On Sat, 24 Sep 2011 05:00:05 +0100, "Johny B Good"
wrote:
>On Sat, 24 Sep 2011 03:31:04 +0100, wrote:
>
>> On Fri, 23 Sep 2011 20:33:56 -0400, clare@snyder.on.ca wrote:
>>
>>> On Fri, 23 Sep 2011 23:52:10 +0100, "Johny B Good"
>>> wrote:
>>>
>>>> On Fri, 23 Sep 2011 20:10:30 +0100, wrote:
>>>>
>>>>> On Fri, 23 Sep 2011 08:06:35 -0400, "vaughn"
>>>>> wrote:
>>>>>
>>>>>>
>>>>>> "Johny B Good" wrote in message
>>>>>> news:op.v18u4so9kd9x7s@fred...
>>>>>> ...>especially as it was supplying a second
>>>>>>> level of UPS protection from the protected mains sockets powered by
>>>>>>> my
>>>>>>> SmartUPS 2000 (a 2KVA 1500W unit) that I keep in the basement.
>>>>>>
>>>>>> This is a good point. I learned the hard way that feeding one UPS
>>>>>> from
>>>>>> another
>>>>>> is a very bad idea. Think of every UPS not only as protection, but
>>>>>> as a
>>>>>> potential POINT OF FAILURE. To reduce the inevitable failures,
>>>>>> simply
>>>>>> reduce
>>>>>> the number of UPSs. I suffered more than one unnecessary server
>>>>>> outage
>>>>>> before
>>>>>> the lesson sunk in. One layer of UPS "protection" is plenty!
>>>>>>
>>>>>> Vaughn
>>>>>>
>>>>> Some UPS units will complain loudly about being fed by another one too
>>>>> - some don't like squarewave input.
>>>>
>>>> And, with good reason on account of the relatively large capacitive
>>>> loading they may present on their mains input, although it's more
>>>> likely
>>>> that the cheap 'stepped' sinewave inverter driving into such a load
>>>> will
>>>> be the first to succumb.
>>>>
>>>> As long as any such cheap UPSes (such as the BackUPS 500 and the
>>>> Emmerson
>>>> Accupower 30) are never used to feed another UPS and only true sine
>>>> wave
>>>> output UPSes (such as the Upsonic UPS600 and the APC SmartUPS 700 and
>>>> 2000
>>>> models), you'll be ok. As you may have gathered, I _was_ quite mindful
>>>> of
>>>> this issue. ;-)
>>>>
>>>>>
>>>>> As for the power usage and the "kill-a-watts" units readings - I gort
>>>>> a replacement for my EM100, and swings between about 33 and 40 watts
>>>>> with the occaisional spike to 70 - I'll run it for a period of time
>>>>> and see what the average consumption comes out to. The old unit
>>>>> indicated over 60 watts, my guess is less than 35 watts. This is a
>>>>> TRUE ONLINE dual conversion UPS, so it will draw significantly more
>>>>> than a standby unit.
>>>>
>>>> If this is the type that uses its inverter full time to power the load
>>>> from a float charged battery, that'll be exactly right!
>
>>> That's the definition of "dual conversion"
>
> I was just confirming that we were both on the same page, as it were. ;-)
>
>>> The new EM 100 registers .21 KwH over a period of 5.1 hours - which
>>> translates to almost exactly 40 watts. The line reads 113 volts on the
>>> new unit, and 116 on my old one.My LAB meter reads 115.6 volts - so
>>> the new meter is going back.
>
> Woah! Hold your horses! Does your LAB meter take a series of samples,
>square each value, sum them over a period of time, preferably over at
>least one or more complete cycles, average the results then calculate the
>square root of that average to accurately measure the RMS voltage?
>
It is called a "true RMS" reading AC voltmeter.(digital, not moving
coil)
> Or, is it just an accurately calibrated moving coil meter with a
>rectifier to produce an average voltage reading which is then calibrated
>by the multiplier resistor to produce a deflection on the scale based on
>the fixed relationship between average and RMS for a sine waveform?
>
> Or, does the meter simply use the peak value of the mains voltage and use
>a multiplier resistor 1.4 times larger than it would be for a constant DC
>voltage to scale the peak to its corresponding rms equivalent (again, a
>correction factor based on the assumption that the mains is a pure sine
>waveform)?
>
> Since that EM100 is already gathering such voltage signed sampling data
>to be multiplied by the corresponding current signed sampling data to
>calculate both positive and negative power values to be averaged then
>summed to produce a net power flow reading, it seems highly likely that
>this data and processing power is not going to go to waste.
I have 2 EM100s One agrees with my lab meter - the other does not.
Reads a full 3 volts less.
>
> There's every chance, within the limits of sampling error, that the meter
>is giving a more accurate RMS voltage reading than a simple moving coil
>meter ever could since the mains waveform (at least in the UK) is quite
>noticeably flat topped (if you have you ever looked at the mains waveform
>on an oscilloscope, you'll know exactly what I mean).
>
Mains voltage pattern here in Ontario Canada is as close to a perfect
sine wave as you will find - no "clipping".
>>>
>>> It also indicates 0.69 amps, which at 113 volts would be 78 watts.
>
> No, that just means it's 78VA which may or may not be 78 watts. In this
>case, the real power figure is likely to be a lot less though not to the
>same extreme you might see with a line interactive switching UPS such as
>the SmartUPS 2000 that I've just been testing 3 digital meters and my
>analogue Metrawatt power meter on[1].
The reading on the EM100 is KwH, Watts, and Amps. The KwH devided by
on-time in hours gives me the 40 watt reading (41.25 watts over 16
hours)- I agree with your 78 VA, rather than 78 watts, because either
the power factor is crasy or the meter is reading peak amps (not
sampling fast enough) and the input current to the UPS is switching at
a frequency that gives a wrong "average" current draw. I'm thinking
power factor, or a combination.
>
>>>
>>> It is now on the "old" EM 100, which is more accurate voltage-wize,
>>> according to my meter. It registers 0.72 amps at 116 volts - which
>>> would be 83.5 watts. We will see what the KwH reading over time says.
>>
>>
>> Well, they both agree with 40 watts, when the KWH is divided by time.
>
> That does seem the more likely figure in view of the fact that the UPS is
>float charging a battery with a steady inverter load on it. Incidently, it
>would be useful to know what the UPS's model number is or it's basic
>specifications.
>
UPS model is (formerly Excide)Powerware Prestige 650.( I thought it
was a 1000)
The sticker says 4.9 amp at 0.9PF at 120 volts, 50 or 60 hz.
>[1] I've just spent about an hour in my basement collecting as much
>measuring data on the four states of operation of the SmartUPS 2000. These
>were rear panel switch in the off position (all functions disabled), rear
[continued in next message]
--- SoupGate-Win32 v1.05
* Origin: you cannot sedate... all the things you hate (1:229/2)
|
