Using a Variac to simulate UPS battery

[Dannyx]

Good day folks. I noticed we have a beefy UPS just sitting there in the corner of the shop, but no batteries. It takes a lot of them judging by its size, but I don't know if it still works. Would it a dumb idea to stick the battery terminals on a variac to simulate a charged battery to see if it fires up ? Of course, I'd use a rectifier and a capacitor (or several) to smooth out the voltage as much as possible and also wouldn't plug the UPS itself into the mains - I'm not that dumb A lightbulb in series with the DC might also be a good idea.

Has anyone done this before or would something pop immediately ? A variac doesn't isolate from the mains, so it would be a shock hazard, but it's just for testing purposes. I imagine if I plug the UPS into the wall, I'd create a short, but assuming I don't and just try to power it off battery to see if the inverter is functional...Any other UPS troubleshooting tips would be greatly appreciated.

[eccerr0r]

Re: Using a Variac to simulate UPS battery

Would it a dumb idea to stick the battery terminals on a variac to simulate a charged battery to see if it fires up ? Yes

Isolation is still the issue... there had been another thread with related issues, in fact you may have started it... At least it seems a lot of UPS designs don't completely isolate the battery from the line and hence they give warnings not to work on batteries when plugged in.

Ultimately I think UPSes, if they don't come with working batteries are generally lost causes - you have to guess whether they work or not. Take your chance and get batteries or not. I think a vast majority of them still work if they're not of the high frequency type, and usually they get disposed of for the sole reason of... yep you guessed it, bad batteries.

If you have an isolated bench psu, you can use that, though I'm not sure my PSU could power my UPS for testing, it only goes to 18V and 15A. My UPS that no longer has any good batteries requires three 6V batteries, so my PSU won't cut it.

[Sametbey]

Re: Using a Variac to simulate UPS battery

You can estimate how many 12V 9A UPS batteries can be inside the UPS. When you plug the UPS into a 220V AC electrical outlet, you can measure how many volts it produces for charging. about how many volts is required (6 * 12V for 72V) You should make the variac output DC with diode and capacitor.

[sam_sam_sam]

Re: Using a Variac to simulate UPS battery

To me this might not be a good idea to do this because of what has been talked about earlier about isolation of Variac I would not mind coming up with something that could be used to check a UPS backup without using batteries to test one but I want to do this safely and blow up something or catch something on fire

[Dannyx]

Re: Using a Variac to simulate UPS battery

As I said in my original post: as long as you don't connect the UPS to mains at the same time, I don't see why isolation should be of concern when using a variac.

[Sametbey]

Re: Using a Variac to simulate UPS battery

As long as you don't connect the UPS to the mains at the same time, there is no problem. however be aware of the danger of electric shock. How many volts will you give from the variac? Are you going to use a rectifier (diode)?

[Dannyx]

Re: Using a Variac to simulate UPS battery

How many volts will you give from the variac? Are you going to use a rectifier (diode)?

I said I'd use a FULL BRIDGE RECTIFIER in my original post (and probably a cap+coil to smooth it out as much as possible, though the UPS itself is likely to have some beefy caps on its battery input anyway). I'm not crazy enough to feed it AC

As for the voltage, I'd adjust it to match the amount of batteries it's meant to work with.

[Sametbey]

Re: Using a Variac to simulate UPS battery

There are 3 12V 9A batteries in the 1000VA ups, I use. Input rectifier circuit failed due to high voltage. i fixed it

[Dannyx]

Re: Using a Variac to simulate UPS battery

Input rectifier circuit failed due to high voltage.

You mean the UPS's internal rectifier for mains, or have you already tried the Variac idea and killed a rectifier trying it ?

[Sametbey]

Re: Using a Variac to simulate UPS battery

İnternal rectifier for mains

[eccerr0r]

Re: Using a Variac to simulate UPS battery

testing UPS implies also testing switchover time..
You'll need massive supplemental filtering, filtering batteries is virtually unnecessary when using low or low enough esr batteries.

[Dannyx]

Re: Using a Variac to simulate UPS battery

You'll need massive supplemental filtering, filtering batteries is virtually unnecessary when using low or low enough esr batteries.

That's what worries me: it may not like the dirty DC coming out a simple bridge rectifier...no idea how I should filter it though. I think I should build some sort of test jig which is always assembled and ready to connect to any UPS for this purpose. Trouble is I think the values for the components (which I'm not entirely sure what they should be to begin with) would also have to change based on the voltage I set on the Variac...

[eccerr0r]

Re: Using a Variac to simulate UPS battery

Due to the low frequency of line voltage you'll probably need quite a bit of capacitance to "match" the voltage stability of a battery, probably 10millifarads would be a start?

My variac can only handle like 10 amps so technically my bench PSU surpasses the current capacity of my variac...

[shovenose]

Re: Using a Variac to simulate UPS battery

Use a bunch of car batteries

[eccerr0r]

Re: Using a Variac to simulate UPS battery

yeah that's what he should do... Or just one car battery...

I wish I could do the same with my 18V UPS.

[Dannyx]

Re: Using a Variac to simulate UPS battery

Use a bunch of car batteries

If I had a bunch of car batts just laying around, I could probably afford proper SLAs

This is intended solely for troubleshooting purposes when I get a big ol' hunk of an UPS with no or dead batteries and I need to test how much of it still works, namely the inverter.

I know this trick works with smaller units and I even tested myself up to 24v using regular power bricks, but when you start going above 36v you have a bit of a challenge to replicate the volts AND amps a ton of SLAs can deliver...

[budm]

Re: Using a Variac to simulate UPS battery

This was what I did in t he lab to test the UPS without having the batteries around, it is just enough to make the UPS run at low AC load.
I use old UPS transformer 120VAC side fed by the VARIAC, the low Voltage side (low Voltage high current winding which was used to be driven by the MOSFET's and get the DC from the batteries) is connected to high current bridge rectifier and lot of high capacitor bank.
for example if the 12V output is drawing 10A, the AC side fed by VARIAC will only draw a little over 1ACA. With is setup you have isolation provided by the old UPS transformer.
I save a lot of old transformer from old UPS to build high power DC power supplies. I am sure you have lots of broken UPS to play with.

[Dannyx]

Re: Using a Variac to simulate UPS battery

Small units are not the issue. Like I said, I was able to test up to 24v battery voltage by using a simple wall-wart, though, granted, with no load, as that would've probably overloaded it. The problem is testing larger units which can easily take 6 batteries in series or perhaps even more, resulting in a voltage which can't be easily replicated, hence where the variac idea came from.

The transformer idea is actually not bad, especially if it's a 230v-110v transformer of a decent size.

[redwire]

Re: Using a Variac to simulate UPS battery

The problem with your approach is the poor regulation of the variac and rectifier.
There's only a few volts difference between a "dead" SLA battery cutoff maybe 10.5VDC and good 12V and charging 14V.

You can variac dial in 13VDC, with the UPS plugged in it jumps to 14V to charge the imaginary battery, and when there is no mains the UPS starts running off the variac and the voltage sags too much.

What happened to your DPS5020 power supply build? I would use that, with an extra output diode in series so the UPS charging does not backfeed and blow up the DPS.

Even 1,000VA UPS needs at least 50A at 24V! When the variac power supply sags to 10V, it will just shut off.

I would test the UPS using a bench power supply, and then do the load test using batteries. It's just caps, mosfets, and relays that fail usually.

[shovenose]

Re: Using a Variac to simulate UPS battery

If I had a bunch of car batts just laying around, I could probably afford proper SLAs

This is intended solely for troubleshooting purposes when I get a big ol' hunk of an UPS with no or dead batteries and I need to test how much of it still works, namely the inverter.

I know this trick works with smaller units and I even tested myself up to 24v using regular power bricks, but when you start going above 36v you have a bit of a challenge to replicate the volts AND amps a ton of SLAs can deliver...

I don't know, I'm broke as heck and have at least four working car batteries floating around