FM transmitter problem :(

[mstechca]

At this moment, I am ignoring low pass filters because I just want to get a clear blank signal out.

Since audioguru gave me an attenuation equation, I used it and arrived with the circuit below.

This transmitter doesn't work, yet I have been playing with the frequency of the receiver from 80Mhz to 150Mhz. In fact, I even tried this with a commercial FM radio!

 

[zachtheterrible]

Try changing the 270 ohm to 2k.

If that resistor from vcc to the base is 20k, the emitter should be 2k. if it is 10k: 1k. and so on.

 

[audioguru]

Reply deleted. I messed up my math. :cry:

 

[audioguru]

Sorry, but No, Zach.
MStechca doesn't say, but if the transistor is a 2N3904 then its DC gain is about 225, so if its emitter resistor is 2k then with a base current through the 20k resistor of a desired 0.365uA, its emitter current will be 82mA and its emitter voltage will try to be be 164V! Of course the emitter voltage can't be so high so the transistor will be saturated so hard that it might not even oscillate.

Even its 270 ohm emitter resistor 's value is high, causing an attenuation of the circuit's positive feedback of only 0.68. So if it was biased properly with a base resistor of about 100k instead of 20k, then it would produce square waves full of harmonics in its output. :shock:


MStechca,
The problem could be that C3, the 10uF audio frequency supply bypass capacitor is lousy at RF frequencies and is actually an inductor instead of a capacitor. A 0.001uF or 0.002uF ceramic disc cap with very short leads is needed at 100MHz. Having two bypass caps at the base of the transistor looks like a good RF supply bypass with them in series, but they are actually a voltage divider feeding back supply voltage fluctuations out-of-phase and therefore cancelling any tendency to oscillate. C1 doesn't make sense to bypass the base to the positive supply. I hope that C4 isn't an inductive at 100MHz polyester cap. It must be a ceramic disc.

The problem could also be the inductance of its coil should have a proven defined core (air), wire size, number of turns and their spacing and diameter instead of having a meaningless value of 0.1uH. :?:

Its 10pF to 25pF tuning capacitor doesn't have much range of adjustment when you add maybe 5pF or more of stray capacitance. Who knows at what frequency it is trying to oscillate. That's why I used a tiny 5-35pF trimmer cap with very short leads in my FM transmitter. :lol:

Maybe it is built on a breadboard with very high stray inductance and capacitance. Not a chance for a VHF circuit.

You are trying to make a VHF transmitter, but didn't you notice that it doesn't have an antenna? :?:

 

[mstechca]

I forgot to add the 6 inch antenna in my original circuit. I just whipped that circuit up on the PC in 10 seconds.

all the caps are ceramic disc with the exception of the 10uF. the 10uF is an electrolytic.

Why are you suggesting to increase the resistance? Increased resistance can reduce current.

The only way the -ve supply can actually reach the rest of the circuit if the capacitors were not present is through the emitter resistor. If this value is too high, the oscillator won't even have enough current to start.

 

[audioguru]

Why are you suggesting to increase the resistance? Increased resistance can reduce current.

If this value is too high, the oscillator won't even have enough current to start.

A transistor oscillator doesn't need much current to start, maybe an emitter current of only 10uA or less. The ratio of the base resistor and its emitter resistor must be correct for proper bias according to its current gain, and the ratio of its feedback capacitor and the emitter resistor must be correct for a proper amount of positive feedback. :lol:

Did you ever get it to work? :?:

 

[mstechca]

All the time before, I used 0 resistance between emitter and ground, and sometimes it worked. I was playing with the base resistor alot.

Why match the base and emitter resistor? If I pick say 100 ohms for the emitter resistor, that means you want me to have 100 ohms for the base resistor. I fail to understand why.

 

[audioguru]

The base resistor cannot be equal to the emitter resistor, they must have the proper ratio according to the hFE current gain of the transistor. Or else the transistor will be saturated or it will be cutoff. That is what biasing a transistor is all about.

I added "The ratio of" a couple of times in my previous post to make it clearer.

 

[panic mode]

if this is wired correctly, all components are ok and traces are kept reasonably short, i cannot see this not working (but it can be on wrong frequency so you cannot find it).

can you show picture of the unit? i'm interested to see coil...

also, how are you checking if this is oscillating?

i always put the circuit very close to good radio and slowly search for
silent spot. once you find one, gently kick/push the coil or whole pcb with something light such as pencil or straw. if the radio makes some noise, you got it - it works. just search a bit more till you find the right
frequency and then you get some decent range. Just for trial keep it close to radio. hope this helps...

 

[mstechca]

My inductor is a commercial 0.1uH inductor.

The only thing which I think may cause a change of oscillator frequency is the fact that my circuit is on a breadboard, however, I either place the next component wire at the socket closest to the location of the last component pin. In fact, I sometimes place two, or even three wire ends in the same socket.

 

[audioguru]

Your commercial inductor might have a high capacitance and therefore might not be capable of 100MHz. Why not make your own like everyone else?

Your breadboard has capacitance between rows of socket jacks and also inductance along every row. No wonder your transmitter doesn't work.

 

[panic mode]

audioguru is right, make coil and put everything on a PCB like everyone else.
breadboarding is great thing for trying things out but there is a HUGE
difference between flashing LED and oscillator running at 100HMz.

 

[mstechca]

a new breadboard???

I only like to do everything on a circuit board when I know I got the circuit working.

Rather than waste circuit boards along with a number of components on each failed attempt, is there something else I could use instead of a breadboard or a circuit board?

and could this also be the reason why I can't get TV stations in on my superregenerative receiver?

I'm doing everything on a breadboard now because it is the only way I can think of doing it without throwing components out quickly.

 

[audioguru]

I use Veroboard for all my circuits and they all work fine the first time.
I wouldn't use a breadboard for anything over 10kHz. No wonder your VHF circuits don't work.

For improvements to the circuits I remove the parts with my solder-sucker and make a new Veroboard layout. My mod4 FM transmitter uses the same parts as mod3, mod2, mod1 and the original. My newest "Veroboard" is made in Taiwan and is very cheap. :lol:

 

[audioguru]

Maybe you haven't seen Veroboard. It is perf board with printed tracks going in one direction. I cut the tracks with a drill-bit and use jumpers, resistors and capacitors in the right-angle direction.

 

[JimB]

If you are trying to build RF circuits, try a bit of "Ugly Construction".

Use a piece of circuit board as a base, connect all the 0v (ground)connections to this.

To mount the components, cut strips of the circuit boad material and divide them up into pads about 5mm x 5mm.

Glue the pads onto the base, I use epoxy, but you can use superglue (cyano-acrylate). Superglue is much quicker but tends to not stick verywell to the copper. Also it gives off some rather noxious fumes when heated with a soldering iron.

Solder the components to the pads and the base board and it will have a MUCH better chance of working that using a plug in breadboard.

Here is my interpretation of "Audiogugus Tx". It is actually running at 70Mhz (intentionally), the output power I measured at 2-3 mW, not much for 500mW DC input.

The RF output is a two turn link winding connected to the BNC connector so that I can connect it to the test equipment.

JimB

 

[audioguru]

Hi Jim Ugly,
Your wires are still too long for VHF, maybe that's why its efficiency is so low. :lol:

 

[JimB]

Which wires?

 

[audioguru]

All wires except the ground wires.

 

[JimB]

Even the red ones with the DC supply, yeh pull the other one!

JimB