this post was submitted on 21 Nov 2024
2 points (100.0% liked)

Self-hosting

136 readers
1 users here now

Hosting your own services. Preferably at home and on low-power or shared hardware.

Also check out:

founded 2 years ago
MODERATORS
 

First let me make sure it's clear that I am NOT trying to extend runtime by connecting two UPSs in series. That's been asked a million times on various forums, and is not what I'm trying to accomplish.

I've had 3 UPS units fail on me in the last 12-18 months, and I'm starting to wonder if it's the power flickers that are doing them in. My power rarely goes out for more than a minute or five, but before it does, it always violently flickers for a few seconds. Those flickers are hell on my unprotected equipment, and I'm wondering if that's what has caused my UPSs to die prematurely (the newest one barely lasted 5 months).

The old ones still function and still seem to do automatic voltage regulation, but none of them last more than 1-2 seconds once they switch to battery. I've tested the batteries, and they're fine; they were also all replaced about 9 months ago.

So, what I'm hoping is that the old ones can sit upstream of the new UPSs to take the brunt of any rapid brownout /surges and keep my new UPSs healthy. They're all pure sine wave and similarly rated.

Thoughts? Warnings/cautions?

top 9 comments
sorted by: hot top controversial new old
[–] vk6flab@lemmy.radio 8 points 2 days ago (1 children)

This happened to me and I called the power company who came and installed some measuring gear and left, came back a few days later, found the fault (in their network) and fixed it. No more flakey power.

[–] ptz@dubvee.org 2 points 2 days ago (1 children)

Power's generally pretty stable. It's only right before an outage that it goes all poltergeist. Could still be an issue on their end, but I'm afraid they're going to tell me "that's just the way it is".

May still call them as my old house didn't have this issue before every outage.

[–] vk6flab@lemmy.radio 6 points 2 days ago

In my experience, outages are exceptions that need to be checked. In a different location I was subject to all manner of shenanigans when I contacted the power company I was asked to stand on my porch and report where a wire crossed the road. It turned out that their map was out of date and I was on the wrong side of a connection. Once they updated their map, no more issues.

In my current location there was an issue with one of the phases in the street. After discovering this they moved me to another phase.

Then I reported another issue a few years later, which turned out to be a fault in the substation.

I don't know in which country you live, but here in Western Australia I've been happy with my local power company. They're responsive, fix faults faster than they say they will and come out if there is an actual problem.

[–] BlackEco@lemmy.blackeco.com 5 points 2 days ago (1 children)

Both Eaton and APC advice against daisy chaining, claiming it would cause extraneous strain to the first UPS in the chain, possibly overloading it.

[–] ptz@dubvee.org 2 points 2 days ago

Yeah, I read similar. And as general advice, that's probably solid.

In my case, both old and new are the same rating and I limit the loads to half of what they're rated for (which is why I have 3 UPSs instead of 2). So overloading them isn't high on my concern list. Was mostly just looking out for any kind of hidden gotchas.

The only issue I've thought of is at least one of my old ones no longer automatically kicks back on when power is restored. That would be a bit of a deal breaker.

Basically I just want to use them as power conditioners since they have AVR and surge protection on the battery output versus just surge protection on the "surge only" outlets.

[–] schizo@forum.uncomfortable.business 4 points 2 days ago (1 children)

This is kinda old information, but my understanding was that there were 3 issues with dasiy-chained UPSes.

The first is that you're potentially going to cause a ground loop, which is not healthy for the life of anything plugged into those UPSes.

The second is that there's a potential for a voltage droop going through from the first to second UPS, which means the UPSes will flap constantly and screw their batteries up, though I'd be shocked if that was necessarily still true for modern high-quality units.

And of course, the UPS itself won't be outputting a proper sinewave when it's on battery, which means your 2nd UPS in the chain will freak out (though again, maybe modern ones don't have that limitation).

[–] ptz@dubvee.org 2 points 2 days ago (1 children)

The first is that you're potentially going to cause a ground loop, which is not healthy for the life of anything plugged into those UPSes.

Wasn't aware of that, but will look into it. Good to know

The other two, yeah, they're pretty much non-issues with modern pure sine wave UPSs. Mine both have AVR and will, within a certain range, just boost an undervoltage at the expense of a little extra current draw (or also the opposite for overvoltages). Since I'm limiting my loads to 1/2 of the rated value, even that extra current should be well within spec.

[–] poVoq@slrpnk.net 5 points 2 days ago (1 children)

I am running a UPS after a solar battery system as it doesn't switch fast enough when the power cuts (despite having a "UPS" mode) and I can tell that the lack of a proper sinewave is still a problem. It seems to work, but the real UPS makes some very strange humming noises when running that way.

[–] ptz@dubvee.org 1 points 1 day ago

Good to know (I've got similar plans to setup a solar battery system)

Though I do have a sinewave inverter I run from my big 48v 30 Ah ebike battery during shorter outages (9 times out of 10, that will get me through outages without having to start the generator). I've plugged my UPSs into it, and they seem fine with it (no weird humming or constantly going on/off battery). I'm a little more than 90% positive that's not what's caused them to fail because I've done that for a lot longer than these have been having issues lol.