analyzing the affect of SMA

[jeanjean]

i use SMA as my rf output

there is some mismath between 50Ω stripline and sma

especially at high frequency


i want to know how to analyse it

thanks

 

[vfone]

The SMA connector is designed to achieve coplanar matching with the outer conductor. In other words, the center contact, dielectric, and outer conductor all lay in the same plane.
The dissimilar coaxial line sizes of the SMA female connectors produce a capacitive discontinuity at the reference plane that increases the VSWR at high frequencies (up to 1.2 at 20GHz).

Here is a trick to fix this discontinuity at very high frequencies.

Offsetting the female contact on the board can provide an inductive compensation for this capacitance. The resulting inductive discontinuity would create a mismatch in the "standard" junction, improving the VSWR.

 

[borich_03]

For analysing mismatching between SMA-socket and stripline you can use 3D EM SOLVER. For example - HFSS, CST and others. Then you can use results for next simulation and tuning.

 

[sergio mariotti]

For my opinion, the transition COAX(structure) to MICROSTRIP(structure) may be analized with the help of a 3D simulator.
But it isn't an easy problem. Lot of articles and AN may be found on this topic.
Look on historical papers of ARFTG.

Different the question about the transition between SMA socket to SMA plug.
This cannot be simulated because no one know the dimension of the gap.
The gap is know as mechanical uncertainty only.
Also the gap depend on both the plug and the socket.

I suggest a sperimental way rather an analitical way. Use a VNA with time domain with precision, compatible SMA connectors. Namely the 3.5mm.

 

[softstone]

draw the SMA in the HFSS, then simulate it.
the result is very exact usually.

 

[sergio mariotti]

draw the SMA in the HFSS, then simulate it.
the result is very exact usually.

Of course HFSS may simulate almost every closed structure, so the coax onnector also.
But i'm still convinced that the problem with SMA coax connectors is related to the mechanical tolerances that became uncertainties. This the reason because the SMA connector cannot be accurately simulated. Only precision connectors can be simulated (7mm, 3.5mm, 2.92mm, 2.4mm, ...)

 

[wjwreck]

yes ,it is easy to simulate the connection between SMA and stripline ,
but how can we simulate the mismatch between sma-k and sma-j?
i had tried but fail.