The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Analog Verification >> Analog Functional Verification >> Ideal voltage source in spectre
https://designers-guide.org/forum/YaBB.pl?num=1257312753

Message started by rampat on Nov 3^rd, 2009, 9:32pm

Title: Ideal voltage source in spectre
Post by rampat on Nov 3^rd, 2009, 9:32pm

Hi

For my test bench I am using ideal voltage source for supply.Now I want to limit the current from voltage source (like LDO).How can I model this in my testbench..
Is there a LDO model in any of cadence libraries ?

Thanks
ramapt

Title: Re: Ideal voltage source in spectre
Post by rajdeep on Nov 5^th, 2009, 4:04am

You can use a series resistance to model the output resistance in a way that whenever the current is beyond a certain limit the drop across the resistance becomes quite a significant drop!! If you spectre ideal components then it would be hard to makethe resistance variable. You can use a constant resitance, but that might not suffice your requirement.

You can use Verilog-A to model such a voltage source where you can make the output resistance variable so that the voltage drop is insignificantly small until the current becomes more than what you want. At that point you make the resistance value to change to a larger value to cause a significant drop across the resistor.

hope that helps..
Rajdeep

Title: Re: Ideal voltage source in spectre
Post by jerome_ams on Nov 24^th, 2010, 9:31am

I agree with Rajdeep: a variable serie resistance is the usual way to get a realistic behavior (you get a significant voltage drop when the current exceed the limit).
Another alternative would be to implement in VerilogA a current limiter, which will clamp the current to a certain value, with no impact on the voltage delivered by your voltage source... This would be a more abstract way, less related to how a "real" LDO would behave. It depends what u need.

Cheers,
jerome

Title: Re: Ideal voltage source in spectre
Post by Geoffrey_Coram on May 13^th, 2015, 8:31am

It's really hard to implement a current-limiter in a way that doesn't cause convergence problems.