I wonder why cables with SMA connectors and SMA adapters become loose all by them selves. I have a stainless F/M SMA adapter (a connector saver) and a cable with SMA connector on my Nano VNA. Too often something is loose when I pick the VNA up.
SMA has polymer at the mating surface and polymers typically have 10x the thermal expansion of metals, so vibration momentarily unlocks the screw and polymer pushes it apart.
This is a general problem with threads. Any combination of thermal expansion/contraction and acoustic/vibration energy input that momentarily overcomes the static friction will loosen the connector. This is why torque wrenches, locknuts, loctite, and safety wire are a thing. If a connector hasn't been engineered for a certain vibration environment and installed/torqued to specification, one should not be surprised when it fails.
Someone should get Boeing in this thread, they evidently need to hear this too.
Did you use a torque wrench? It shouldn't loosen if you used a torque wrench There are a couple different torque settings for sma. Use .45 Nm for brass body connector and general lab use. Use 0.9 Nm for steel body connectors that may be subject to higher vibration or are semi permanent.
Someone should get Boeing in this thread, they evidently need to hear this too.
If you don't install the bolts you don't need to torque them.
Did you use a torque wrench? It shouldn't loosen if you used a torque wrench There are a couple different torque settings for sma. Use .45 Nm for brass body connector and general lab use. Use 0.9 Nm for steel body connectors that may be subject to higher vibration or are semi permanent.
No, I didn't. I tried to avoid over tightening the connectors. Probably part of the problem.
If they are so loose you are noticing it, I imagine you are getting some rotation of the cables. In turn, rotating the center pins inside one another.
The NanoVNA is nice in this regard. You can develop all sorts of bad habits and the worse that happens, you collect some bad data, prematurely damage a few parts and your out $50.
Did you use a torque wrench? It shouldn't loosen if you used a torque wrench There are a couple different torque settings for sma. Use .45 Nm for brass body connector and general lab use. Use 0.9 Nm for steel body connectors that may be subject to higher vibration or are semi permanent.
No, I didn't. I tried to avoid over tightening the connectors. Probably part of the problem.
You can't overtighten an SMA connector by hand, and an 0.45 Nm torque wrench won't damage good connectors when used properly.
However you can damage an SMA connector by mating it if there is debris in the threads. So make sure you connectors are clean. Then you should be able to screw them on by hand smoothly and easily until fully mated. Finally use an 0.45 Nm torque wrench to go an additional ~ 1/8 - 1/4 turn.
Doing this shouldn't damage the SMA part itself but it is possible to damage the solder joint of a PCB mount connector. Bulkhead connectors in circular cutouts can also have their retaining nut loosen when torquing or disconnecting SMA connectors. Many connectors have additional flats for a fixed wrench (often 5.5 mm) to provide counter torque on the female side of the connection and prevent these problems.
sma connectors are a good reason to replace your hand with clamps using biotechnology
more N and BNC connectors!
I recently repaired a Wavetek 3000 a.k.a. Motorola R-1010A. It must have 25 SMA connectors, some chassis mount, some on hardline and some as cable ends, even right angle style. They were mostly all loose. If you tighten them finger tight you still need either a proper torque wrench or simply use a wrench to turn them not more than another 1/4 turn. B.T.W., if you ever work on one of these units BE AWARE the connections on the bottom DO NOT include ground!!! Many of the modules are sourced with + and - 18 volts as well as + 7.5 volts. The ground connection is ONLY via the securing screw under the chassis and unreliably by the cables with the (loose) SMA connectors. If you disconnect the cables for testing AND remove the securing screw you have eliminated the ground reference and will almost surely damage the module when powering up!!! What a stupid design. The ground reference for the power supply regulators is not on the regulator board either but is carried by a single wire that is part of a multi-pin molex connector.