ADS difference between SP simulation and Transient simulation

Status
Not open for further replies.
S

shuma394

Newbie level 6
Joined
Dec 11, 2014
Messages
11
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Visit site
Activity points
64
Hello.
I built a LNA at 300GHz.
so I got 21dB gain (S-Parameter simulation)

because i want to make a detector, i need a time domain graph.

but LNA didn't work.

so i sweep LNA at 50~400GHz, i got a graph attached.



why there is difference between SP sim and Tran sim ?

how i can solve this problem ??
 

It's possible that your circuit oscillates at around 75 GHz and in Transient simulation the circuit's performance is affected by this oscillation. Check your current and voltages and check for beginning s11 and s22 if their magnitudes become positive at around 75 GHz.
 

I just checked gain from input, output peak at every 10GHz
 

voltage gain or power gain? S-parameter simulation is only a small-signal analysis, while transient includes large signal analysis. Nevertheless, check your current and voltage waveforms from transient analysis.
 

I attached some result.

i tried to HB simulation again, i got different result but i didn't change anything ..

i don't know how i solve this problem..

there are current and voltage waveforms.


(HB result)

(SP and Tran result)
 

So for S-parameter results you take a 50 Ohm terminated source and for transient an ideal voltage source?
 

I revised circuit, but there is no change..



i simulate ideal BJT at TRAN and SP.
there is difference.




i change load's R, because matching.
why gain is different ? because SP is only small signal ??

i think active element make a difference between SP and Tran simulation.
and if i find this difference, i can solve first problem ..
 

Just in brief, from your first diagrams (transient ?) oscillation is visible. Second diagram, the voltage gain = 20*log (x) not 10*log (x), did you roughly calculated your voltage gain of your ideal common base ampl with gm*Rl?
 

By the way, for s-parameter analysis place proper dc-blocks and dc-feeds at your sources and terminals.

- - - Updated - - -

And last thing to remind, s21 is the forward voltage gain under matched (Zref) condition. I attached an ideal simulation with calculation of s21 and voltage gain, you see it works under matched conditions, otherwise it is not comparable.
 

Attachments

  • 1.jpg
    465.7 KB · Views: 156
  • 2.jpg
    208.1 KB · Views: 146
Last edited:

Thank you for your help.
i have some questions.
in your circuit, load R is very large.
because check output voltage, Open circuit setting ?

my goal is check demodulator's performance.
and i need amplfying before demodulating.
if i connect amp with demodulator(input R is 50ohm) instead load, gain will be change ..
 

The load has a high value because of the high output impedance of the ideal transistor and to meet the matching and show the relation between s21 and voltage gain. I justed wanted to show that you can only compare s21 and voltage gain under matched condition.

Is your demodulator a real circuit? Maybe it has some additional reactive input impedance, if you connect a simple 50 Ohm resistor, its characteristic is constant over frequency and includes no reactive components.
 

I'm sorry very late.
I tried many things for finding difference between HB and TRAN.
I checked other components.
R, C, L show same results.
But transmission line and transistor show differences.

I attach results of transmission simulation.


I made a very simple circuit for test.

This S21 exactly same with result of SP-Sim.

left is HB ans right is TRAN results.

I tried to deduct same Vout, Vin, Power Gain, very hard to me.

Is there some errors ? or mistakes ??
I don't know why transmission line and transistor have difference between HB and Transient.

- - - Updated - - -

i attach other result(change to 50ohm transmission line)

<- this is HB result
<- Transient result

i think i miss something when simulate transient sim ..

- - - Updated - - -

And I find some problem.

it's okay until 250GHz.
but 250.1 GHz , it's strange
 

shuma394 said:
And I find some problem.

it's okay until 250GHz.
but 250.1 GHz , it's strange
Click to expand...

I can confirm the problem, it's the the ideal tline component, if you set F till 250 GHz, everything fine, 251 GHz you get problems. If you take instead the TLIN4 component everything works fine, also for 300 GHz. Later more.
 
I think TLIN4 is well.
what is difference between TLIN, TLIN4 ??
 

I contacted Keysight, They say ideal TLine don't have algorithm upper than 250GHz.
And they will correct this problem at next version of ADS.

Thank you for your help.
 

Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…