Why SMA connector impedance below 50 ohm? Thank you.

[jianke]

Why my SMA connector impedance below 50 ohm? Thank you.

Hi All,

I designed a SMA connector in HFSS to excite a microstrip line. It is constrcuted based on the Datasheet from its manufacturer (SouthWest Microwave.inc).

It is 50-ohm connector. However, HFSS simulation shows that its impedance is only 48.16 ohm. ADS-LineCalc's result is 48.13 ohm, with the same dimensions and dielectric constant (2.1). Why they are smaller than 50 ohm.

If you are available, can you help me check my HFSS model. The model and simulation results are attached. The datasheet is following link. Thank you very much.

https://mpd.southwestmicrowave.com/products/product.php?need=launchAccess&item=16
**broken link removed**

 

[albbg]

I think 48 ohm is not so far from 50 ohm. VSWR is 1.04

 

[jianke]

Hi albbg,

Thank you for your reply. The problem is I am not sure if 48 ohm is its real impedance, or there is something wrong with my HFSS model and let it below 50 ohm.

 

[Andryuha]

2 Ohm is a typical difference between simulations and experiment. Actually, it is very good. The connector is real and made of real components with real shape. and all those oxide layers, dielectric nonuniformities, etc. so, of course after design and simulation, the real technology is tweaked, which is not reflected in datasheets.

If you prepare to some experiment, tweak the dielectric to get the best results. simulate it with excessive precision first at designated frequency to be sure.

BTW. Some SMA connectors are really not high frequency. so those dielectric parameters can be actual on 10MHz, not 5 GHz. Check this. On 10 GHz and above just rare SMA connectors achieve passport performance. Use K/2.92 connectors.

Oh, and double-check your transmission line with rectangular waveport. See help for integration line and Z estimation method reference.

 

[jianke]

Hi Andryuha, Thanks so much for your reply. "tweak the dielectric to get the best results"? Can you explain how to do it?
In Datasheet, the dielectric used is “Dielectric: Virgin PTFE Fluorocarbon Per ASTM D1710”. In HFSS, I chose "Teflon" correspondingly whose dielectric constant (er) is 2.1.
With the same conductors' diameter, er has to be changed from 2.1 to 1.945, in order to change the connector impedance from 48.16 ohm to 50 ohm.

Whether my modification is proper and reliable? If not, what should I do? Thank you very much.
I will send to fabrication and measure my circuit, so I have to simulate more carefully.

 

[Andryuha]

I think, for some foam dielectric filling the epsilon can go nown a bit. It is possible. I don't see a big problem with this. Anyway, you should not care about the phase shift, because connector introduces random shift every time you turn the bolt.

 

[jianke]

Hi Andryuha, Thanks again for your reply. One final question is about what you said, "double-check your transmission line with rectangular waveport".

Why should we check it again with the rectangular waveport? And, what items or performance should we check, compared with the coax-waveport excitation?
Thank you!!! Really need your help.

 

[Andryuha]

I just wanted to say that the culprit can sit between the connectors. I don't trust microstrip calculators too much until I simulate the line with high precision on the designated substrate.