The connector lifetimes are tested by some manufacturers. When I was working on USB (electrical) testing at Microsoft we had test jigs for testing connectors to failure and I can vouch for Microsoft's connectors having lifetimes exceeding 25,000 insertions.
We would also test how many miles a mouse would travel before the glides would wear out.
we had test jigs for testing connectors to failure[...] 25,000 insertions
That's likely to be the exact reason why the reliability specs don't line up with reality: your test jigs are almost certainly going to be inserting the plugs directly with exact precision, whereas in practice people are going to just "jam it in" and apply significant side-loads as well as misalignment and levering forces.
It's not hard to design a connector that'll last 25k precisely aligned insertions by a test jig but will consistently fail after a few cycles of being mated by typical "hamfisted" users.
>That's likely to be the exact reason why the reliability specs don't line up with reality: your test jigs are almost certainly going to be inserting the plugs directly with exact precision, whereas in practice people are going to just "jam it in" and apply significant side-loads as well as misalignment and levering forces.
as someone who was involved with endurance/reliability testing of other physical products, I think you'd be surprised.
Randomized jitter and movement patterns are often replicated and considered.
I agree with the parent's comment. I also worked with automated test setups and did human tests. Nothing compares to a human tester. We tested small remote controlls and it was amazing to see that a human hand can inflict more damage to the product that an automated test setup. (the remote was given to different persons just to push the buttons until we reached the desired number of button presses).
Because you can search for what connector testing machines look like, and they all look like they'll just destroy the connector if it isn't aligned correctly.
This is true. A test system will only apply force from one or more predefined directions. A user, on the other side will not take care that he inserts the connector at 45 degrees +/- 3 degrees.
I guess most damage actually does not happen during insertion itself but when someone trips over the cable while the plug is in the socket. Simply because of its bigger size that force is then applied to a bigger grip surface and thus lower per square millimeters on USB-A while the forces will be much bigger with C
The fragility comes from how easy it is to damage them by handling them incorrectly. The fact that you can insert them properly 25000 times without damage doesn't change this.
How can a large usb A port be weaker than a tiny usb c port?
If you attach a keyboard and throw it out of the window the usb A port will remain ok whereas the usb c port instantly gets damaged.
My phone is 4 years old and it's type-c port completely destroyed for the last year, it barely connects (and cables are good, verified on the newer hosts). Even we assume I've connected it 2 times per each day every day, ot would mean about 2000-3000 insertions tops. This is an order of magnitude worse than that fantasy 25k number.
Another phone with 1000-2000 insertions over life time is also showing port degradation.
And I charge my phones in 99% of cases at home, stationary and undisturbed.
Then the tests must be faulty. At least, they don't reflect the real-world experience of me and my friends. USB-C has a much higher failure rate for us.
We would also test how many miles a mouse would travel before the glides would wear out.