How should I interpret the impedance of an SMA connector? [closed]

Question

I am a mechanical engineer that has been tasked with verifying that a bunch of RF elements in a circuit are impedance matched. I have a source component with a 50 ohm output resistance that connects via SMA to another element with a 50 ohm input resistance.

The data sheet says that the connection is SMA 50 ohm. Forgive my ignorance, but is this referring

1. to strictly the characteristic impedance of a single metallic connector where the coax cable has an additional 50 ohm impedance, or

2. is the 50 ohm figure referring to the cable and the two connectors in total?

From what I have read today, if it's case 1, then that is an appropriate impedance match, but if it's case 2, then I would effectively have an output resistance of 100 ohms connected by 50 ohm coax to a 100 ohm input impedance and have an impedance mismatch. Thanks!

Answer 1

It's both the impedance of the connector alone, and the impedance of the connector plus cable, and the impedance of the connector plus cable plus another connector, and so on.

The characteristic impedance of a 50 Ω system is still 50 Ω no matter how many cables and connectors you put in it--as long as you terminate it appropriately. So with or without the cable, the connector's impedance is 50 Ω.

It would be difficult to measure that accurately, of course, since the connector is so small. You'd need to use pretty high-frequency test signals to see any impedance mismatch with no cable whatsoever.

---

Characteristic impedance seems to be one of those things people new to EE have a hard time wrapping their heads around, so I encourage you to research the matter more--I'm sure there are other questions on here that you would find educational. Other resources like wikipedia and your textbook of choice (The Art of Electronics is a popular one, but I don't recall how comprehensive its treatment of characteristic impedance is, and my copy is on a shelf at work right now) would likely be of help as well.

Answer 2

Transmission lines work like this- they have a characteristic impedance that is (ideally) a certain real number, in this case 50Ω.

When you connect two transmission lines (and here the connector is like a very short transmission line) together properly, you get one longer transmission line that is still 50Ω. This can go on to infinity, but since you probably don't have an infinitely long cable, at the end there will be something that looks (electrically) like a 50Ω resistor, taking the place of the rest of the infinite cable. Same thing at the other end. If those terminating impedances are not correct you'll get reflections off of one or both ends.

You can't easily measure the 50Ω of a connector or cable without some specialized (and often very expensive) equipment such as a TDR oscilloscope. If you connect a multimeter to any reasonable length of transmission line it will read open circuit.

The good news is that cables are marked, and if the connector is correctly designed and made from the correct materials, you can distinguish (say) 50Ω from 75Ω simply by making mechanical measurements and comparing with the datasheet. See Neil_UK's answer here

Answer 3

For cables and connectors, the 50 Ω impedance refers to the geometry of the connection. The idea is to maintain the same impedance all the way from source to destination.

For simple coaxial elements, it's a function of the ratio of outer and inner diameters, and the dielectric constant of the material the insulator is made from. To verify a connector, you either need to trust the manufacturer and supply chain (not damaged, right part number), or whip out a vernier caliper, measure, and compare with the datasheet.

If you can see tracks above ground on a board (microstrip), then for FR4 substrates, a 50 Ω track is roughly twice as wide as the substrate thickness, other substrate materials may have different ratios.

The actual electronic boxes and antennae that get connected are designed to generally present a 50 Ω impedance at their connector. Caveats - when operating, and often only over a limited frequency range. This generally cannot be measured with a multimeter, needing an RF network analyser to verify. Again, if they are working, and properly sourced, the datasheet should tell you what they are designed to do.

Answer 4

The SMA connector by itself does not have an impedance of 50Ω, but it is specified to work properly in a 50Ω transmission line. A transmission line essentially is the arrangement of connectors and cables here with a special property as becomes clear in the following.

A connector, as well as a cable have parasitic capacitance and inductance. Think of the body and the pin as the two plates of a capacitor, the same way with the shield and inner conductor of the coax cable. Also, lines of cabling build magnetic fields and thus have inductance. To transport RF-signals these parasitics are extremely nasty, they distort the signal. However, a coax cable has a property called wave impedance, and - somewhat fantastically to the uninitiated - all the parasitics disappear when the end is shorted via a resistor of the same value (a.k.a. "terminated"). So if you connect 50Ω across one end of "50Ω cable", as seen from the other end there are no parasitics anymore, whatever the length, and whatever the number of pluggable connections.

RF transmission lines are always constructed this way, and if you want to make them pluggable, you need connectors that do not disturb this feature. Suitable connectors are specified with the appropriate wave impedance, 50Ω for SMA connectors.