So I have been playing around with the following transmitter on a breadboard:

my Q1 is a PN3563. my C1 is 1nF. the electret microphone is replaced with a speaker and a 4.7uF capacitor in series. my R1 is 3.3K. my R2 is 220 ohms. my C4 is 1nF. and my supply is 6VDC from a power adapter.

The transmitter DOES work well when the antenna is connected to the collector or the emitter.

The problem I have is that I tried my radio link (from my transmitter to my superregen) on several different frequencies. My radio can receive the signal well, especially when it is outside the FM band. the problem is if I don't have the transmission frequency set to slightly higher than the FM band (say 109.something Mhz), then I cause interference to the TV and the TV is 10 meters away.

From all my tests, I went through channels 45 to 60 inclusive, and at least one of those channels had faint interference (enough to distort fine images, but not enough to distort the station) while my transmitter was on.

All my transmitting frequencies were definitely higher than 108Mhz.

so I was wondering if there is any way to narrow the bandwidth of the transmitter above, because the transmitter if set to < 108Mhz can flood my other commercial radio across several stations, almost overriding normal radio stations, and it is about 1 meter away.

 

I mentioned in your other thread and I showed a simulation of the waveform clipping at the bottom at the collector because the transistor has way too much base current (the value of R1 is much too low) so the transistor is saturated a lot of the time. The clipping produces harmonics that cause interference to high TV channels.
Try 47k to 220k for R1.

I found that a voltage divider of two resistors at the base worked best.

 

audioguru is right. a voltage divider at the base. the transistor should have a bias at the base, and not to create its own bias by forming a divider through the emitter with the emitter resistor that depends on the hfe which varies too much.
on the other hand, should you use the suggested electret mic without additonal series cap, maybe the bias voltage will be just right.

 

...has way too much base current (the value of R1 is much too low) so the transistor is saturated a lot of the time.

I never thought 3K was too low. I can see < 1K being too low.

Try 47k to 220k for R1.

Is there any math behind this that determines why 47K to 220K is the best?

I found that a voltage divider of two resistors at the base worked best.

I was considering that, but I'm afraid if I use that, I think will cut some of the audio, since the resistor (if added) will be in parallel with the capacitor and speaker. The capacitor and speaker are in series.

The reason why I want to use a low value is because I want to transmit a far distance on one frequency. Having clear reception for long distance is great as well.

 

Hi,

If you carefully read the schematic page, he explicitly states that if you opt for a higher supply voltage than 3V, then you must increase resistance on both R1 and R2. You should roughly be using a 10K for R1 and 680 Ohms for R2. Perhaps you already knew this and are compensating for the fact that you have a different transistor in there.

Steve

 

In all applications, I can understand that I increase several resistance values if I increase voltage, because of ohms law.

I'll try 10K for R1.

I don't understand why 680 ohms should be used for R2.
Doesn't a higher R2 lower the transmitter power?

 

Mike,
R2 limits the maximum current through L1/Q1.
When Q1 hits saturation, your transmitter starts "splattering" all over the frequency spectrum, causing interference. Look back at Audioguru's simulation he so kindly posted for you, and notice the upper trace on the simulated oscilloscope. That's what the output looks like on the collector. Ideally, that should be a pure sine wave - but since Q1 and/or L1 are saturating, it's very distorted. Increasing R2 will help a great deal to keep Q1 operating in the linear region, thus reducing the tendency to saturate.

 

I can't remember where Audioguru's simulation is.
In the meantime, I will just have to do some more research on transistors.

 

The PN3563 transistor has a huge variation of hFE of from 20 to 200. Each one will need to have is base bias resistors calculated for it.

A 2N3906 transistor was specified for the circuit because it works up to at least 300MHz and has an hFE variation of only from 100 to 300.

If the hFE is 200 and you want a max voltage of 2V across the 220 ohm emitter resistor then the max current is 9.1mA, the base current is 45uA and the base resistor should be (6V - 0.7v - 2V)/45uA= 73.3k ohms, not 3.3k ohms. A huge difference.

Changing the transistor to a 2N3904 that has a typical hFE of 200, changing the 3.3k resistor to 56k and adding 12k to ground at the base makes the transmitter work much better.
Its output has very little harmonics for low interference, its output voltage is 4.5 times higher and its output power is 20.25 times higher.

A 12k resistor across an 8 ohm microphone makes no difference. A 40 ohm resistor across an 8 ohm microphone makes a very slight difference.

The transmitter doesn't have a bandwidth too wide. Your commercial radio is cheap so it doesn't have a tuned input stage. So the transmitter is overloading its input which causes its input to be flooded. My Sony Walkman FM radio receives my FM transmitter all over the dial when the radio is close and is set to Distant. When the radio's input is attenuated by switching to Local then my transmitter is received normally on only one spot on the dial when it is close.

My home stereo FM tuner and my car radio are not overloaded when my FM transmitter is close.

 

How did you get 45uA for the base current?

--- edit: never mind. I divided emitter current by hFE.

 

you mentioned that the hFE of my transistor is between 20 and 200, a range of 180.
you mentioned that the hFE of your transistor is between 100 and 300, a range of 200.

Mine has a smaller range of hFE.

 

I think he means in terms of ratios, so, 20 to 200 is 1:10 and 100 to 300 is 1:3.

Steve

 

you mentioned that the hFE of my transistor is between 20 and 200, a range of 180.
you mentioned that the hFE of your transistor is between 100 and 300, a range of 200.

Mine has a smaller range of hFE.

Your transistor has an hFE range of 10 times!
One with a current gain of only 20 needs a huge base current.
One with a gain of 200 needs a tiny base current.
Each transistor will need its base current adjusted fot it.

My transistor has an hFE range of only 3 times.
The circuit can be designed for a typical transistor with a gain of 200 and will still work pretty well if a trans=istor has a gain of 100 or a gain of 300.