how to tell good sma connectors from bad?

[mrf184]

The imported connectors from digikey are quite expensive, starting at around $2 each, and when I bought them they came with the label "made in china" so I wondered about whether they can be sourced elsewhere.

The SMA connectors on taobao come in two price ranges, ~$0.2 and ~$0.5, the former having fake gold plating and the latter real gold plating. The $0.5 ones pass acid test fine but I'm not so sure about its mechanical accuracy. I only noticed that they seem to use a bit more torque during mating than the imported connectors.

Is there a easy way to tell whether a connector has good mechanical accuracy? How would you go about measuring this? I'm talking about SMA female connectors so the pin height measurement doesn't really apply?

[ChristofferB]

I guess the "proper" way would be to get some mechanical drawings of a known good brand SMA connector - with tolerances, and then check those values against measured values on the cheap one with calipers/micrometer.

It's going to be difficult to measure the quality of the metal body.A few of the cheap connector failures I've seen have been the connector simply shattering due to poor material quality.

[Co6aka]

PIM tester. Smokes out the bad ones every time.

[Andrey_irk]

SMA connectors are standardised, so every one of them must meet the interface dimentions specifications. For example, take a look at MIL-STD-348. Most manufacturers offer special tools, which help to quickly check position of the center contact. This is vital, because if there is excessive protrusion of the center pin or radial misalignment, you may damage not only this connector, but also the one you mate it with.
Here is what maury offers:
https://www.maurymw.com/Precision/Connector_Gage_Kits.php

Also there are the quality of the thread, insertion and withdrawal force etc.
So, you can get a gage kit, thread gauges to chech the thread. Or you can use a measurement microscope and measure everything, but it'll take a lot of time. The quicker, but more expensive option is custom tools for every demention to control.

Anyway, in my opinion if you are concerned about the quality, then you'd better buy quality connectors. It will be cheaper, than to do the job for manufacturer.

 

[tkamiya]

I wonder if cheap SMA has actual GOLD plating on it.  What I have seen appear to be too yellow to be real gold.  If it is brass or some other material, corrosion may become an issue later on.

[Ice-Tea]

Well, in the olden days I would say 'weigh them'. That was untill I learned that some headphone manufacturers would actually add bars of aluminium to fool people like me 

Joking aside, for a lot of stuff a heavier unit tends to point to better quality.

[xaxaxa]

I wonder if cheap SMA has actual GOLD plating on it.  What I have seen appear to be too yellow to be real gold.  If it is brass or some other material, corrosion may become an issue later on.

gold plating is easy to test for with salt & vinegar; add salt to vinegar until saturated, then put a drop of that on the surface of the connector; let it dry and if it leaves a discolored spot the gold plating is fake; if it leaves only some dried salt the plating is likely real.

[CopperCone]

Well they might also have a specification for surface finish.

When working with metal there are a few options. These include a sample gauge, where you have the same material polished or finished to different levels and you hold that kit up against the metal to see which one matches visually and by way of touch, similar to trying to match paint colors in a department store where you have the printout card with the various scales on it, but since your surface finish scale will have the same material you can clean both surfaces with an approved solvent and touch them with your finger for nonoptical comparison.

You can use various hardness meters to determine surface hardness.. There are like 6 different ways of doing this.

I believe you measure the thickness of the coating by using eddy currents or ultrasonic aparatus.

You do a chemical test to crudely determine composition, you would need to look into taking scrapings and doing XRF on the sample (can be done cheap) to determine alloy composition. Xrf can be done at home but not sure if to the required precision to meet the spec.

You can also measure surface roughness with special expensive meters but keep in mind surface roughness is kind of like measuring noise, you basically band limit it by choosing span in an chosen area but you get mean and peak levels.. I.e.a polish on a scratched surface may make the average good with extreme outliers. It is a difficult measurement to really perform. To get an idea you need to loook how surface plates are calibrated.

Then you also have mechanical tollerances and any residual factors like materials stress and crystal structure which you would do an acid etch test using a metalurgical microscope to measure.

Unless you make it yourself quite a bit will go into fact checking chinese shit.

A high polish will increase corrosion resistance but a deep scratch will be some kind of impedance mismatch and nucleation point for processes like tarnish or coating flaking etc

[TheUnnamedNewbie]

There are two steps:

First, verify conformance to the most important tolerances like center conductor dimensions with a proper connector tester gauge. These are things like this: https://www.maurymw.com/Precision/Connector_Gage_Kits.php. You have to do this before going to the next step because if it fails this test you might end up destroying much, much more expensive connectors.

If they pass this (And only then) you get the connectors on a cable or board of high quality, and with a VNA measure it a number of times. This will give you 1) the performance of the actual connector in terms of insertion loss and return loss vs frequency and 2) the repeatability of said performance. From that you can figure out if your connector meets specification or not, or rather, if it is good enough for your purpose (does it really matter to you that a connector for your drone antenna has 10 dB instead of 25 dB return loss at 10 GHz?) .

A quick visual inspection by someone who has a lot of experience around various grades of connectors can often tell you a lot already (in my experience a lot of the cheap male connectors have just horrendous center pin dimensions, and you can easily see this by eye if you know what they are supposed to look like).

In addition, not all good connectors have gold plating. In fact, none of the top-of-the-line connectors in our lab (apart from the ones on the 3.5 mm cal kit) have gold plating.

I'm not sure most cheap connectors have gold plating. I think they are often made out of some form of brass, and just giving it a good polish might make it look gold.

In the end it all comes down to applications. What are you getting these connectors for, how much do you need, and who are you getting them from. And it is not just the manufacturer that matters. Even the big names will make connectors that are not great, because they are good enough for say WiFi, and cost 1/4th of what it would cost to make a connector that gives you the best performance at 12 GHz.