Maker Pro
Maker Pro

Implementation of Amplifier in LTSpice

robi10101298

May 25, 2020
21
Joined
May 25, 2020
Messages
21
Hello, I need to simulate the following circuit in LTSpice. I made the scheme but I don't know what value to choose for R2 and Cinf in order to have theoretical gain Av=1500. The following constraints are imposed: Vcc=10v; VCM=1.5V; vg it's a sine wave with a frequency of 1Khz and R1=15kOhms. I need to simulate the circuit(transient, AC sweep), choosing an appropriate amplitude for vg in order to minimize output signal distortion(I put 100 uV but idk if it's correct..), and to specify a method for decreasing the voltage gain (without changing R1, R2, VCC or VCM). Implement the change and re-simulate the circuit in order to demonstrate the effect. I've attached the .asc file for the ltspice.
 

Attachments

  • Project EDC.asc
    2.7 KB · Views: 11
  • 96234628_10213667725885840_7038115341192396800_n.jpg
    96234628_10213667725885840_7038115341192396800_n.jpg
    55.5 KB · Views: 9

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
Here are a first few tips:

  • The task description states "Cinf is a capacitor which behaves as a short-circuit at the frequency of vg". Think how the value of Cinf can influence the gain when it is effectively a short-circuit.
    To find a realistic value for this capacitor (1 F is not realistic) I personally would assume an impedance of less than 1 % of R1 as a short circuit. Knowing the frequency of vg allows you to compute the capacitance.
  • To find the value for R1 you should ideally analyze the transfer function of the circuit with R2 being the parameter. Then you can find R2 from this transfer function.
    If you cannot analyze the circuit, you can find a value for R2 by making a batch of simulations with varying values for R2, changing R2 in the process until you have reached the target gain.
    Note: This is not the way a simulator should be used. Normally you use the simulator to verify your theoretical result.
  • Select an appropriate amplitude for Vg such that the amplifier operates in the linear region without creating distortions. A very small voltage is sufficient as you will evaluate small signal gain only.

As a note on the side: your ASC file will not simulate. The base of Q8 is not connected, although it looks as if in the schematic. But when you drag Q7 you'll see the open connection: The wire from the base of Q8 is not attached to the base of Q7
upload_2020-5-25_20-28-47.png

I'm not sure about the voltage source Vcc. It seems to be a DC source but shows an AC symbol. If you encounter issues when simulating, swap Vcc for a DC source as used for Vcm.
 

robi10101298

May 25, 2020
21
Joined
May 25, 2020
Messages
21
Here are a first few tips:

  • The task description states "Cinf is a capacitor which behaves as a short-circuit at the frequency of vg". Think how the value of Cinf can influence the gain when it is effectively a short-circuit.
    To find a realistic value for this capacitor (1 F is not realistic) I personally would assume an impedance of less than 1 % of R1 as a short circuit. Knowing the frequency of vg allows you to compute the capacitance.
  • To find the value for R1 you should ideally analyze the transfer function of the circuit with R2 being the parameter. Then you can find R2 from this transfer function.
    If you cannot analyze the circuit, you can find a value for R2 by making a batch of simulations with varying values for R2, changing R2 in the process until you have reached the target gain.
    Note: This is not the way a simulator should be used. Normally you use the simulator to verify your theoretical result.
  • Select an appropriate amplitude for Vg such that the amplifier operates in the linear region without creating distortions. A very small voltage is sufficient as you will evaluate small signal gain only.

As a note on the side: your ASC file will not simulate. The base of Q8 is not connected, although it looks as if in the schematic. But when you drag Q7 you'll see the open connection: The wire from the base of Q8 is not attached to the base of Q7
View attachment 48385

I'm not sure about the voltage source Vcc. It seems to be a DC source but shows an AC symbol. If you encounter issues when simulating, swap Vcc for a DC source as used for Vcm.
Hello, thanks for the modified scheme! The file will simulate but will not give satisfactory results, for Cinf our teacher said that 1farads it's more than enough for this simulation, Vg amplitude I set to 100uV which I think it's more than enough. My problem it's with R2, I computed as being 968kohms and idk if it's correct, my theoretical gain should be Av=1500.
 
Last edited:

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
for Cinf our teacher said that 1farads it's more than enough for this simulation
For a simulation this is fine. In the real world you'll have some problems finding such a capacitor useful for this kind of application.
My problem it's with R2, I computed as being 968kohms and idk if it's correct, my theoretical gain should be Av=1500.
What do you get? How do you determine the gain?
I get gain = 955 ... So your determination of R2 is not correct.

Tip: swap Vcm and Vg on the right input of the amplifier. The resulting voltage at the base of Q2 is the same (voltages add up anyway), but you have easier access to Vg with reference to gnd.
 

robi10101298

May 25, 2020
21
Joined
May 25, 2020
Messages
21
For a simulation this is fine. In the real world you'll have some problems finding such a capacitor useful for this kind of application.

What do you get? How do you determine the gain?
I get gain = 955 ... So your determination of R2 is not correct.

Tip: swap Vcm and Vg on the right input of the amplifier. The resulting voltage at the base of Q2 is the same (voltages add up anyway), but you have easier access to Vg with reference to gnd.
  • ICq5=ICq6=ICq7=ICq8=(Vcc - Vbe)/R2;
  • ICq1=ICq2 = [ICq7 / 2 ];
  • ICq4=iCq2-current of Q5;
  • ICq3=ICq1-current of Q3-current of Q4;
  • Beta in my case it's 524.9 because I'm using an BC847C transistor.
  • A=Vo/Vg = Gm_q2 * Beta_q5 * R1Gm_q2=ICq2 / Vt that it's approx 40*ICq2
  • So, Av(1500)=40*IcQ2*Beta_q5*R1
  • IcQ2=IcQ7/2
  • iCq7=(Vcc-Vbe)/R2=9.4/R2
  • We get that 1500=40*9.4/R2*515.4*15000, and the final result it's R2=968.952kohms
So basically the value computed for R2 it's not the same as the value from simulations...
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
What gain does your simulation give you? How do you determine this gain?
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
Please, my question is not that difficult: You expect gain = 1500, Which value do you get? How do you determine that value from the simulation?
 

robi10101298

May 25, 2020
21
Joined
May 25, 2020
Messages
21
Please, my question is not that difficult: You expect gain = 1500, Which value do you get? How do you determine that value from the simulation?
As I said, I determine the voltage gain in the simulation by substracting the maximum amplitude from the minimum one. I think this is the correct answer, if not, please help me to Better understand how can I use this...
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
As I said, I determine the voltage gain in the simulation by substracting the maximum amplitude from the minimum one. I think this is the correct answer,
Gain is not determined by subtraction but by division: gain = vout/vin.
What is the value you get?
 

robi10101298

May 25, 2020
21
Joined
May 25, 2020
Messages
21
Gain is not determined by subtraction but by division: gain = vout/vin.
What is the value you get?
I've updated the scheme in order to not float Q7. So, I ve measured the gain and it's approx 1000. How I did this? I simulated between Cinf and R1 and I got that it's almost 1000.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
As I said in the other forum, you cannot make a high gain amplifier without any DC negative feedback because the tiny offset voltages are amplified 1500 times.
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
you cannot make a high gain amplifier without any DC negative feedback because the tiny offset voltages are amplified 1500 times.
True for any real world amplifier. In the simulation where every component is ideal this works.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
My simulation shows the DC voltage feeding the output coupling capacitor saturated at +9.9974V without negative feedback and a reasonable lower DC voltage with some DC negative feedback.

Q8 does not have enough base current since without negative feedback its base voltage is only 0.316VDC.
 
Top