Question about LC circuit equivalency

Go To Last Post
9 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I should know this, but are these two LC circuits equivalent in terms of frequency response and other characteristics? (Pretend that MMMMM is the symbol for an inductor.)

Circuit 1:

Avcc ------MMMMM------|----- +5V
             L        |
                      |
                    C -
                      -
                      |
                      |
                     GND

Circuit 2:

Avcc --|----MMMMM------- +5V
       |      L        
       -               
       - C            
       |               
       |               
       |               
      GND              

Maybe a better way to phrase the question would be as a sort of multiple choice. Are these circuits:

A) Functionally identical?
B) Slightly different, but close enough for hobby work?
C) Totally different?

The reason I ask is because it's easier for me to wire up circuit 1 than circuit 2 on a breadboard, so if they are reasonably similar in terms of frequency response I'd rather use circuit 1. I'm interested in understanding the frequency response of the theoretical circuits, rather than the practical issues that can arise from using a breadboard.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The 'in' is 5V, the 'out' is AVCC. So the bottom one is a 2nd order lo pass filter, 12 db per octave. The top one is just an L in series with VCC. (which is a 1st order lo pass, 6 db per octave)

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, depends on whether you are talking about an "ideal" circuit or a "real" circuit.

In an ideal circuit, the voltage source has zero source impedance (resistance), In that case, the C of Circuit 2 will have no effect. So, the only frequency dependent part will be L. In circuit 1, the series L impedance goes up with frequency, the shunt C impedance goes down with frequency. They make a voltage divider that divides more and more as frequency goes up; good filter!

In the real case, Vcc has some series resistance AND probably some series inductance. These tend to be small to moderate in value. So, with circuit 2, the combination of the source resistance and C will cause some attenuation at higher frequencies. The series L will increase with frequency, so that it, also will provide more attenuation at high frequencies. Circuit 1 will behave only slightly different than it did with no source resistance?

So, which is "better"? The generally recommended one is Circuit 1 because it does NOT depend on the source resistance of the voltage source. In fact, with some voltage sources (LDO regulators, in particular), adding shunt C CAN cause the regulator to start oscillating. This is a very NOT GOOD situation.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jim, you have it backwards. You are reading from left to right. In actuality, 5V is the 'in', and the 'out' is AVCC, so read from right to left.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Circuit 2 is a correct second order LC low-pass noise filter for AVCC like in the application notes,

Circuit 1 presents just a series inductor for AVCC and is horribly wrong because it has high impedance at high frequencies, so every time AVCC current changes rapidly the voltage will change too so AVCC voltage has ripple, noise and ringing perhaps.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Correct. I did not read the schematic carefully enough!

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
In fact, with some voltage sources (LDO regulators, in particular), adding shunt C CAN cause the regulator to start oscillating.
warning: useless academic drivel ahead.
Is this only if it shifts the resonant frequency towards one of the driving frequencies (50/60/100/120Hz for linear; harmonics of switcher f for SMPS)? ie: 1/√(LC) ~ ω_source. eg: 100uF & 0.1nH gives > 10MHz pole. 0.1F (BFC) & 4mH gives 50 Hz pole.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

To put in in simpler terms -

For an LDO (which was the ONLY thing I was talking about), most have a chart showing the range of stable load capacitance. Some have both minimum and maximum limits. That is, if the load C is too BIG, it will oscillate!

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the advice. I moved some things around on the board and got it wired up the right away.