UPS output is 192 volts??

[chux0r]

Some of you might already be tired of my "I bought a crappy UPS from eBay" story... but maybe not.  Who knows.  Anyway, I thought it might be interesting to build a 48V turbine based around a UPS, since I'll deal with lower currents and the UPS will handle automatic switching from grid power, if that's what I want to do and such.

Anyway, so I bought this old junker 1.5kw unit on eBay, and after much coaxing and caressing I managed to get it going.  It even charged its long-dead batteries (eventually) and I'm using it right now on battery power on this PC to type this message.  Yes, if you're reading this, it means it lasted long enough.  In fact, I don't think it's quite fair to call it a junker because it's overengineered like crazy.  It looks like it was designed by the Russians to survive nuclear attack.

The reason I'm using my PC as a test load is because as I was load testing with incandencent lights, I noticed they were almost twice as bright on battery power than grid power.  I thought I might eventually have glass to clean up around here.  I put my voltmeter on the UPS output and it said 192V.  Now, I know there's some issue about RMS readings for sine, modified sine, and square waves... but this seems rediculous.  My PC and monitor have funky power supplies that say "input: 100-250V" so I don't think there's any problem there.  And it's working just fine.

So my question... uhh.... is this a known failure mode of an inverter to dump out 60% more voltage than it should?  It seems awfully weird, but then I'm new to this stuff.  Is there something that I might be able to... fix???  Yes it's an open-ended question, I suppose.  Where should I start looking?

[tecker]

  Put a scope on the output if you 've got one and see what the signal looks like .

no brand listed in your post so unless someone has dealt with that problem I'm going to

say unplug your pc from the 192 and check it out first. get the model number and go point to point with the inversion circuit to clean it up .

[Opera House]

It may be your meter.  Less expensive ones are calibrated for pure sine wave whih you inverter does not put out.  It may be looking at the ringing peaks.  Did you measure this with a load on it like a light bulb?  That would be the real test.

[chux0r]

Yup, I tested it with no load, with the lightbulb load, and with the PC load.  There's fluctuation, but it's only within a few volts.  Also, during the lightbulb test, as I mentioned, a 60W bulb looks bright as a 100+ watt bulb.  I can do this on line power while looking at the bulb, and then when I unplug the UPS it switches to battery and the light bulb becomes intensely brighter... so I'm not imagining it.

[chux0r]

Ha... you must mean "buy a scope."  Yes, that would be extremely cool.

The unit is a UPSONIC 1500.  A cold-war-relic-looking thing with a giant iron transformer.  One battery post is connected directly to the center tap, and the inverter circuit connects to both ends.  I have to assume that the inverter circuit alternately feeds each winding to convince the transformer to generate AC, and I further assume that the waveform coming out should be fairly nice as a result.  I suppose I can probe around on the inverter and see what I find, but I wouldn't know really what I'm looking for.

Then again, not knowing what I'm looking for is how I fixed it in the first place... so...  why not?

[tecker]

 I see haul it down to a electronics shop and get a look.  I think you'll find a few spikes too many the caps on the secondary aren't clipping the voltage .You might be able to shotgun( replace everything ) the inverter circuit and still have a cheap unit.

[chux0r]

Hm... a friend I work with looked at the caps and mentioned about old caps this style wearing out prematurely.  The thing weighs over 200+ lbs, so I'm disinclined to haul it anywhere... but I'll ask around and see if anyone can loan me a scope so I can check it out.  I think the excessive voltage is causing it to not handle surges very well, but it may also be because the batteries are fairly weak.  We'll see.

There's lots of regulators and whatnots with part numbers, so it should be fairly simple to swap everything out and see what happens.

I wonder if replacing everything in the inverter circuit would make it quieter!  Holy crap, the 60Hz hum... or buzz... or rather grinding... is incredible.

[Nando]

CHUXOR:

You have a very old UPS, it uses the power transformer ( heavy 60 Hz mother) for power generation, also it seems that the circuit that regulates the output is not controlling the output since the working voltage is from 42 to 57 volts ( 48 battery voltage range -- So the voltage feedback seems to be non-operational and the other circuit may try to keep it high.

The Semiconductors may be power transistors and not power MosFets (I may be wrong).

Too far to assist you in the trouble shooting

The wave form may alter the voltage reading if you do not have an RMS reading voltmeter, some use peak voltmeters and give the RMS equivalent.

Get a friend with a Oscilloscope -- YOU NEED IT.

By the way where are you located ?.

Regards

Nando

[whatsnext]

I have a 2500watt UPS and all I hear when it's on is the fan and the incesant beeping when it tells me that is 'doing it's job'.

John........

[tecker]

  A pound per volt it could double as a door stop .

[tecker]

  or maybe a concrete vibrator

[baggo]

This may be a silly question but are you sure that it's not a 240 volt model? I've had a look at the Upsonic website and all the models they appear to sell are 240 volt output( they are based in Australia ) The 192 volt output may be due to duff batteries. Just a thought,

John

[baggo]

Forget that. Just found your diary entry about the UPS and looked at the Ebay description. It's definitely 120 volt. Looks like the output regulation is at fault as suggested earlier. If it's a true sine wave output like the later versions it will be well worth trying to repair it.

John

[bob golding]

certainly wont do any harm to change the smoothing caps. big silver cans about 6 inches long and 4 inches dia or thereabouts. i have a 1000 watt sine wave inverter with 4 of them in. if it has been sitting around  for a while the caps will go bad. hope this helps.

bob golding