URGENT (PLEASE HELP) AC Analysis Matrix Singular Error

[freakcrazy]

Hi all (sorry to post on a new thread again Yao) , my project is due tomorrow at 9 am. Therefore please please please help me on this circuit.

I can't perfom AC analysis on the circuit for that there is a matrix singular error. May anyone please tell me what is wrong ??

Please! Thank you!!!

[RLC]

I don't know what you are trying to do but the circuit will not simulate because the ground near the negative side of V1 is not actually connected to anything. If you connect the ground to the wire on the negative side of V1 the circuit simulates like a champ.

[freakcrazy]

Hey RLC thank you for replying!!! I think I did what you said; it doesn't work!

I attached my revised circuit here. Did I do it right? if not, may you please attach your corrected version of my circuit?

[RLC]

You still didn't connect the ground - you have to draw in a wire from the ground to the existing wire you are trying to ground. You can't just shove the ground against the wire to get a connection.

[freakcrazy]

Hey man, I tried your circuit for simulation. I still gives singular matrix error. :mansad:

[RLC]

Yes, it will fail on an ac analysis which I did not do the first time because I didn't know how you had it set up. I just simulated what was showing. Anyway, the reason for the singular matrix is that you have put the integrator into the circuit backward. I'm attaching your file updated to reflect that as well as a simplified version that I think does the same thing.

[freakcrazy]

RLCCCCCCCC!#@#!@!#@#@#!@#!@@#!!#@#@!@!# THANK YOU!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! YOU ARE MY GOD NOW!!! ha gosh, THANK YOU for going through the trouble to simplify the circuit for me!!! I THANK YOU WITH MORE THAN WORDS CAN SAY!!! you can't imagine how happy I am right now, after spending 2 weeks gruelling over errors after errors. So the integrator has sides as well!!! sigh...

RLC may I please ask you ONE more thing. I tried all the variables for the output (at least the current ones), none of them gives me a resonant peak; this means that none of the outputs I chose correspond to the current through the multimeter. The current appeared on the multimeter shows resonance (which should be at 50 Hz - 60 Hz; any other frequencies should give a larger current. (therefore it's anti-resonance)). How should I perfrom this AC anylsis to so that the output is the multimeter current? (sorry for my extreme novice questions)

[RLC]

OK, I added a current probe to measure the current thru the multimeter (You could also use a current - controlled voltage source) but no matter what I look at I can't get an ac analysis to give a non-zero gain.  I simplified the circuit some more and set the option for the node numbers to show. There is something fundamentally wrong in the area of the ac voltage source V1 and the voltage controlled capacitor. As long as the voltage source is in parallel with the capacitor, the capacitor current will be driven only by the voltage source. The capacitor voltage will always be equal to the voltage source V1. The current source A4 doesn't accomplish anything - its current will just be sunk into the voltage source V1. I thought maybe you really meant to measure the voltage controlled capacitor current and feed that back into one of the inputs of A4 but I can't really tell what you are trying to accomplish. Maybe this will give you some ideas.

[RLC]

I think I may have figured out why the ac analysis doesn't work. According to the help it only works on LINEAR circuits. Sure enough, if I take out the square of V1 and just make it unity gain, the ac analysis works. If you can't accomplish what you want to do with a linear circuit, you may have to just calculate the ac gain point by point at multiple frequencies.

I usually use the Bode Plotter to do this kind of thing but it will not work with this circuit either.

There is still something screwed up about the voltage source - capacitor thing but fixing it probably will not make the ac analysis run.

[freakcrazy]

Gosh...sigh. So close of getting it to work. Thanks alottt RLC. I really appreciate what you done for me. Maybe my circuit is just incorrect. Thanks alot man and take care. :manhappy: