Hi,
I had a 7 port USB2 hub (DUB-H7, legacy version) give up the ghost, here. Always wanting to know "why", I pulled it apart.
Of course, as with most low cost consumer kit, very little inside! All the work is done by a standalone USB2 controller (ISP1521). Some FETs to switch power to each port. And an LDO to power the USB controller (3.3V). A sprinkling of discretes...
LDO was showing ~2V on the output with the normal 5V input. Lifting the output leg (now unloaded) saw the output rise to the expected 3.3V.
First thought is the device has gone into current limit (?) meaning the USB controller is toast. Presumably, something crept in on one of the USB connectors and ate its lunch. However, I don't know that for sure (yet) -- I'll drag out a milliammeter and try to get an idea what the *actual* load on the regulator was (prior to my lifting the leg).
The other possibility is LDO gave up the ghost for thermal reasons (?).
[The ISP1521 claims an Icc of 492mA (typ) with 7 HiSpeed devices connected. IIRC, the LDO (1735-33) is rated at 500mA. (Gee, if this was SOFTWARE, it would be called a *bug*! But, as it is HARDWARE, it's called "profit margin") :> ]I use several of these, here (most machines don't have enough USB ports and the ports are in inaccessible locations!). Recently, moved several of them off of their "switched (AC) outlets" to be "on" 24/7/365. So, its possible that things are just getting too warm from continuous use (?).
Regardless, I'd be interested in sorting out what the real cause of the failure may have been so I can prepare for "repeats" on the other units still in service (or, do something proactive to reduce their failure risk).
I figure I can throw another regulator on the next Digikey order "just to see" -- it's trivial to replace (unlike the USB controller!) and gives *some* information as to what's happened.
Thx,
--don