I need a UPS that will power a printer at idle (published load of 2.5W at 120VAC) for at least 8 hours, if practical.

I'm having trouble finding reliable information as to UPS's behavior at low loads. The ones I've tested shut down after a period and I'm suspecting they shut down due to the load being too small to keep the UPS engaged.

I considered setting up a load bank using C7 bulbs (Christmas lights) but the smallest commonly available bulb is 4 watts.

Is there a practical (not too costly or too time-consuming) way to build a load bank to simulate a load in 0.5 or 1.0 watt increments? Or any alternative to achieve the same?

Any suggestions would be appreciated.

 

I'm having trouble finding reliable information as to UPS's behavior at low loads. The ones I've tested shut down after a period and I'm suspecting they shut down due to the load being too small to keep the UPS engaged.

It would be strange for a UPS to shutdown because it required a minimum load. I have some UPS's that aren't being used, but I keep them plugged in to keep the batteries charged. I've never noticed any of the ones I've had on with nothing plugged into them turn themselves off.

Is there a practical (not too costly or too time-consuming) way to build a load bank to simulate a load in 0.5 or 1.0 watt increments? Or any alternative to achieve the same?

Using an LM317, powered by an appropriate adapter, with a fixed load resistor and varying the output voltage to get the required dissipation comes to mind.

 

It would be strange for a UPS to shutdown because it required a minimum load. I have some UPS's that aren't being used, but I keep them plugged in to keep the batteries charged. I've never noticed any of the ones I've had on with nothing plugged into them turn themselves off.

I don't think that the UPSs will actually turn off, but some of them do stop supplying battery power. APC claims (according to one support rep) that none of their units are designed as such but three out of the four APC models I have tested cut off their output when used with low loads.

I was testing with a small LED lamp (about 1 watt or less) and all three would run about 70 minutes and then cut their output. I've read in other forum threads where people report a run time of much less than that. Others claim never having observed such behavior.

There is a YouTube video that shows how to bypass the low-load shutdown circuit on one particular brand/model.

 

but some of them do stop supplying battery power

The battery isn't used unless line voltage drops or browns out.

There is a YouTube video that shows how to bypass the low-load shutdown circuit on one particular brand/model.

Once you modify the device, all certifications become void. Does it also tell you what to do if your modification causes a fire?

 

The battery isn't used unless line voltage drops or browns out.

Yes. Sorry that I omitted the information that I intend to use the UPS to supply power via its battery to equipment in the event of a power outage.

Once you modify the device, all certifications become void.

Yes, of course any warranty would be void and I was not endorsing such modification. I included that information as part of my point that apparently some UPSs - when disconnected from the mains - will stop supplying battery power when they don't sense a certain minimum load.

 

Yes, of course any warranty would be void and I was not endorsing such modification. I included that information as part of my point that apparently some UPSs - when disconnected from the mains - will stop supplying battery power when they don't sense a certain minimum load.

I don't believe most of the so-called experts on YouTube say. The material isn't vetted and they're either trying to monetize your time and/or they like the sound of their voice.

 

s there a practical (not too costly or too time-consuming) way to build a load bank to simulate a load in 0.5 or 1.0 watt increments?

Just built a bank of power resistors controlled by toggle switches to connect the resistors in parallel as needed.

 

Afew things to keep in mind. Lower cost UPS units output a MSW (Modified Sine Wave) rather than a TSW (True Sine Wave) as a basic wall outlet does. You want to make sure the UPS outputs what you need for your load. Every UPS I have worked with outputs power anytime there is an absence of mains power as long as the load does not exceed the maximum power rating of the UPS. How much power depends on the UPS rated power and how long depends on the battery.

While I see no reason to increase the load if you want to increase load, do as suggested and place a few power resistors acting as a load bank. Just off the cuff a 12 watt load at 120 VAC would be 120 / 12 = 10 Ohms making sure your load resistor(s) are rated about twice the power in watts than needed to reduce and distributed heat. Keep in mind as you increase load on the UPS you decrease the battery run time.

Again, I have never seen a UPS which had a load minimum, maximum yes but not min.

Ron

 

Just off the cuff a 12 watt load at 120 VAC would be 120 / 12 = 10 Ohms

You need a better cuff.
10 ohms will give a load of 12A or 1.44kW.
For 12W you need 0.1A or a 1.2kΩ load.

 

You need a better cuff.
10 ohms will give a load of 12A or 1.44kW.
For 12W you need 0.1A or a 1.2kΩ load.

Whoops, let me fix that. Thanks for the correction.

Thinking a bit further I have had printers which run on 120 VAC mains and some that run on a wall wort. If the thread starter has the latter they may be able to just use a battery backup and forgo the UPS.

Thanks Again
Ron

<EDIT>

You need a better cuff.
10 ohms will give a load of 12A or 1.44kW.
For 12W you need 0.1A or a 1.2kΩ load.

</EDIT>

 

virtually all small ups's use the same 12v7ah (7ah @ a 20hr rate as per standard) lead acid cells , smaller ones 500va and under usually only use 1, up to 1500va use 2 to 4 depending on the unit as a general rule .discharge rate derating curves for the batteries can be used to calculate expected run time by looking at the ups specs for its runtime at the rated load then use the battery curve to calculate the time given the load you want to use
http://www.eurosep.com/fichiers-joints/web-bpb12070y.pdf
this battery above will be close in spec to any consumer ups battery, you will just have to scale for the number of batteries that the ups uses

 

btw no consumer ups's i know of drops load because they are too light, they simply start inverting when they detect ac loss and keep running till ac is restored or their battery reaches the low voltage cutout threshold which is usually around 10.5V, some units this values is hard coded and some you can set the cutout but you shouldnt go lower as deep discharge will kill your lifecycle count
your led test terminated because the battery was drained, they all had the same runtime as they all have around the same efficiency and use the same batteries

 

ignoring inverter efficiencies:

as an example for runtime calculation lets use a 750va 450w tripp-lite ups
its rated at 10min runtime at 200W, 200w / 12v = 16.6Amp load on the battery not the ac load

A single battery is rated 12v7Ah (as most ups's use) at 20hrs (standard ah time) or 84 watts in 20hrs which is 4.2 watts per hour / 12volts = .35Amp load

The Amp-Hour rating or c-rating is a measure of the rate at which a battery is being charged or discharged. It is defined as the current through the battery divided by the current draw at which the battery would deliver its nominal rated capacity in one hour.

so back calculating the amp load of .35a / 7ah battery rating = .05ca . looking at the battery spec sheet discharge curve( i posted a link previous) a .05ca gives a 20hr runtime rate as expected

so for the ups it would be 16.6amps / 7ah battery rating = 2.37ca
from the battery curve a 2.37ca gives a runtime close to the rated 10minutes

so in your case you want to run a load of 2.5W / 12 = .208 Amps
.208 / 7 = .03 ca and since there isnt a curve lower than .05 we know it should last longer than 20 hr so any single battery ups will more than fill your 8hr requirement, probably closer to 30hrs, if the printer comes out of standby though thats a whole new kettle of fish...

 

In my experience any UPS with a 7AH battery has been junk. Those batteries are grossly overloaded (off the chart on a battery spec sheet) and just don't last very long (in run-time or service life). There was an APC Personal Powercell B (250 VA) at the office loaded only with a Linksys ethernet switch, and even that battery was dead in a couple of years.
I don't think you'll find any off-the-shelf UPS that will run 8 hours with a small load. The UPS wastes a non-trivial amount of power just making that high voltage "modified" sine wave. Get a small UPS with a dead battery and temporarily hook up a reasonably healthy car battery. Measure the actual current when the UPS is inverting, and calculate how many ampere-hours would be needed. Then obtain an appropriate sealed-lead-acid or LiFePO4 battery (or a wet deep-cycle battery if the location allows it) and hook it up externally.