FM Circuit in Multisim

Thread Starter

aragon1971

Joined Apr 7, 2008
44
HI friends

i draw a circui in multiism for production of a FM signal but not working

than k you in advance

George
 

Attachments

DickCappels

Joined Aug 21, 2008
5,944
Your oscillator does not have the necessary feedback. This is often done by putting an emitter resistor in series with the emitter and a small capacitor from the collector to the emitter, like this:
1572951633548.png
Note that the base needs to be bypassed to ground.
 

Bordodynov

Joined May 20, 2015
2,420
You bypass the oscillating circuit with zero resistance of the voltage sources. The varicap must be energized via a resistor. For example, use a 100 kOhm resistor.
 

DickCappels

Joined Aug 21, 2008
5,944
@Bordodynov Good catch -the voltage sources V1 and V2 are shorting the collector to ground through the 15 uf capacitor. In Spice unless a resistance is specified, voltage sources are 0Ω.

Putting a 100k resistor in series is a good solution but that huge C2 15 uf capacitor would need to become much smaller so that the modulation input has some bandwidth.
 

Thread Starter

aragon1971

Joined Apr 7, 2008
44
Thank you
i put a scope in collector of transistor but the signal have constant frequnecy
i attach the multisim file

George
 

Attachments

Bordodynov

Joined May 20, 2015
2,420
And where's the 100 kOhm resistor I'm offering? I advise you to read textbooks and look for practical schemes with varicaps. Then you will understand what I mean.
 

DickCappels

Joined Aug 21, 2008
5,944
It's great that you got it to work!

In real life you will need a resistor and/or choke to keep the modulation source from hurting the operation of the oscillator.
 

DickCappels

Joined Aug 21, 2008
5,944
I didn't look closely at the waveform and only took your word for it. I would expect the L/C circuit to oscillate at about 1.6 MHz. All I see is 100 Hz which is your modulation. The RF oscillator is not oscillating.

At what frequency do you want this to oscillate? Right now it is around the top of the AM broadcast band in most regions -1.6 MHz.

The capacitor from the emitter to collector that I suggested (5 pf) is good for 60x that frequncey but around 1 Mhz the impedance is probably way too high. Either you need to make your tank circuit resonate around 100 MHz or change the cap (C2) to 300 pf. You also need to add that 100k resistor between the modulation source and the varactor. The larger capacitor(C2, 300pf or so) along with the 100 k resistor will cause a high frequency roll off of your modulation signal at about 5 kHz. You can adjust that after you get the RF oscillator working.

In real life the 15 uf capacitor would probably work but for these frequencies we don't use such large capacitors because the construction of such large capacitors makes them not suitable for high frequency use -too much stray inductance is the main issue. Ceramic capacitors are most common in RF circuits. When you get up to high power Mica capacitors are superior.

You are only going to be modulating the capacitance across the tank circuit by (probably) hundreds of fempto farads so you can safely couple your varactor with 10 pf or so. That would also increase modulation bandwidth a lot.

Also, I don't think Spice cares, but in real life you should have a bypass capacitor, something like .001 uf ceramic for 100 MHz, 0.1 uf at 1 MHz are good values.

I see that you are using a PNP transistor. That means that your varactor is forward biased. I didn't catch that before. You need to reverse your varactor so that it is reverse biased so it can act like a varactor instead of a simple diode. You will also have reverse V2.

You might also want to swap the values of R1 and R2 to lower the current. That might be important when using real-life parts.
 

DickCappels

Joined Aug 21, 2008
5,944
Put a 0.1 uf across the battery (its needed in real life so its good to get used to using and seeing it.

Let's use a known transistor, the 2N3904 which is a common NPN. For this reverse the battery so that the negative terminal is grounded and the collector circuit supply is positive +12V.

Referencing component labels in your schematic of post #12:
Q1 2N3904 (NPN not PNP)
C1 39 pf
C2 330 pf is ok.
C3 15 pf
R1 10k
R2 2k
R3 470
R4 100k is ok, gives your modulation input a 5 kHz high frequency roll off.

This should get you close to 10.6 MHz once you subtract the varactor's capacitance from C1..
 

Bordodynov

Joined May 20, 2015
2,420
I'll give you some more advice. It is necessary to put a capacitor, at least 1 nF, between the transistor base and the common wire. Then it will work both in Multisim and in life.
 

DickCappels

Joined Aug 21, 2008
5,944
He's right. I forgot to include that capacitor. The base must be at ground for AC. If it doesn't work after that maybe Spice gives your battery parameters like a real battery (that would be strange) in which case a similar capacitor should be placed across the battery.
 

Thread Starter

aragon1971

Joined Apr 7, 2008
44
thank you agaain for your reply ... and your time with the last change the outpout in the collector is a constant signal..
if you have multisim i can sent the ms file
1573116709025.png
 

DickCappels

Joined Aug 21, 2008
5,944
I don't have muitisim.

I must have been tired last night because I forgot to suggest that you change the L1 to 4.7 microhenries so it will oscillate near 10.6 MHz.

What frequency do you get on the collector?

Also when you vary the bias on the varactor do you see a change in frequency?
 
Top