RF pi network design with LTspice question

Thread Starter

gonzalezd91

Joined Sep 6, 2019
6
Hello everyone I'm new to this and to circuit design so thank you for your help. My main issue is that my pi network circuit on LTspice does not match the same results as the breadboard circuit i tested. Same components and same parameters, I attached a screen shot of the pie circuit from LTspice, the components consist of: 2V input and 1MHz frequency AC source, with one 100pF, two 220pF, and a third 47pF capacitors, and two 1 microHenry inductors.
At 1MHz frequency the simulation gets about 2.3 V output, the breadboard gets 1.8V output
At 14MHz frequency the simulation gets about 5V output, and the breadboard gets somewhere around the mV (very different)

If anyone could help I would greatly appreciate it, im assuming the error comes from something missing on the simulation circuit i just don't know what.

Daniel
 

Attachments

crutschow

Joined Mar 14, 2008
23,523
If you want the network to be driven by a 50Ω source, then the 50Ω resistor must be in series with the voltage source (generator) not from the input to ground.

You simulation results are likely different because you don't have any parasitic capacitance and inductance shown in the simulation.
For example, the stray capacitance across the inductor, and the wiring inductance (which can have a significant effect at 14MHz) are not shown.

How was the breadboard constructed?
Post a picture.
 

Thread Starter

gonzalezd91

Joined Sep 6, 2019
6
If you want the network to be driven by a 50Ω source, then the 50Ω resistor must be in series with the voltage source (generator) not from the input to ground.

You simulation results are likely different because you don't have any parasitic capacitance and inductance shown in the simulation.
For example, the stray capacitance across the inductor, and the wiring inductance (which can have a significant effect at 14MHz) are not shown.

How was the breadboard constructed?
Post a picture.
Once again thank you I'll attach the breadboard image.
 

Attachments

crutschow

Joined Mar 14, 2008
23,523
Yes, a plug-in breadboard is absolute inappropriate for any high frequency circuits. :eek:
Besides what RPL suggested, you could just solder them together in the air with short leads, in a 3d rats-nest style.
That will have a minimum of stray impedances.
 

Thread Starter

gonzalezd91

Joined Sep 6, 2019
6
Thanks a lot guys ill look into doing this on the coming monday ill let you guys know the outcome and further question, im really glad i asked this question! you guys rock!
 

KL7AJ

Joined Nov 4, 2008
2,223
Hello everyone I'm new to this and to circuit design so thank you for your help. My main issue is that my pi network circuit on LTspice does not match the same results as the breadboard circuit i tested. Same components and same parameters, I attached a screen shot of the pie circuit from LTspice, the components consist of: 2V input and 1MHz frequency AC source, with one 100pF, two 220pF, and a third 47pF capacitors, and two 1 microHenry inductors.
At 1MHz frequency the simulation gets about 2.3 V output, the breadboard gets 1.8V output
At 14MHz frequency the simulation gets about 5V output, and the breadboard gets somewhere around the mV (very different)

If anyone could help I would greatly appreciate it, im assuming the error comes from something missing on the simulation circuit i just don't know what.

Daniel
What is the response in the very center of the passband?
 

KL7AJ

Joined Nov 4, 2008
2,223
If you want the network to be driven by a 50Ω source, then the 50Ω resistor must be in series with the voltage source (generator) not from the input to ground.

You simulation results are likely different because you don't have any parasitic capacitance and inductance shown in the simulation.
For example, the stray capacitance across the inductor, and the wiring inductance (which can have a significant effect at 14MHz) are not shown.

How was the breadboard constructed?
Post a picture.
Not to mention parasitic resistance in the inductors. Even though you probably can't measure any DC resistance, at RF the skin effect will be significant.
 

Thread Starter

gonzalezd91

Joined Sep 6, 2019
6
Sorry for the late response, I finally had time to make the rat nest circuit and I'm now comparing the results in LTspice simulation with the 50 Ohm resister placed in series with the source, their results are still poor compared to the simulation especially at 14 MHz still resulting in mV output.

If i remove the 50 Ohm resistor i get a 3 times increase in the simulation, on the rat nest test without the 50 ohm terminator gives me a 7.4 times voltage increase.

For the pass band as i changed the frequency would have max voltage peaks @ 968 kHz, 1.03 MHz, and 2.78 MHz being the highest voltage output (1.7 V from a 2 V input)
 

Thread Starter

gonzalezd91

Joined Sep 6, 2019
6
Also is it safe to assume that the simulation doesn't always give realistic results? Maybe it doesn't account the parasitic and skin effect impedance.
 

Thread Starter

gonzalezd91

Joined Sep 6, 2019
6
In addition to what @crutschow said in post #2, you also need a 50Ω load resistor at the END of your circuit, so 50Ω IN and 50Ω out.
Oh I assumed it was just for the IN, alright ill adjust that in the simulation thanks!
As for the RLC parasitic effect i wouldnt know how i can find the exact numbers to put that in LTspice however i can just place random numbers to see how it effects the output.
 

Bordodynov

Joined May 20, 2015
2,421
I'll give you some advice. You have not considered the influence of the observer on your scheme. You are not God and you cannot bypass the use of real devices with parasitic parameters. At a certain frequency, inductance is of good quality. You can simulate this value with a sequential or parallel resistor (individually or together), which are the inductance parameters in LTspice. You can take a look at an example of a 1 µH throttle parameter from the LTspice library. I advise you to use the Q=20 -100 quality first. Rser=2*pi*Freq*L/Q or Rpar=2*pi*Freq*L*Q.
 
Top