USB hub winks out when I turn *off* a non-computer connected device plugged into the same outlet

Factual Questions
1

TL;DR: I have a cat fountain and a USB hub plugged into the same power strip. When I turn the fountain off (and only off), it occasionally knocks the hub offline (the hub eventually auto-reconnects). I realize I can rewire the office space, but I’m curious as to why this is happening.

This is a weird problem. My two monitors (24" flat screens), USB hub and small speakers are plugged into an outlet strip. The outlet strip is plugged into a UPS via a 14 gauge extension cord (the cord has never felt warm). The UPS also powers my router, modem and VOIP modem on the battery side and two printers on the surge-only side. My computer and other items are connected to a different outlet (and different UPS). The circuit is 12 gauge wired to a 20amp circuit breaker that has never tripped.

When the problem arises, the speakers are off (either muted or nothing is playing) and both printers are asleep.

I might be able to work around this issue by doing some significant rewiring, but that wouldn’t answer the question of what’s going on.

One of the items plugged into the power strip is a cat fountain that sits on my desk (I work from home so cat management is a must). To make things easier, the fountain plugs into an extension cord with a remote switch. I have a few of these in the office (very helpful in avoiding climbing up shelves when restarting modems and routers), and have swapped out cords to make sure it’s not contributing to the issue. The fountain is powered by a low-voltage pump.

The switch is important to me because the pump isn’t quite silent, so I turn it on and off when a thirsty cat ambles by.

Most of the time, nothing happens when I switch it off. But every once in a handful of times, my powered USB hub loses its connection to my PC. I only have a mouse and keyboard connected, which means every once in a while both stop working for about a minute or so before the USB comes back on line. This is a minor frustration, one that I’m hoping has an easy fix.

Regardless of an easy fix or a rewiring, I’m curious about what’s going on. There’s never an issue when I turn the fountain on, only turning it off.

(My guess is that turning the fountain on causes a tiny drop in voltage or something on the line. Perhaps the UPS compensates? Then when I turn it on, there’s a slight spike in voltage further up the line from the UPS. By the time it detects it and reacts, something inside the USB hub freaks out and protects itself from too-high voltage. But that’s guessing more from imagination than knowledge.)

2

Turning it off causes an inductive “kick” due to the collapsing magnetic field in the motor.
That surge upsets the hub.

3

This is quite likely right.

And what you are doing is wrong in 2 ways:

  1. Don’t plug other things (especially motors) together with computers. (This is why~)
  2. Those cat waterers are designed for the flowing water to cool the motor. Frequent on-off cycles will cause it to burn out sooner.
4

Thanks. It sounds like my intuition was somewhat in the right direction (a confusing typo in it had “on” twice instead of “off” the second time). Besides learning more about DC Motors, I’ve been putting “inductive kickback” and similar to understand what’s actually happening — the real point of the OP, not just how to fix it. So far the cleanest description I’ve read is “When the energy source is disconnected from an inductor, the inductor tries to maintain current flow because current cannot change instantaneously through an inductor. An inductor converts current to voltage as it releases energy stored as magnetic flux to voltage” from a Quora answer, but I’ve a ways to go to fully appreciate what’s going on.

On the ‘fix it’ side of things, right now it’s plugged into a power strip that’s plugged into a UPS. If I figure out how to get it directly plugged into the surge side of the UPS, would that likely handle the issue? Or does that depend on how the UPS is wired?

I will tell the cats that they can have one flowing drink per day (two on weekends). Actually, I don’t cycle it on and off all that often, mostly because even with an easy switch it’s easier to let it run while one then another comes by. But good to know I shouldn’t turn it on and off in short succession.

5

Probably not.
The UPS handles brownouts & surges coming into the house from the power grid. But this surge is coming from the little pump/power supply for that cat waterer – that’s already inside the house, after the UPS, so it can’t be stopped by the UPS. The surge comes out of the cat waterer, through it’s plug into the power strip, along the strip and back up the cord into the hub, causing the problem. The surge never goes into the UPS at all – both those devices are connected together after the UPS,

To fix this, you need to plug the cat waterer into another power circuit that can only reach your hub by passing through the UPS (which will stop it). Like a different power outlet in the house, so the surge will come out of the waterer, go back through the power outlet to the main electrical panel of the house, nd then from there to the power outlet the UPS is pluggfed into, and into the UPS, which will stop it before it reaches your hub.

6

We traced the surge … it’s coming from inside the house!!!
I finally got around to re-working the cabling/wires in the office. The cat fountain is now plugged almost directly into the wall — but turning off the fountain is still randomly knocking the USB hub offline.

Since cats are involved, it’s a slightly splashy water fountain, it involves collapsing fields, and it sometimes does/does not happen, I’ve taken to referring to it as an issue with the “collapse of the wave function.” My apologies to physicist everywhere.

The outlet it’s now plugged into is a 20-amp circuit with four receptacles. There are two UPS devices plugged into the same circuit/set of outlets — one for my PC and one for my desk (two monitors, USB hub, speakers, etc.).

My best ill-informed guesses are that the surge from turning off the outlet is either A) travelling along the wires and causing a local disturbance to devices near those wires or B) that it’s going into the outlets and then back out of the outlets through to the other devices plugged into it.

If it’s the latter, am I correct in thinking that since the surge is now behind the UPS (which should block the surge), then it’s likely that the UPS (CyberPower CP685) no longer provides surge protection and should be replaced?

To complicate matters, that UPS needs its battery replaced. Devices plugged into it work fine, but if there is a momentary lack of power it cuts out and needs to be turned off/on. Could the dead battery be a reason the surge is still getting through and knocking the USB offline? (The battery is on order – at half the cost of a new UPS, so I’m not inclined to replace the UPS without reason).

If it’s neither of those cases, does that imply some sort of issue with the circuit or receptacles?

Thanks!

7

The motor is 12vdc and plugged in via a wall-wart. Will an inductive surge make it past that transformer? Bonus question: Isn’t there an inductive surge when the motor is turned on?

8

Adding a transformer to the equation makes everything more complicated. First of all, those “wall-wart” boxes my not be simple transformers - they may be a “switching” power supply, or may have a DC rectifier and capacitor filter in them. So, hard to know what it will do.

As for your second question: No. Inductors oppose change in current. So, when you apply a voltage to an inductor, the current will ramp up from zero until it reaches the point where the current is limited by the resistance in the windings. So, no voltage spikes are generated.
The reason that you get a “kick” when you turn OFF the voltage is because: inductors oppose changes in current. When you disconnect the voltage, the magnetic field will collapse and “try” to keep the current flowing through the inductor.

9

Isn’t the “surge” side of the UPS just a glorified circuit breaker? The motor may not be enough to trigger it, even if it affects your hub. Does it change when the hub is plugged into the battery side, assuming the battery holds ANY charge at all?

Also, does your hub absolutely NEED to be powered? If you just unplug the power cord from it, most computer peripherals (mice, keyboards, webcams, etc.) should still keep working. You only need that additional power if something is using a lot of power off USB, like if you’re trying to quick-charge a smartphone from it.

It looks like your UPS is the “single conversion, line-interactive” kind. It’s unclear to me whether it’s actually doing anything at all to your DC power unless it detects some anomaly in grid power… maybe beowulff can explain?

10

Yeah, “line-interactive” UPS’s don’t do any power conditioning (other than some simple buck/boost with a transformer). So, they are not great at preventing spikes from getting through. But, as the OP noted, the UPS is supposed to have surge suppression, which should take care of this, especially for something as small as a cat waterer. I wonder if the interference is being transmitted magnetically, and getting picked up by some USB cables. I know that I can reliably wake my desktop up by walking up to my desk and touching it in the winter…
So, the exact cause of the hub upset may be hard too track down - it could be on the line side or the USB side.

11

Really? That’s not just you accidentally bumping the mouse sensor or some such?

This will be my new explanation why “just hit it a few times” always seems to work. It’s not silly, it’s a manual pentadigital inductive kicksurge!

12

I have an old Steelcase metal desk, and it’s the static discharge that is waking the computer for sure.

13

I’m intrigued that something as small as a cat fountain can have such an effect. My entire computer setup is plugged into an outlet with a window air conditioner (it’s on its own circuit, and it’s the closest outlet that’s properly grounded), and I’ve have never had an issue, even before I got a UPS. Granted that’s not a DC induction motor, but the compressor startup current has to wreak havoc with that circuit.

Anyway, I’d consider getting a new/better hub. Whether the hub itself is the problem or the power supply, it seems to be a lot more fussy than it should be. Cheap no-name electronics from China tend to be weakest when it comes to power conditioning and protection. Lack of capacitors, no fuses, minimal circuitry, that sort of thing.