Strange sounds from Smart-UPS 1400

[haarp_apc]

Hello there,

I recently got my hand on a Smart-UPS and intend to use it for my machines It's this one:
http://www.apc.com/resource/include/techspec_index.cfm?base_sku=SU1400RMI2U
I use the 230V version (European). It may be intended for racks, but that doesn't mean I have to mount it, does it? 😉

Well, since the batteries in this thing were swollen and completely dead, I got a new battery tray and connected it. To make sure that the batteries work fine and that this old unit still works fine, I connected the UPS output to a Variac with a hair dryer connected to it, so I can slowly ramp up the load and see how the UPS handles. Now, on AC this thing works like a charm (except for the transformer humming very loudly), but as soon as I switch to the batteries, I get a strange behavior. When the Variac goes over approx. 120V (should've been about 150-200W load), the UPS makes strange vibration noises, besides the humming. Because it's difficult to describe, I simply recorded it.
See attached file, strange sounds starting at 10s. the humming is also very load, especially on battery backup mode, but for some reason the mic didn't record that.

What could this be? I'm scared that this thing might blow apart or otherwise harm the attached hardware. Could it be that UPS don't like inductive loads like a Variac?
Thanks in advance!
And please, no "get a new one" posts, because I barely could afford even the batteries for this thing.

Message was edited by: haarp

[Erasmus_apc]

These units can be used as rack or tower configurations. Since it is the international model it requires international H,N,G three-wire 230v input. You mentioned 120v, which it should never receive. If it does this at 230v international with x% amount of load...that could be an inverter issue or, possibly the load I suppose. Smart-UPS are generally fairly tolerable of all but the most noisy motor-driven loads, or loads with very large amounts of feedback due to inefficiency or malfunction.

[haarp_apc]

Hi,

The UPS was connected to 230V, like intended. But I plugged a variac (variable transformer) into the UPS output, which took 230V as input and output 120V to the load.

But I think I got it figured out. Some hair dryers seem to create A LOT of distortion because they use diodes when not running at full power. This was most probably the cause of the noises I heard from the UPS.


I've got another question. The UPS is running fine right now, but the remaining battery time is a bit strange. I'm using apcupsd to read that time. At 20% load, I've got about 30 minutes left (this itself is not even half of what the batteries should be able to endure) At 30% load it goes down to 16 mins and 8 mins at 40%. When I approach 50%, the runtime really starts to crumble, falling down to under 2 minutes. Theoretically, 2 times the load should merely half the runtime, not cut cut it in third or worse. Is the UPS really inefficient or why is this far from linear? Yes, I did a runtime calibration after changing the batts.

Message was edited by: haarp

[BillP]

Runtime isn't a linear equation. If you take a look at the runtime charts posted on the APC website, you'll see what I mean. But like you said, you should still be getting more runtime than that. Did you purchase APC batteries for replacements?

[haarp_apc]

You're right, but they're close to linear 😉

As for the batteries...I have to admit that I bought my own. I mean, come on, APC slaps a sticker on it and sells them for 4x the price of the regular batts. I don't have that kind of money.
I just checked the real runtime. On a very high load, the UPS will immediately start beeping in 1s intervals to let me know that the runtime will soon be depleted. However, the device still runs for minutes and minutes.... Somethings wrong with the runtime calculation here.

[haarp_apc]

I have it figured out. I followed the procedure on this website:
http://www.joeteck.com/apc-repair.htm
and now the runtime seems more sane again. 🙂

But I'm curious. What is the background of this modification? What does it actually do?

edit: I found this:
" Explanation: The battery constants give the battery status via their
life expectancy. If the battery ages and is “exhausted”, the constant is
overwritten. The management software calculates the runtime of the UPS
with these constants. If the battery is now replaced, a self test must
be done with the new batteries. Through it, the red battery replace
indicator goes out and the battery constants should be reset to the
standard settings. This does not occur in some cases. Therefore, the
constants must be reset manually. "
http://lists.alioth.debian.org/pipermail/nut-upsdev/2006-June/000938.html

Message was edited by: haarp

[BillP · Accepted solution]

Whatever works. 😉 Then again, I'd be careful with those instructions, since a) changing other settings in there can really screw up the UPS and b) there are different constants for each UPS. I wonder if they're out there on Google somewhere...