Inconsistent Connection verification with end of run micro usb cable questions

[NeverTheEasyFix]

I am trying to replace a long, expensive 40-foot usb3 cable ending in a micro3 upangle connection with a single extension cable and a roughly 30-foot extension/repeater cable and a short 6-foot end of run cable. It gets damaged all the time by operators (these go to scanning cameras) and it's much easier to replace the exposed bit rather than pulling a whole new cable at a cost of almost $200 a piece.

I have ordered a number of extension cables (of various types - repeaters, ones with an auxiliary plug for more power, and just generic extensions) and haven't had any issues reading 5V at the end.
I could only find 1 or two vendors that offered an up-angle usb3 micro cable (it has to be in the up angle configuration for how they mount into the camera and housing)

I picked up a passmark PM065 to do some testing and I'm lost as to the outcome.

PC -> 6-ft up-angle micro -> PM065 = cable tests out good
PC -> any extension variant -> 6-ft up-angle micro -> PM065 = Only shows 'USB3 enumerated' and pm065 doesn't register on usb3test. If I plug in partially, will register and run usb2
PC -> Most extension variants -> usb micro that came with PM065 -> PM065 = cable tests out good (One brand of powered cable just shows "usb 3enumerated" and doesn't register on usb3test at all - but if I plug it in partially, it will register and operate at usb2)

I get the same results trying to use the cameras with the various cable arrangements. They will register on any extension with the passmark cable, or on the 6-ft cable if directly connected.

Is there a reason the micro cable that comes with the pm065 would work on just about all the extensions but the cables I'm trying to use for end of run would only work when they are the sole connection between the PC port and the pm065?

Is there a max cable length / delay I'm right on the edge of that the pm065 or usb3 in general is not liking?
The one cable the pm065 does not work on with its own cable is a 40-foot powered cable (although windows dings like it sees a device connected, it never shows up).

Thanks

[David (PassMark)]

USB2 is slower and the max cable length is around 5m (in practice I am sure many lower quality cables wouldn't work this far however). But if the cable is good, the limit is 5m.

USB3 is faster and the limit is more around 3m, if the cable is good. It is easy to make a bad cable (with no shielding, or the wrong amount of twisting the data lines). Some USB3 extension cables might not even join all the wires. ie. the might only be USB2 extenders and leave the USB3 data lines unconnected. Type-C cables are even harder to make well.

My guess is that your 6-ft up-angle micro cable is marginal.

There are many different type of powered extenders. e.g. optical, ethernet & repeaters. It isn't surprising the behavior is different.

Joining cables is bad, generally speaking. Independence differences and reflections ruin the signal quality (plus the problems of more resistance, less voltage, signal attenuation, and potential interference that you would get with a long cable as well). You can also have latency issues with long cables, or extenders that add delay.

[NeverTheEasyFix]

I don't have any issues with just the 6-ft cable. I wasn't sure if maybe I was reaching a particular length of signal degradation or something. And it only registers as USB 2 if I partially insert it (so the USB 3 pins don't touch). It almost seemed like either pins were crossed so it wouldn't actually recognize USB3 but because it saw some pins it wouldn't default to 2. But oddly I had no issues getting significant speeds using a thumb drive for testing. So I wasn't sure if there was some internal value or parameter in the passmark tester that might get triggered by a particular amount of delay or interference.

Unfortunately, with how often this camera gets knocked or inattentive operators step up on the equipment and catch the cable and break it, I'm sort of being forced into a better solution than having to re-run a $200 cable every other week.

I did get a response from the engineering team from the company that makes the 6-ft cable, since they engineer most of this stuff for camera support. They said in their experience, for some reason long repeaters or powered extensions tend to not work on their own, but running a 10-ft repeater followed by a passive extension has been fine to significantly longer lengths. I have some more cables on the way that I'll hopefully be able to test out.

[David (PassMark)]

USB3 specs require a voltage (on VBUS wire) to be 4.45V to 5.25V and SuperSpeed signaling is typically 400 mV to 1200 mV peak-to-peak (differential).

If the voltage at the USB source port was just 4.8V, then you can have a 0.4V drop over a 3m cable and be out of spec. (28AWG, 3M cable, 1A current).
And of course the chips in different devices will have different tolerances for their minimum voltages. Some might not even deal with the required 4.45V. So when you are on the edge of the spec working / not working, it can be expected that different device behave differently.