The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> Ideal Balun
https://designers-guide.org/forum/YaBB.pl?num=1034815521

Message started by William Ngai on Oct 16^th, 2002, 5:45pm

Title: Ideal Balun
Post by William Ngai on Oct 16^th, 2002, 5:45pm

Hi,

Based on the paper "A Test Bench for Differential Circuits", I created an "Ideal Balun" and used PSS to simulate the 1dB CP of a PA. However, I could not get the simulation to converge. After replacing the "Ideal Baluns" with VCVS, the simulation converged.

Does it means this "Ideal Balun" add extra simulation load to Spectre or make it more difficult to converge?

Thanks,
William

Title: Re: Ideal Balun
Post by Ken Kundert on Oct 17^th, 2002, 12:16am

William,
An ideal balun, if connected properly, will not create any convergence difficulties. I suspect you have done something wrong. The most likely source of the problem would be not defining the common mode input. Have you provided a DC path to ground from the common mode input?

If you cannot find the problem, post the netlist for your test bench and we take a look.

-Ken

Title: fRe: Ideal Balun
Post by William Ngai on Oct 17^th, 2002, 9:51am

Hi Ken,

Since there should not be any DC path to the input of my PA, I just leave it connect to an NC. I tried connect thru' a cap of 10nF to ground but it didn't help.

Thanks,
William

Title: Re: Ideal Balun
Post by Ken Kundert on Oct 17^th, 2002, 11:16am

William,
Something must define the common mode voltage at the input of the PA. If not the test bench, then it must be the PA itself.

When your circuit is in the larger system, what common mode impedance is your circuit expecting to see when looking out from its input?

-Ken

Title: Re: Ideal Balun
Post by William Ngai on Oct 17^th, 2002, 2:50pm

Hi Ken,

Make it clearer:

The balun is feeding into the Driver amp which has input DC blocking caps. The common mode input of this input balun is connected to a DC voltage at 0V (it supposes to see inductors//drains in common mode from previous stage). The PA is connected to the Driver amp and the output of PA is connected to the ouput balun. Since the outputs are directly taken out from the Drain inductors. The internal common mode is well defined and should see the indutors in parallel the the drain impedance. There should not be any DC common mode voltage being contributed by the o/p balun and so it's common mode input was connected to NC.

Thanks,
William

Title: Re: Ideal Balun
Post by Ken Kundert on Oct 17^th, 2002, 9:56pm

William,
From your description it sounds like you are doing everything right. I recommend carefully checking the subcircuit you created for the ideal balun assure that there are no errors. By not connecting the common mode terminal of the output balun, you are forcing the output currents in the p and n terminals to be exactly equal and opposite. Is there anything in the circuit that would prevent that from happening? If those things do not lead anywhere, you could post the netlist for the top-level of your circuit and perhaps a description of how the simulator is failing. Someone might be able to make a suggestion from that.

-Ken

Title: Re: Ideal Balun
Post by William Ngai on Oct 29^th, 2002, 10:42am

Hi Kent,

I've got it fixed, the simulation time doesn't differ by much. Probably I've set to different convergence criteria.

Thanks for you suggestions and sorry for my late reply,
William