Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Tuning a series LC circuit, different ways?

Status
Not open for further replies.
neazoi

neazoi

Advanced Member level 6
Advanced Member level 6
Joined
Jan 5, 2008
Messages
4,160
Helped
13
Reputation
26
Reaction score
15
Trophy points
1,318
Location
Greece
Activity points
37,224
Hello I have a series L/C circuit connected to a vfo. I am trying to figure out different ways to tune it. Right now I am thinking the next ones:

1. L stationary, C mechanically tuned (variable capacitor)
2. L mechanically tuned (inserting or removing cores), C stationary
3. L stationary, C varicap
4. L electrically tuned (varying magnetization of the core by a closely placed electromagnet), C stationary
5. Combinations of all the above


Any other methods that you can think of?
 

If you vary L you will affect the bandwidth of the tank as well. That is why it is preferred to vary C to tune a LC tank.
 
Last edited:
E-design said:
If you vary L you will affect the bandwidth of the tank as well. That is why it is preferred to vary C to tune a LC tank.
Click to expand...
Do you refer to the Q of the tuned circuit?
 

neazoi said:
Hello I have a series L/C circuit connected to a vfo. I am trying to figure out different ways to tune it. Right now I am thinking the next ones:

1. L stationary, C mechanically tuned (variable capacitor)
2. L mechanically tuned (inserting or removing cores), C stationary
3. L stationary, C varicap
4. L electrically tuned (varying magnetization of the core by a closely placed electromagnet), C stationary
5. Combinations of all the above


Any other methods that you can think of?
Click to expand...

A problem of method 1. is the microphonism of variable capacitors when subject to movement or vibrations.
For this reason, method 2. was used in old car radios.
Method 3 is used in modern radios.

Another method is using a variometer, i.e. a series combination of 2 coupled inductors with mechanically variable mutual inductance, like this:

DSC_5035.jpg


Ltot = L1 + L2 + 2*M where M=k*sqrt(L1*L2) and k can range between -1 and 1 .

Regards

Z
 
  • Like
Reactions: neazoi

    neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
Yes, a variometer, well known from the crystal radio era.
I like the varicap method but If you listen to a varicap oscillator signal on an SSB radio it shows tone fluctuations, possibly from fluctuations in rf voltage. I have tried varicaps in 2-3 different circuits and they all have these fluctuations. The "real" LC circuit is much more tone clear :(
I was hoping to find a way to avoid these harder to find and expensive variable capacitors with reduction drives
 

Yet another possibility is the reactance modulator: a transistor (or a tube in the old times) presenting a reactance (inductive or capacitive, according to topology) whose value is varied with the transconductance of the device. Transconductance is controlled by the operating point of the trsansistor.
 
neazoi said:
Do you refer to the Q of the tuned circuit?
Click to expand...

When C is varied in an LC circuit, the resonant frequency changes but the bandwidth remains constant.
This is important because it is imperative that the bandwidth of an LC circuit remains the same throughout
the tuning range. If L rather than C is varied, however, the bandwidth would vary throughout the tuning
range of L. Specifically, if the circuit's resonant frequency is doubled by reducing the value of L, the Q of
the circuit is halved and the bandwidth is quadrupled.
Conversely, if the circuit's resonant frequency is halved by increasing the value of L, Q doubles and the
bandwidth is reduced by a factor of 4. The condition of a varying bandwidth is highly undesirable. This
is why an LC circuit is usually tuned by varying C rather than L.
 
E-design said:
When C is varied in an LC circuit, the resonant frequency changes but the bandwidth remains constant.
Click to expand...
It's not correct..For series resonant LC circuit
Q=(1/Rs)xSQRT(L/C) and BW=Fo/Q
So, when you change C or L, the bandwidth will also be changed.
 
Last edited:
  • Like
Reactions: neazoi

    neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
zorro said:
Yet another possibility is the reactance modulator: a transistor (or a tube in the old times) presenting a reactance (inductive or capacitive, according to topology) whose value is varied with the transconductance of the device. Transconductance is controlled by the operating point of the trsansistor.
Click to expand...

I would like to see an example of this technique

- - - Updated - - -

BigBoss said:
It's not correct..For series resonant LC circuit
Q=(1/Rs)xSQRT(L/C) and BW=Fo/Q
So, when you change C or L, the bandwidth will also be changed.
Click to expand...

Maybe that is why there are so different L/C ratios that can be used for a frequency range, but only some are usually chosen. For example one does not use a very small cap in combination with a very low inductance and vice versa
 

The reason that the bandwidth does not change when C is varied is that the Q of the circuit
varies in direct proportion to the resonant frequency. That is, if the resonant frequency is doubled by
reducing the value of C, the circuit Q also doubles. Conversely, if the circuit's resonant frequency is halved
by increasing C, the circuit Q is also halved. Since BW = fo/Q, it is obvious that there can be no change
in the circuit's bandwidth as C is varied throughout its tuning range.
 

Attachments

  • Adj_LC.gif
    Adj_LC.gif
    240.9 KB · Views: 137

In the following, let's consider series and parallel circuits having 3 elements (L, R and C) and narrow-band aproximation (i.e. Q>>1).
G=1/R is conductance.


For a series circuit, the impedance is

sL + R + 1/(sC)

The 3-dB bandwidth is Δω=R/L, i.e., it does not change with C.


We have the dual situation as follows:


For a parallel circuit, the admittance is

sC + G + 1/(sL)

The 3-dB bandwidth is Δω=G/C =1/(RC), i.e., it does not change with L.


Regards

Z
 
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top