Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Deviation between simulation and measurement of matching circuits

Status
Not open for further replies.

ueckid

Member level 1
Joined
Mar 22, 2017
Messages
39
Helped
2
Reputation
4
Reaction score
3
Trophy points
8
Location
Osaka, Japan
Activity points
411
Hi all,

I am now designing a power amplifier (PA) at 2.4 GHz using microstrip lines as input and output matching networks on a Rogers 4350B substrate. This PA was designed using the harmonic termination technique. Simulation substrate parameters of this substrate are given in its datasheet. Input matching network (IMN) was treated to 2f0 while output matching network (OMN) was treated up to 3f0 using open-circuited stubs. The simulated results were very good based on EM simulation.

However, I am now facing an issue of large differences in impedances between EM simulation and measurements at f0, 2f0 and 3f0 for both IMN and OMN. We measured impedances of the fabricated IMN and OMN using a VNA from Keysight (DC-26.5 GHz) with input power of -30 dBm, IF bandwidth of 20 Hz. VNA was calibrated using a full 2-port mechanical SOLT calibration kit. SMA connector was not de-embedded in the measurement. The differences become more significant at high frequencies range.

As you see in the attached file, magnitude of transmission and reflection coefficients among Schematic, EM and Measurement results are not so deviated. However, the vector impedances on Smith chart among these results are terribly deviated. As you see that Schematic and EM results agree well but the measurement results are deviated. I think that SMA connectors don't play a major role for this deviation.

Attachments are descriptions of my design.

Could anyone suggest solutions for my issue?

Thanks alot in advance!
Slide1.JPG
Slide3.JPG
Slide5.JPG
 

There might be few variable that impact consistency of the electrical response.
Manufacturing tolerances
Component Tolerances
Measurement Errors

The results generally look not too bad but I believe there are calibration and measurement errors.
 
  • Like
Reactions: ueckid

    ueckid

    Points: 2
    Helpful Answer Positive Rating
Measurement Errors

To be more specific: different reference plane, i.e. extra line length included in measurement. Test your measurement response whith a known device (open or short) and see what you get in Smith chart.
 
  • Like
Reactions: ueckid

    ueckid

    Points: 2
    Helpful Answer Positive Rating
Thanks a lot to @BigBoss and volker@muehlhaus for your kind replies and helpful suggestions,

In the case of measurement errors, do you think this is due to additional electrical length of SMA connector because I didn't de-embedd SMA connector in the measurement?
If so, could you please suggest an effective method to de-embedd the SMA connectors from measured data?

Thank you in advance!
 

Perform calibration with appropriate reference plane or use port extension.
 
  • Like
Reactions: ueckid

    ueckid

    Points: 2
    Helpful Answer Positive Rating
The influence of SMA connector can be neglected @ 2.4GHz.
It has an influence over 10GHz and beyond..
 
  • Like
Reactions: ueckid

    ueckid

    Points: 2
    Helpful Answer Positive Rating
Perform calibration with appropriate reference plane or use port extension.

Thanks for your recommendation but I have calibrated at the end of the coaxial cables (reference plane) using a full 2-port SOLT calibration kit.

- - - Updated - - -

The influence of SMA connector can be neglected @ 2.4GHz.
It has an influence over 10GHz and beyond..

Thanks for your kind reply.
Do you mean the cause should be fabricating tolerances of microstrip lines instead of SMA connector?

I think I have made an appropriate cable calibration using the standard calibration kit. I don't where the measurement errors come from!? Could you please kindly suggest some detailed causes of my issue?

Thanks in advance!
 

Did you include the SMA connector in your EM simulation? If not, die you try to compensate it's electrical length by applying port extension?

I believe however that your initial conclusion "reflection and transmission - not so big deviation, impedance - terrible deviation" is incorrect. You overlook that the former measurements are presented in logarithmic dB scale while the latter uses linear scaling. Reflection is just another way to show impedance.
 
  • Like
Reactions: ueckid

    ueckid

    Points: 2
    Helpful Answer Positive Rating
Did you include the SMA connector in your EM simulation? If not, die you try to compensate it's electrical length by applying port extension?

I believe however that your initial conclusion "reflection and transmission - not so big deviation, impedance - terrible deviation" is incorrect. You overlook that the former measurements are presented in logarithmic dB scale while the latter uses linear scaling. Reflection is just another way to show impedance.

Thank you for your kind reply!
I didn't include SMA connector in the EM simulation and I have not yet compensated it in measurement using port extension method. I am trying to do so. However I wonder that if SMA connector plays a main role for this deviation at 2.4 GHz?

For the impedance deviation, I think that this is due to the phase shift or phase difference between EM simulation and measurement (this may be caused by an additional electrical length somewhere which I have not yet figured out) while the magnitude of reflection and transmission between EM simulation and measurement is not so different as shown in the attached pictures.
 

For the impedance deviation, I think that this is due to the phase shift or phase difference between EM simulation and measurement.

You just answered yourself to your initial question...
 
  • Like
Reactions: ueckid

    ueckid

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top