I’m running a rather small homelab and am hunting for a good UPS to help keep everything running smoothly. My top priorities are:
- Just enough battery life to keep things running until they can be shut down
- Compatible with open source software for monitoring and automated shutdown
Would I have better luck getting a used one and a new battery, or a brand new unit altogether? Anyone have one they don’t need anymore, on that note? 👀
Thanks for the advice!
This will depend wildly on what you are planning to put onto it
That said I have a 2 cyber power 825va (I think that’s the model, not sure). It’s like 450watts each iirc. I got them 2 for 1 for about $120 new. One has my server/nas, for which it’s grossly underpowered (maybe 7-10 minutes of runtime, at best), and one powers basically everything else critical in my rack (modem, switch, poe switch, etc) and powers that longer but still not as long (my primary switch is a business switch that was pulled from an ewaste place for nothing, like $15, but it’s got 48 gigabit ports and 5 10gb ports! But it also uses a shocking amount of power).
They work great for my use case. I live in a rural area with a horrendous power grid so I lose power about once every 6 weeks. As a result I have a (very pricey, can’t recommend unless you lose power a lot like me) whole house generator with automatic transfer switch. When power drops out the generator kicks on and switches the house over to generator power which takes about 45-90 seconds, so I really only need these to keep my gear on for that period. Beyond that it’s generator monitoring and if the fuel supply for that is running low network gear is shutdown to conserve power
In a perfect world where I was financially independent I would probably upgrade the server one to at least a 1500va to ensure my storage pool could fully stop and everything could shut down even if power was lost
But most ups will work with monitoring in one way or another. APC and cyberpower work with the apc daemon (probably others) which can easily be implemented into all kinds of software and has support in mac, Linux, windows
Determining battery life depends greatly on load. Rough calculation with power supplies of gear connected, better calculation with something like a kill-a-watt or multimeter and taking a reading for a little while under load, add it all together and add 20-30% to be safe. APC, cyberpower, etc have calculators for this
Buying used can be okay but you do have to be comfortable changing the battery. Additionally there is the risk of something being wrong with it of course, they’re not bulletproof. They’re usually pretty decent though, the bigger thing is that they’re just really expensive to ship, even without batteries
Unless you really know what you’re doing DO NOT measure current on a wall outlet with a multimeter. Specially not with a cheap one. That can pretty easily break your hardware, burn your house down and kill you, not necessarily in that order.
just get a clamp one?
That’s better, but you still need to have single wire to loop it around, which is not normally accessible. And at least in here the term ‘multimeter’ spesifically means one without a clamp, so you’d need to wire the multimeter in series with the load and that can be very dangerous if you don’t know what you’re doing.
Also, cheap ones often are not properly insulated nor rated for wall power (regardless of your voltage), so, again, if you don’t know what you are doing DO NOT measure current from a wall outlet with a multimeter.
You mean parallel right
Parallel won’t show current load of a device. Even a clamp type can be thought of as serial, it’s just picking up the EM field instead of actually carrying the current load across the device.
Something in parallel will be powered by the same source, with it’s current load independent of the other device.
(And yes, I had to think about it for a second, it’s not always immediately intuitive for me either.)
Pfff current I was brain-dead, yes.
There’s smart plugs that measure current, I have some Emporia plugs at home.
Series is more accurate.
Voltage is measured in parallel and current in series. You need both to calculate the power.