S22 not matching simulation

[arwen16]

Hello,

I had designed an LNA using an HBT from Infineon on ADS and had it fabricated and tested. The S11 measured was a good match with thr simulation results. However, the S22 measured looked like a flipped version of the S22 simulated. I don't know how to debug this issue.

 

[volker@muehlhaus]

. I set the mesh frequency to 6 GHz and the number of cells/wavelength= 80. The EM simulation is taking a really long time (more than a day). I want to understand if I am doing this right. I wish to reduce the simulation time.

There is no need (and no advantage) to set 80 cells/wavelength, it is enough to use 20 cells/wavelength and combine that with edge mesh. Note that you can enable edge mesh selectively per layer or per polygon.

 

[arwen16]

There is no need (and no advantage) to set 80 cells/wavelength, it is enough to use 20 cells/wavelength and combine that with edge mesh. Note that you can enable edge mesh selectively per layer or per polygon.

I did do that. My measurement still does not match the EM simulation. I tested the matching networks independently, and they match the EM simulation. However, after integrating everything, the results do not agree. When I feed the EM model of the matching network with the required load through a DAC, the result obtained resembles the pattern obtained during measurement of the LNA. But the EM model of the LNA and the measurement results of the LNA do not match.

 

[volker@muehlhaus]

I did do that. My measurement still does not match the EM simulation. I tested the matching networks independently, and they match the EM simulation.

So the difference is in the active device?

I suspect that some amount of inductance is included in the HBT S-parameter data -- whatever they used during measurement to create the ground path. If you now include the full ground path in EM, you might over-estimate inductance in that path (your layout's ground path L + HBT S2P measurement ground path L).

So maybe you can go to circuit simulation and tweak some +/- delta L to see how that affects overall results.

 

[dick_freebird]

I'd expect it has something to do with what's attached
to the output port. Something not represented in your
simulations. Perhaps an attribute of the test equipment.
Perhaps you want to focus on that with a VNA to see if
your modeled "bench" represents your simulated, or
how it does not.

 

[D.A.(Tony)Stewart]

Time to calibrate your 2 port test set with SOT (short , open, through)

 

[arwen16]

I

So the difference is in the active device?

I suspect that some amount of inductance is included in the HBT S-parameter data -- whatever they used during measurement to create the ground path. If you now include the full ground path in EM, you might over-estimate inductance in that path (your layout's ground path L + HBT S2P measurement ground path L).

So maybe you can go to circuit simulation and tweak some +/- delta L to see how that affects overall results.

Yes, its when the active device is introduced.

I used the design kit of the HBT and not the Touchstone file.

So the difference is in the active device?

I suspect that some amount of inductance is included in the HBT S-parameter data -- whatever they used during measurement to create the ground path. If you now include the full ground path in EM, you might over-estimate inductance in that path (your layout's ground path L + HBT S2P measurement ground path L).

So maybe you can go to circuit simulation and tweak some +/- delta L to see how that affects overall results.

Yes