said:
I've looked for off-the-shelf RF transmitter/receiver modules to do what I want, but the ones I turn up all seem to be designed to transmit digital data, so they can't be simply interfaced with an analog microphone signal.

If anyone has any suggestions on a reasonably easy way to put together a legal, reasonably reliable, short-range wireless microphone RF link, of good enough quality for live vocals, I'd love to hear them. Plain old FM may be entirely adequate quality .....

Motorola made a chip ( now obsolete ) for early LF FM cordless telephones, the MC1376, in an 8 pin DIP package. Someone, somewhere, probably has a stash of them. They were common enough that Radio Shack sold them for a while.

It has a VCO tuned with an LC circuit and can run between 1.4 and 14 MHz; the VCO is directly modulated by audio to produce the deviation; a transistor is on the chip and good for about 25-30 mw output.

There may be some Part 15 bands where you could use this directly, otherwise you would have to use doublers or triplers to get to the FM broadcast band. The dicey thing about frequency multiplication to the FM band is that if you do it wrong, you can wind up with spurs, or even the main signal, in the aeronautical band and this will bring you unwanted attention in a big way, in a huge hurry, if it is detected.

You can't rely on the LC circuit at the output of the multiplier ( or, I don't ) - you still have to have some way of knowing you are tuning on the correct harmonic. I don't have a spectrum analyzer and have always used a grid dip / wavemeter to accomplish this, as hams did for decades. I still have a homebrew two meter, frequency multiplier type, transmitter on the air. Used it last night.

I have the datasheet for this chip that I can scan. I have some, but never got around to using them.

Probably not a beginner DIY project, but certainly doable. Or just use it at LF in whatever band the old cordless telephones used. I think they were either right below, or right above, 160 meters.

Win W5JAG
 
RF is a whole new ball game that requires at least one or two notches higher education than most audio hobbyist have. There are plenty of people here that know all the ins and outs of cascodes, current mirrors and Zehner diodes and may have heard of H parameter but when it comes to RF a totally different set of parameter become relevant such as scattering and admittance parameters. Most audio hobbyist never heard of these.
By many RF is also referred sometimes as black magic, when they realize in some industry RF engineers work with signals that are well below the noise floor. There is no measurement equipment in existance today that can measure such signals except the object that these signals originated from.
That makes this very academic and very difficult to work with and most people that realize what headache this can be leave this field for something more logical like digital electronics.
I don't blame them.
 
when it comes to RF a totally different set of parameter become relevant such as scattering and admittance parameters.
It's all about where in the enormous range of RF frequencies you happen to be working. Before I was a teenager, I was successfully building AM and shortwave radio receivers, and didn't experience any more trouble with them than building audio circuits. Okay, trying to align IF transformers without a proper RF signal generator was a bit of a pain in the rump, but some patience and careful listening was enough to get a good-enough result.

At that age, I didn't know much theory beyond Ohm's law. But building 500kHz - 7.3 MHz stuff didn't really need you to understand RF-specific esoterica. I just kept wires short and put some thought into layout.

That was then. Nowadays, I've known Maxwell's equations for a long time, and how to use them. The mathematics of complex hybrid parameters or textbook transmission line theory isn't a problem.

But none of that gives me the ability to sit down and design, say, a 400 MHz transceiver. I wouldn't even know where to begin with something like that!

in some industry RF engineers work with signals that are well below the noise floor.
Not only in RF. You can do that with AF and even DC signals, too. Back in college I've used lock-in amplifiers to extract weak signals with frequencies in the audio band, signals otherwise buried in noise.

...most people...leave this field for something more logical like digital electronics.
For me, digital electronics hardware is so logical that most of the fun of understanding it is missing. :D

Writing software can be more fun, because it's not quite as logical as AND and OR gates, and there are usually multiple ways (algorithms) to tackle the same problem.

But trying to get good-sounding guitar distortion out of a handful of glowing glass bottles, by tweaking just a few capacitors and resistors, that is completely illogical. Your left-brain cannot easily solve a puzzle like that, the creative and subjective right-brain has to get involved too, and therefore, it can be a lot more fun!

-Gnobuddy
 
kct said:
. . .when it comes to RF a totally different set of parameter become relevant such as scattering and admittance parameters. Most audio hobbyist never heard of these.
Most radio hobbyists have never heard of these either. RF is hard, harder than audio, but let us not make out that it is too hard. I was successfully tinkering with RF from my mid teens, although it was only later that I realised just how little I knew early on. I have DIYed an FM tuner and a VHF transmitter, but no scattering parameters passed through my calculator - lots of impedances did, though.
 
For me, digital electronics hardware is so logical that most of the fun of understanding it is missing. :D
The "interesting" thing about digital is that at high speed it is essentially RF. A high speed signal flows through a trace and the return current flows through the ground plane, not as a diffuse current in the general direction back as it does for DC, but it flows back directly under the signal trace. And no, I don't know why it does that. I also don't know how a length of wire in air or a vacuum has some finite impedance to ground at RF frequencies.

This is a now-classic book on digital - the "black magic" subtitle is indicative of the lack of understanding in the area, and the authors endeavor to fix that:
High Speed Digital Design: A Handbook of Black Magic: Howard Johnson, Martin Graham: 9780133957242: Amazon.com: Books
 
The "interesting" thing about digital is that at high speed it is essentially RF.
The PCBs in the 'phones and PCs we never give a second thought to these days are almost miracles of high-frequency design.

Look back at pictures of the old Cray supercomputers - their cylindrical shape was dictated by the speed of light (slowed down by the properties of cables and dielectrics and striplines). The fastest thing in the universe was already inconveniently slow for the worlds fastest computers - and that was decades ago. :eek:

The speed of light (and therefore RF signals in PCB striplines) hasn't increased since then, but electronics has become much, much smaller, which has helped enormously, by shortening the distances that signals have to travel.

Still, it boggles my mind that you can plunk down $100 or so, and walk off with a gigahertz-clocked computer the size of a large USB thumb drive.

Neither Cray supercomputers nor USB-stick computers come anywhere near being feasible DIY electronics projects, needless to say.

-Gnobuddy
 

Attachments

  • cray1-large.jpg
    cray1-large.jpg
    123.5 KB · Views: 435
  • Intel_Compute_Stick.jpg
    Intel_Compute_Stick.jpg
    118.1 KB · Views: 424
The PCBs in the 'phones and PCs we never give a second thought to these days are almost miracles of high-frequency design.

In my other life (RF engineer at Motorola) I laid out several of those PC boards. Some were iDEN phones (Nextel, Telus, and others), some were two way radios for police, fire, and military applications. Look carefully at the tiny traces on a PC motherboard, especially around the processor and memory modules. You will see little zig zags in some of them such that all the bus signals are equal length. When you are clocking several signals at once across a board, you want them all to arrive at their destination in phase (at the same time).

It's all about where in the enormous range of RF frequencies you happen to be working.......But none of that gives me the ability to sit down and design, say, a 400 MHz transceiver. I wouldn't even know where to begin with something like that!

When I started at Motorola in 1973 the highest frequency two way radio available worked in the 450 to 512 MHz frequency range. Looking back now, it's a miracle that some of the designs out there even worked, especially the old vacuum tube stuff.

Cellular wasn't even a dream then, and the forward thinkers began looking at 800 MHz for two way radio in the late 70's, but the research didn't go very far because the frequencies between 470 and 870 MHz belonged to UHF TV in the USA. It would take FCC action to steal channels 70 through 83 from TV to pave the way for cellular phones, and 800 MHz two way radio. I was one of the engineers working on the first generation 800 MHz stuff in the early 80's and we all said that 800 was easier that the UHF stuff around 400 MHz.

Today there are all sorts of "building block" chips out there that have 50 ohm inputs and outputs and work over wide frequency ranges. It is possible to build a two way radio just by "connecting the blocks." Companies like Mini Circuits made this possible. They make mixers, attenuators, and RF gain blocks that can be connected together like Lego blocks. A lot of stuff in their catalog is actually made by some of the big names in RF like HP (or whatever name their chip group goes by now, Avago?), RFMD (now Qorvo), NEC and others. Some of it is available in little metal boxes with connectors so that you could build something without even laying out a board, although it would be large and expensive.

During the last 10 years at Motorola I worked in a research group that designed state of the art chips for two way radios. Those chips need to be tested, and EValuation Boards need to be designed. Some of these EVB's were sections of a two way radio, or even fully functional transceivers.

I was one of the people designing the EVB's and test systems. The MiniCircuits, DigiKey, Mouser, RFparts, and Richardson web sites, and a PCB layout tool was often the only design tools I needed to design these things......OK, sometimes I had to fire up ADS, a complete RF design tool from Agilent(now Keysight), but I didn't use it often. It does have a cool RF optimizer. You draw up an RF schematic, set the design goals, turn the optimizer loose, then go home for the evening. Sometimes you would come in the next day and find a completely optimized design ready for layout. Sometimes you would get a nastygram from the simulator stating that it couldn't converge....change the circuit, change the goals, or start over.

I once had an argument with a professor at college when I was working on my masters degree. She told me that I would have to learn this linear algebra crap to become an RF engineer. I explained to her that I was an RF engineer (designing cell phones at the time) and the only math that I used was the calculator in Windows. The hard stuff is built into ADS.