• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Connect FSK modulator with fm transmitter circuit?

nwar1994

Member
The output is on the antenna. I have a Kaito KA200 fm receiver. For the first oscilloscope image.
The output is after the coupling capacitor for the waveform generato for the second and third oscilloscope images.

The square wave is the waveform generator. It has a lot of noise when it is on and off. Also, the modified colpits oscillator doesn't have a stable frequency.

I'm starting to believe that the 2N3904 can't handle the high oscillations.
 

Attachments

RadioRon

Well-Known Member
I' ll spend some time studying these, but on first glance, I see that perhaps you haven't attached the waveform generator correctly. If it is the type that has a 50 ohm output, and you attach this as shown "waveformgenerator.png" (the "output" point) then you will be upsetting the DC bias to your transistor because there is a DC path from your circuit through the generator. If this is the case, you need to inject with a DC blocking capacitor in line with your signal. Something like a 0.1 uF ceramic may be fine. I also noticed that C8 may be much too large a value and is shorting your af to ground. This capacitor should be lower value, like 1000pF.
Can you please specifically describe where you measured "DCOFF" and "DCON"?
 

nwar1994

Member
The signal generator should be coming from C4, doesn't that block the DC input?

DCOFF and DCON is the output junction of C4,R1,C8,Q2 when the DC is on or DC is off.
 

RadioRon

Well-Known Member
OK. I understand now. Yes, if the signal generator is fed through C4 that should take care of blocking the DC. It would have been clearer if you had marked on your schematic where the generator was feeding into. I do not assume anything.

Perhaps you can verify that your scope was AC coupled when you recorded those traces? DCOFF shows the output of the generator, which looks fine. DCON shows the output of the generator superimposed on RF, which is to be expected if your oscillator is indeed oscillating. It makes sense that there would be some 85MHz carrier visible at the base of the transistor. So, I don't see anything wrong with DCON, other than I can't see any DC bias (because the scope input is likely ac coupled).
The antenna output scope trace looks ok to me, although I have no idea what Channel 2 of the scope is showing.
 

nwar1994

Member
Channel 1 is the probe, channel 2 is the output of waveform generator.

I am able to tune and confirm there is a signal coming from the transmitter, but I am not able to hear the proper audio. There is a large amount of static. The resonant frequency is not stable. That may be why? I also am testing on a breadboard.
 
Last edited:

RadioRon

Well-Known Member
I recommend switching the waveform generator to sine wave output, roughly 600Hz and turn the amplitude way way down to zero. Then, while listening to your transmitter signal on the FM receiver, slowly turn up the waveform generator output. It is possible that the audio is simply far too strong and is causing too much FM deviation and also excessive distortion. To find out, do the procedure above, trying to keep the audio input to the transmitter as low as possible and carefully listening to the result on the FM receiver and note whether the modulation is much clearer and there is less noise when modulation is very low.

If your modulation is too large, this could also make the carrier appear unstable. It is not a good idea to try and modulate the transmitter with a square wave as this will guarantee an excessive amount of deviation.

Also worth mentioning is that you need to be sure you are receiving your carrier and not a spurious sideband or other spurious signal. When you hold the FM receiver near your circuit and leave the waveform generator completely disconnected, your signal will cause all of the usual background hiss that you hear on an FM receiver (when receiving nothing) to disappear and you should hear absolutely nothing. We call this "full quieting" of the FM receiver. You can check if it is your signal causing the FM receiver to fully quiet by switching your transmitter on and off. Also, you will likely hear a tinny sound when receiving your transmitter carrier and when you hit the transmitter with your finger. The hit will cause your inductor to vibrate slightly which in turn FM modulates the carrier. This is an obvious thing that you will hear on the FM receiver. We call this effect "microphonics" because the transmitter circuit is actually acting a bit like a microphone.

Verify your carrier by tuning your variable capacitor and noting that the carrier moves to a different frequency on the FM dial.
 
Last edited:

RadioRon

Well-Known Member
One other thing to comes to mind is that this simple transmitter circuit is also easily modulated by noise on the power supply rail. To isolate this kind of noise, use a battery rather than an AC to DC power supply. Batteries are noiseless, whereas other power supplies can have a lot of 60Hz ripple, 120Hz ripple and high frequency ripple if using a switching regulator. They can have other noises too. So use a battery to check that.
 

RadioRon

Well-Known Member
Another point worth mentioning is that when you put a scope probe at the antenna output, or any point on the tank circuit, you are effectively adding some capacitance to the tank circuit and so altering the frequency of oscillation. So the frequency that you measure on the oscilloscope is probably different, perhaps lower, than the oscillation frequency when the probe is not attached. You can estimate the effect by checking the capacitance of the probe, which may be on the order of 5 to 15 pF. Calculate how this capacitance, in series with the 10 pF output capacitor, would affect the tank circuit if placed in parallel with C1.
 

nwar1994

Member
It was the probe. I was also able to quiet the fm transmitter. I checked that the frequency changed, and it responded when I touched it/turned it off. My issue is that the baseband signal is not being transmitted. There is no audio on the output
 

RadioRon

Well-Known Member
Well, its good to know that your RF oscillator is working ok. It shouldn't be too hard to figure out why no audio. It is certain that if there is audio voltage at the base of Q2 there will be FM modulation.
 

nwar1994

Member
You're right it works. I will connect the 555 timer now. I hope that the oscillatory current doesn't mess up the 555 timer
 

RadioRon

Well-Known Member
How much AF voltage did you need at the base of Q2 to make a normal volume in the FM receiver? That's a good number to know as it is your Modulation Gain.
As I mentioned, you should filter the square wave from the 555 to round it off as best you can because you shouldn't be feeding a square wave to an FM modulator, it will cause your bandwidth to be huge. You can vary C8 or the voltage divider resistance preceding it to do this.

Also, add some bypass capacitance (if not already there) directly on the Vcc of the 555 timer. User ceramic capacitors, maybe 0.01 uF.
 

nwar1994

Member
So I added some bypass capacitors. This helped remove the noise from the 555 timer. It is transmitted properly. Unfortunately the transistor is modulated and transmitting somehow even if I don't connect the C4 capacitor on the FM Transmitter schematic.

About 50 mA was good enough. Should i just use a first stage lowpass filter?
 
Last edited:

RadioRon

Well-Known Member
The modulation from the 555 will couple to the transmitter through unwanted paths depending on a few things. First of all, if you have built the circuit on a plug type breadboard (white plastic with the strips with tiny holes for pushing wires into), then you will have lots of parasitic capacitance and this can provide a path for the 555 output to get to your oscillator. You can test this theory by building the 555 oscillator on one board and building the HF oscillator stage on a different board and then connecting them together with an AF wire and a ground wire.

The next path is through the DC power supply. The older 555 circuit (the one that is not made of CMOS) is infamous for being noisy. Since it draws current from the power supply in sharp spikes, it is hard to provide enough bypass capacitance to keep it quiet. Sometimes a little bit of series resistance is needed in the Vcc line to help the bypass capacitor do its filtering job at higher frequencies. One way to test the idea that there is 555 noise getting through the Vcc line to the modulator is to use a separate battery for each stage, to completely isolate their power supplies. If this makes the problem go away, then the unwanted coupling is through the common power supply. Usually, when we use a more modern CMOS version of 555, this problem is much less of an issue, since the CMOS version requires much less current.

The 555 signal can be coupling to the oscillator through the ground line as well, for roughly the same reasons as it might couple through the Vcc line. The ground currents flowing from the 555 can be rich with spikes, and this current flowing also through the ground side of the HF oscillator can cause small voltage spikes to affect the oscillator. Isolating the two stages and feeding power in a star or hub-and-spokes grounding method may help. You have to make sure that the ground current from the 555 does not flow through the HF oscillator stage.

It would be helpful to see your circuit. Can you post a close-up photo showing exactly what your circuit looks like?

For the lowpass filter, just try a single order type (simple series R shunt C) for now and see if the results are good enough. Be aggressive with your choice of cutoff frequency, slow the edges down.

You wrote 50 mA. Did you mean to say 50 mV? And is this peak-to-peak or RMS?
 

nwar1994

Member
So you're saying plug in a resistor to the Vcc of the 555 timer?
I believe it is coupling through ground since I disconnected Vcc to 555 timer, but there was still a square wave output.

If it is coupled, it's useless to filter the output signal since the unfiltered signal is coupled. I also will be putting this onto a prototype board

Yes I meant 50 mV PP
 

Attachments

Last edited:

RadioRon

Well-Known Member
The idea of adding a resistor to the VCC of the timer is simply to create an RC low pass filter on the power supply line to help isolate the two stages. However, it may be premature to do this just yet. I think you need to separate the 555 circuit completely from the HF oscillator circuit, meaning that they are built on separate boards and they are powered by seperate batteries. It would be better if you built the HF oscillator on hand cut pcb (the "dead bug" method) example:
https://spectrum.ieee.org/geek-life/hands-on/with-the-dead-bug-method-hobbyists-can-break-through-the-highfrequency-barrier

It doesn't make sense that the 555 continues to oscillate without any Vcc. Be sure to disconnect the resistor that connects to the Arduino as well.

50mVpp seems like a good value.
 

nwar1994

Member
I tried to build the 555 timer on a separate breadboard. For some reason when connecting both, it created a lot of noise which is why i decided to build the 555 timer on the same board. I did not use separate power supplies as I was unsure of the outcome. Also, the lowest voltage for the 555 timer with a voltage divider was about 250-300 mV PP

If I discover it works on a separate supply, would that mean for the pcb I would need two separate batteries? What if the two batteries have different voltages due to external conditions?
 

RadioRon

Well-Known Member
Using two separate batteries is a temporary thing to help figure out what you need to do to fix the noise and excess coupling problems. Once you understand it all, you can likely go back to one battery.

Since you previously measured that 50mVpp was the right voltage for audio to input to the modulator, obviously you need to reduce the timer output further by changing the resistor values in the voltage divider. If you find that you have modulation even when the 555 output is disconnected from the modulator (or the shunt resistor of your voltage divider equals zero), then you still have unwanted coupling through other paths and need to fix that.

When you say it created a lot of noise, how and where did you observe this noise?
 

nwar1994

Member
Yes so having two separate power supplies fixed the issues. Would it suffice two have to separate battery supplies in the PCB circuit? The coupling capacitor should block the DC quiescent voltages from the power supplies
 

RadioRon

Well-Known Member
You can certainly choose to use two batteries all the time, if you like. This is a reliable fix to the problem. The blocking cap does block the DC so you don't have to worry about different DC voltages. It might be easier to do this than to fix the coupling problem, but that's up to you.
 

Latest threads

EE World Online Articles

Loading

 
Top