Alcun Atirutan BBS

Sooo I just had a UPS do exactly the opposite of what it's supposed to do.

I have a small CyberPower CP375JP that I use for networking equipment. Basically just fiber CPE, switch, and WiFi AP.

I just had a tiny power glitch. Just a fraction of a second, the kind of thing many PSUs survive on their own. This caused all my UPSes, including this one, to briefly go into discharge of course. It was so brief that none of the others even had a chance to beep at me.

Unfortunately, the battery in this UPS was dead. It has never beeped at me before, so clearly it doesn't do battery self-tests. Bad.

Unfortunately unfortunately, the battery was *so* dead that instead of just discharging very quickly, it immediately went into fault mode. A fault mode where it just screamed at me, *with the output perpetually off*, even as power had come back on within a fraction of a second. No fallback to bypass mode. WTF.

So yeah, don't buy the CP375. It fails at the most basic duties of a UPS.

Guess I'm in the market for a small UPS that isn't total junk now...

@marcan APC's quality has become dangerously bad the last few years, ever since they got bought out by schneider electric.
https://www.youtube.com/watch?v=dIEM2bG8mOQ

Omron BZ35LT2 seems to be roughly the same form factor and actually claims automatic battery tests in the datasheet, so maybe I'll go for that.

Edit: APC BE425M-JP is smaller (fits better where I want it) and also claims auto tests in at least some docs, plus it's cheaper. Let's try that.

@developing_agent oh is that when they got so bad! Interesting. They used to be so good, then a few years ago I bought a few and gradually realised how bad they are. Irreplaceable batteries?! One that has a well-known “sorta goes goes weird” situation… very disappointing. Another that just fails to work sometimes, despite a fresh battery.

I also have a small Eaton UPS which I really like. Simple, standard USB interface. Much nicer price! I’m hopeful it’ll prove more reliable.

@j I was looking at eaton ones a while back instead of APC, but it's going to have to wait until I have more cash on hand. The ones with true sine wave at the wattage I need ran like $600. Most of the computers I own cost less than that. 😂

@developing_agent @marcan I wonder what good alternatives there are. The last time I took something apart from Triplite it was full of cheap aluminum wiring. Are there any good makers of UPSes now, or do we need to roll our own like HBPowerWall?

@developing_agent @marcan Hah! I hadn't even watched the video by the time I said that, and what do you know, ALUMINUM WIRING IN THE TRANSFORMER!

@kazriko @marcan

Eaton make similar small personal UPSs, and so far their reputation is intact. The units cost 2-3x as much though.

@kazriko @developing_agent I don't think cheap aluminum wiring in the transformer is really a problem, as long as it does the job and is rated for the load.

On the other hand, those solder crimps... yeah, that is bad, very bad.

@marcan @developing_agent Aluminum wiring is still a really bad idea, they used to be used in cheap mobile homes before 1974 before they were banned, mostly because they have high resistance and overheat.

@kazriko @developing_agent Resistance is inversely proportional to cross section. Aluminum wiring is used all the time in industrial applications like high voltage power transmission. As long as the gauge is correct, it's fine. Any wire of too small a gauge for the material will have high resistance and overheat.

On the other hand, copper-clad aluminum (often seen in cheap Chinese power cables) is dangerous because it looks like copper when it isn't, and most often is passed off as pure copper which will result in underspeccing the gauge and therefore dangerous use. But that's largely a problem for "standard" wiring, especially molded cables where the whole insides are invisible.

If you're a transformer manufacturer and you want to use aluminum and you know how to solder it properly, and it meets the specs you need, there's no problem.

@psykose @developing_agent @marcan Mostly because it's a sign that something is being made super cheaply, especially when it only has maybe a meter of wires, and it's a multi-hundred dollar power distribution bar...

@psykose @developing_agent @marcan I was Extremely disappointed when I took that thing apart and found that it was cobbled together from cheap consumer-grade wall sockets and aluminum wiring. It made me want to get commercial grade wall sockets and copper wire and rebuild the whole thing, just keeping the outer shell of the rack power distribution box..
replies
0
announces
0
likes
0

@psykose @kazriko @marcan There's a lot more to it than just highschool lessons on resistivity. One of the things I've personally had to deal with is the nightmare that results when you mix copper and aluminum wiring. In contact the two cause severe corrosion over time and eventually housefires. Copper is used throughout modern electronics and without appropriate care using aluminum near it WILL cause corrosion and electrical faults.

@psykose @kazriko @marcan This happens far more often than people think with aluminum wiring and for the most part it's ignored. It can be as simple as scratching through the gold plating on a PCB to the copper beneath when attaching an aluminum cable to a lug. Personally had a microswitch in a 3D printer made by printerbot (USA) where it's been completely consumed by corrosion, simply because they used aluminum stranded wire and soldered it directly to a microswitch with copper legs.

@psykose @kazriko @marcan If APC are willing to cheap out on aluminum windings, not take the care required (not even do a real crimp and hope the solder sticks!) and nearly start an arc fire inside this guy's home?

Banned. For Life.

@marcan @kazriko @developing_agent I wonder if the copper-clad cables would actually be fine for AC given the skin effect. I'm not sure how strong the skin effect is for 50-60hz though.

@Nefsen402
@marcan

The skin effect isn't relevant. This isn't about resistance. It's aluminum's mechanical properties, and the total lack of quality control to catch an intentional and dangerous design decision.

This wasn't a simple manufacturing defect. The transformer maker or APC *decided* to use aluminium, then hoped and prayed that their solder blob would stick or their woeful twist job would hold together. Either way, APC quality control never caught it before nearly staring a fire.

@Nefsen402 @marcan The only thing positive I can say about it is they didn't use copper on the wingdings and aluminium for the lead, or vice-versa. The combination of the two has been known for decades to cause housefires when precautions aren't taken to prevent corrosion.

Given that APC/the transformer maker thought a twist and a solder blob would properly whet and secure aluminium, I doubt they'd apply the required paste/spacers/whatever to prevent corrosion.