If I were to increase the modulation frequency to say, 100 kHz, how would this affect the bandwidth?




*edit* - So by increasing the modulation frequency, I would be increasing the bandwidth. If I decrease modulation frequency then I will be decreasing bandwidth. For example,

Modulation freq of 1 kHz --> square wave has around 5 times the fundamental frequency of modulation, 5(1 kHz) = 5 kHz

fc = 1000 kHz
fm = 5 kHz

LSB = fc - fm = 995 kHz
USB = fc + fm = 1005 kHz

So now, we have data transmission only 10 kHz wide.

Does this sound right?

 

OK,for convenience,I used the "5x" relationship,as it is a generally used "rule of thumb",to get an idea of the bandwidth needed to reproduce a reasonable square wave.

Depending on the source,your data may still have higher frequency components,even if you reduce the fundamental frequency.
This is largely determined by the rise time of the square wave,which you can determine with your Oscilloscope.

Tr(secs) = 0.35/BW(in Hz).,hence bandwidth BW =0.35/Tr
(If Tr is in uSec,then BW will be in MHz,& vice versa)


Your Receive side may be critical as to rise time.
If you increase this,(reduce max freq components),you may have trouble decoding your data.

Even if you have achieved a reduction in bandwidth,it will all go for nought if you "cut carrier" --this will cause you to radiate harmonics of 1MHz across the bands---not a good thing!!

 

Thanks for your input, I really appreciate it.

How would my I be radiating 1 MHz harmonics across different bands since it is a sine wave (with ideally one frequency)? Would the stray radiation be due to the square wave data signal?

What would be a good solution to this problem? Some sort of filter?


Thanks again,

Brent

 

If you "cut carrier" it is because excessive amplitude of your modulating signal causes the carrier to disappear,rather than to just reduce to a very low level.
At the point where the carrier amplitude reduces to zero,the,sinewave is distorted,& will produce multiple harmonics of,in your,case, 1MHz.

Overmodulation can occur with any type of modulating signal,including sinewaves.
It is easiest to imagine in the time domain,as you can see the carrier being "cut".

Have a look at the Oscilloscope picture which you posted previously.
I'm not sure if the small higher frequency signal visible at the point where the display would normally be at its minimum,is 1MHz,or just an artifact of the 'scope's display.
You are certainly close to,if not at,carrier cutting.

The best way to avoid this problem is to reduce the amplitude of the modulating signal.

RF & modulation is a fairly complex field,& you can't just "try things" without a bit of background study.

 

It is a complex field and i'm hoping to take a RF engineering course before I graduate in May. I have rudimentary knowledge of modulation and signal processing(it was presented briefly in a Signals and Systems class I took). However, I never learned anything practical, so I think the best way to learn is to try to build a circuit and see how it goes. I really didn't know anything about sidebands and bandwidths until I started building this.

I think what I will do is try to reduce the base voltage going to the modulating transistor and take a look at it with the Oscope.

 

I found a better solution. By limiting the current into the collector of the modulation transistor, I can reduce the amount of modulation. I added a 1k resistor to the collector, and the simulations so far look promising.

 

I connected a power supply to the wrong terminal on my oscillator and it burned out. Progress will cease until I can find a new one...

 

Just an update:"I have decided to scrap this design and go with a completely redesigned transmitter using a colpitts crystal oscillator connected to a power amp with an impedance matching network. I ordered the parts a few days ago, so they should be here soon.