The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> responsivity simulation of AM detector
https://designers-guide.org/forum/YaBB.pl?num=1487835028

Message started by Kyle on Feb 22^nd, 2017, 11:30pm

Title: responsivity simulation of AM detector
Post by Kyle on Feb 22^nd, 2017, 11:30pm

Hi everyone,
I'm working on a 20GHz amplitude modulation detector right now. And I want to get the responsivity with spectre. Does anyone knows which analyses I should use?
Here is what I think will work.

1. Use hb simulation to get the number of input power. And then use transient simulation to get the voltage amplitude of output signal.
Is it right? Is there anyother ways to do it.

2. Seem the envlp analysis could do it.
If I try to plot voltage magnitude in spectrum domain. there several parameters I need to define. But I'm not clear how to define them.
Nyquist half-bandwidth:
Frequency bin width:

thank you guys if anyone knows about this.

Title: Re: responsivity simulation of AM detector
Post by Andrew Beckett on Feb 23^rd, 2017, 10:14pm

If by "responsivity" you are asking about the gain of the detector, then this could be done with harmonic balance directly (or maybe hb+hbac if the linear conversion gain is sufficient). I'm not entirely sure what style of AM detector you're talking about, but I see no reason why you'd need to also use a transient or envelope simulation to measure this.

Andrew

Title: Re: responsivity simulation of AM detector
Post by Ken Kundert on Feb 24^th, 2017, 12:01pm

I don't think there is any need to limit yourself to harmonic balance. Shooting methods should work fine and in general give better accuracy.

You can measure responsivity with a simple PAC analysis where the analysis frequency is the input frequency and the output frequency is baseband (use sweeptype = relative and observer the -1 sideband).

Alternately you can use a PXF analysis where the analysis frequency is baseband (use sweeptype=absolute and observe the +1 sideband).

-Ken

Title: Re: responsivity simulation of AM detector
Post by Kyle on Feb 24^th, 2017, 2:27pm

Andrew Beckett wrote on Feb 23^rd, 2017, 10:14pm:

Thank you for the reply.
Voltage Responsivity= output volatge(baseband)/input power(RF).

The detector is a kind of doubler detector as shown in below figure.

Btw, when I use hb, there is always a error that "fund frequency 20GHz does not divide evenly into the stimulus frequency 'ammodfreq' 80MHz"
I don't understand this error.

Title: Re: responsivity simulation of AM detector
Post by Kyle on Feb 24^th, 2017, 2:32pm

Ken Kundert wrote on Feb 24^th, 2017, 12:01pm:

Hi Ken, thx for the reply.
By using ac simulation, I think I can't get the right value of the input power. What do you think?
Thx

Title: Re: responsivity simulation of AM detector
Post by Ken Kundert on Feb 25^th, 2017, 5:35pm

You definitely would not use AC analysis. Must be PAC or PXF.

All of these are small signal analyses, so their input amplitude is of no real consequence as long as it is not zero. However you do need a large signal present that the small signal 'modulates'. If you just perform a simple PAC analysis, you are adding a single sideband, which means that you are modulating both the amplitude and the phase of the 'carrier'. Alternatively, you can use these modulators to apply only a specific type of modulation to the input.

-Ken