The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Modeling >> Behavioral Models >> Transition filter issue in VAMS model
https://designers-guide.org/forum/YaBB.pl?num=1494325399

Message started by Gp on May 9^th, 2017, 3:23am

Title: Transition filter issue in VAMS model
Post by Gp on May 9^th, 2017, 3:23am

Hi All,

I am facing the following issue with transition filter.
I give the input V(out)<+ transition (in,20n,5n,3n). In this case, time delay=20n, rise time=5n & fall time=3n. In that case TON=10n.

Now my output generating the following way:
time delay=20n, rise time=5n & fall time=3n. But TON=8n. It's vary with different value of fall time. So, I think it should be same (10n).

So, is it my right way or not? if not, then what is the right concept?

Thanks.

Title: Re: Transition filter issue in VAMS model
Post by Ken Kundert on May 9^th, 2017, 1:43pm

Seems to be a lot of missing information in this question. For example, what is TON?

Perhaps you can think about your question from the perspective of somebody that is not familiar with what you are working on, and then fill in the missing details.

-Ken

Title: Re: Transition filter issue in VAMS model
Post by Gp on May 9^th, 2017, 8:42pm

Hi Ken,

Here TON is ON time. I think,
Pulse width=Rise time+fall time+TON. Is it true?

Title: Re: Transition filter issue in VAMS model
Post by Ken Kundert on May 9^th, 2017, 10:54pm

I'd like to help you, but you are giving me nothing to work with. You say it is the on time, but what is turning on?

It might be more helpful if you gave us some waveforms. For example, what is the input, what is the output, how is the output not meeting your expectations.

-Ken

Title: Re: Transition filter issue in VAMS model
Post by Gp on May 9^th, 2017, 11:33pm

Hi Ken,
Here is the VAMS code and also attached waveform of 5ns ON time. My issue is, ON time of input is 10ns and output is 5ns. SO, why changes this time.

All other parameter rise time and fall time are met expectation.

`timescale 1ns/1ps
`include "disciplines.vams"
`include "constants.vams"

module transition_1(out,in);

input in;
output out;
wreal in;
electrical out;
real Vin;

analog begin
V(out)<+ transition (in,20n,5n,3n);
end

endmodule

Title: Re: Transition filter issue in VAMS model
Post by Ken Kundert on May 10^th, 2017, 9:40am

Okay, that I can answer.

The transitions in the output begin 20ns after the corresponding transitions in the input.

The fact that the output only sits at its high value for 5ns where as the input is high for 10ns is due to the 5ns rise time.

You are using an unusual definition of TON. In real circuits the output would approach its equilibrium point slowly and so you would generally define TON using a threshold, perhaps the 50% point. If you did that you would get something closer to 10ns. It still would not be exactly 10ns because of the asymmetry in the rise and fall times.

-Ken

Title: Re: Transition filter issue in VAMS model
Post by Gp on May 11^th, 2017, 1:45am

Thank you very much.