[Chis] wanted an FM transmitter bug device packed in a mint tin, and that is just what he made. Featuring only 17 discrete parts, running off of a 3volt coin cell battery, and small enough to fit inside of a plastic mint box and still have plenty of room for the mints.
Using a simple design the audio enters the device through a electret microphone and hits a transistor for amplification, the signal is then passed to the oscillator section of the circuit which features an LC tank type design which generates the carrier frequency and mixes that with the signal for a range of about 100 feet indoors.
Each section is broken down into steps where it is thoroughly explained with animations, theory, and simulation, if you are wondering how a transistor, wire, and capacitors make an FM transmitter, or if you would like to just make the final project, schematics, pcb files, and assembly instructions are provided as well.
Join us after the break for a short video and be sure to check out the other radio transmitters we have featured as well.
[youtube=http://www.youtube.com/watch?v=alTPa7MLf9o&w=450]
Very nice. detailed, covers everything.
In the first schematic, there are 2 places where the wires cross, and there should be a dot showing that they’re connected.
I’ve also heard that one problem with this type of transmitter is that the center frequency can drift with temperature. It would be nice to see a simple circuit like this but with a quartz crystal or ceramic resonator substituted for the LC tank. It still would allow FM through capacitive loading, the same way as is done here.
otherwise, good job.
Simpler IS Better. What no Arduino?
I agree with the crystal, but if you don’t have one, simple oscillators can be loads of fun.
Good article.
There are typos in it – including one where it says a frequency is 5000 kHz instead of it actually being 5000 hz. Hopefully the author can fix them, then this will be a very good resource to learn a bit how these circuits work.
It would be nice if the author could explain why a BC547 transistor is necessary over the 2n2222. In America, bc547’s are harder to get than 2n2222’s.
@M
Showing connections with a dot is one valid way to make a schematic. There is another equally valid way to draw a schematic where connected wires simply cross. Unconnected wires which must cross in the schematic are then displayed by putting a little curvy spot in one of the wires at the point they cross.
Your way with the dot has probably become the more common method in recent years. Maybe because it’s the easier method to implement in a CAD program? His schematic isn’t necessarily wrong though.
Leif,
The old school method of lines crossing with a curve is not used in industry these days. I believe they stopped teaching that when I was at college in the late 90’s.
Industry practise is to have connected lines joined with a dot. BUT, it is also bad practice to have four wires meet at the same place as this looks like a cross over.
Instead we would have a staggered junction where no more than three wires meet at any node.
Altium designer software (again industry standard) only allows this method of staggered joins. It’s not possible to create a four way node.
@js
Maybe he isn’t in America? Or maybe some of the readers aren’t? His design works for me, I am in America but I have a lot more BC547s in my junkbox than 2n2222s. I bought a bag of 50 for about $1 on eBay because I wanted to build something a guy from Sweden posted online.
My point is someone who takes the time to share his design online shouldn’t have to explain himself, why he chose a part that is easier to obtain in one part of the world over another. Also, it doesn’t matter anyway when what he chose was dirt cheap and easy to obtain online.
@js: Comparing datasheets, BC547s have lower noise figures and higher gains than the 2222s.
@Leif: I do see your point. Sure, they shouldn’t have to explain why they chose one part over another. But, if you’re instructing someone in an article like this, a few simple words would help a lot of people who don’t normally have access to much more than mainstream parts and don’t know how to look up specs. BCs aren’t normally sold in the US except for odds and ends in most places.
I have used this design, or something similar to it years ago. The one I made had a really high audio gain. I could hear people whispering very clearly about 50 yards away without a parabolic reflector. The main problem was that when they got louder, the frequency dispersion went wider. The center frequency was about 100 MHz and would spread out to take up the whole band if someone started shouting into the mic.
@Josh
I do see how hard to find parts might not belong in some forums. That was a big problem in the old pre-internet days with a lot of the electronics magazines. Actually, often that was how the author got paid for the work… by offering the hard to find part for sale.
Besides that BC547s are easy to find online, the other half of my point was that I thought BC547s ARE mainstream. Maybe not in the US but if there was one I missed the notice that Hackaday or Pyro Electro are US only sites. I thought that in Europe were the common, mainstream hobbyist transistor that 2n2222s are here. Was I mistaken?
I also see BCs a lot more if I try to collect parts to reuse from junkbox items. I almost never see a 2n2222 in something which isn’t homemade.
I am from the US and always have been but I find a lot of articles/blog posts online from other parts of the world with projects I like to build. The UK and New Zealand are pretty big for this. Hackerspaces as I understand the story started in Germany. This makes me thing there is a lot more to the maker community than just the US.
Any Hack that “works” is worth looking at.
Locales etc does influence the conventions for both documentation and parts. Same applies to Freq assigns and power V&F norms. These differences are often a Very Good Thing. As speaking in a hypothetical case: A FM BC Band xmitter for a country that covers a range outside another Locale’s assigns can become inherently “noncontending” for an open freq in countries where that’s possible.
YMMV and check with resources to assure you’re not operating on a freq not suitable for that mode/powerin your Locale of course.
As for spec of parts. Depends on if it’s a “parts on hand” day or if ordering stuff is ok. One of the most underharvested sources of Semiconductors and RF useful bits is CFL ballasts. I’ve been polishing some prototypes and let’s leave it as- they’re worth exploring for RF bits.
@Chris: lowering voltage to powered electret mikes has tamed such over deviation issues. A few series diodes or an old trimpod filched from “somewhere” has been useful as the gain Vs Deviation points are not linear for circuits like yours.
@Oren: I suppose you’re talking to me. The circuit I had was pretty much the same thing, but operated on 2 AAs instead of a coin cell. I found out that the mic was a special type (it’s been 20 years, so I can’t remember what) with high sensitivity. When I got it, I just thought it was a regular one. When I got the internet, I was able a ton more info on it than the old ECG and NTE catalogs had.
It can be quite frustrating when a “parts on hand” project turns into an on order project. This many times has led to component substitution. In my experience, most 2n parts have a nearly equal BC part, though the pinout may be slightly different. I may just build one of these with altered RC values to use as a fox. (Yes, I’ll run it licensed). Might be even better hidden in one of these http://cyberoptix.com/radio.php
Yep- was a bit distracted by house guests etc. the 20 years bit is why our keeping lab books is a VERY Good Idea. I’ve often regretted not having done so and having to reverse engineer stuff I’ve Hacked together THIRTY years ago has been exasperating.
As in ” Oh Frak, I made this- Thirty years ago, why can’t I recall the color codes, Wh/Or-Wh/Bk were the POTS but WtF are the 4 identical solid copper lamp cord wires FOR?”
That was me fixing my phone ring>lamp flasher. The 4 coppers were ac mains from wall plug to an extension cord end. That had been snipped off when reclaiming it from my old shop’s bench-20 years ago.
Documentation and Lab Notes would have been easier than unwrapping the half roll of tape from the relay base- and retaping it.
Learn from my mistakes. Keep a Lab Book- Use smaller books and ARCHIVE the originals after copying. Copying to some Blogging site if it’s stuff you;re CC or Open Sourcing! Back to the “Dead Tree” Lab Books- Date/Initial pages frequently.. Consider SCANNING it’s pages frequently and emailing to yourself. You never will wish you had NOT saved a page of such notes :>
@M central frequency drift caused by thermal effects is small problem for such configuration compared to drift caused by battery voltage drop or capacitive loading of antenna ie hand extending to reach the frost mint. The later could be partially corrected by adding RF amplification stage to separate LC circuit form antenna.
Don’t even think of using a digital tuner to receive it. A good dial tuned car radio is best.