[Tfug] UPS sizing

[Bexley Hall]

Hi, John,

--- On Mon, 1/5/09, John Gruenenfelder <johng at as.arizona.edu> wrote:

> Well, certainly the oldest stuff was knob and tube. 
> Actually the first time I had seen something like that 
> in person as there aren't many places in Tucson

Yeah, there are homes in New England that were around
long before AZ was a "settled territory"  :>

> old enough.  :)  How exactly this was melded with the newer
> circuits I am unsure, considering that the power requirements 
> of 2009 are significantly different from those of 1890.

Most municipalities grandfather in older wiring re:
newer (electric) Code.  Some places require an electrician
performing work on your premises to bring existing wiring
"up to (latest) Code".  Needless to say, this is rarely
done (since it can dramatically increase the cost of getting
*anything* done).

But, wire is wire is wire so a circuit installed 50 years
ago will still carry electricity to your appliance; you
just might not have all the amenities you expect (i.e.,
safety ground, current carrying capacity, too few circuits,
etc.)

<shrug>

If you look at most "modern" houses (i.e., last 20 years)
you'll find that even they don't often have "good power"
considering how much is used nowadays.  E.g., most of the
homes I have seen have a single GFCI circuit in the house
that *only* handles the outdoor, garage and bathroom
outlets (all on the *same* circuit!).  Run a hair dryer
in *one* of the bathrooms and anything running outside in
the yard (e.g., XMAS lights) or in the garage (e.g., a
freezer/refrigerator) or in some *other* bathroom (e.g.,
a curling iron) can easily blow that breaker.  A neighbor
was never able to run his microwave oven without tripping
a circuit (most kitchens don't have enough circuits for
all of their loads)

Current Code requires GFCI's on each of the kitchen
(outlet) circuits, bathroom, etc.

<shrug> 

> >Possible.  I've seen APC units "click on" (i.e., go into
> >backup mode) spontaneously.  I've always assumed it was the
> >result of a brown-out -- though have never been able to
> >correlate this with "dimmed lights", etc. <shrug>  As long
> >as it didn't happen too often, I just shrugged it off (I've
> >got enough *other* stuff to do so why go looking for issues?)
> 
> Ahh... I'm glad you mentioned this.  Back in the old
> house my UPS would do
> this click-click about once per day.  NUT would make a note
> of this in the syslog each time.

But, it would just say "UPS switched on" (whatever) and not
*why* it did?

> Until now I hadn't realized that this hasn't occured once
> since I moved back.  I had thought that perhaps it was some
> sort of self-test, but since it has stopped that can't be it.

No, I'm sure it is the UPS reacting to *something*.  But,
since the "user I/O" on a UPS is pretty limited... :>
it's kind of hard to figure out what it is doing and *why*.
And, unless you can *notice* something happening wrt the
AC power (e.g., lights blinking), there's no easy way for
you to determine if, indeed, what the UPS is reporting is
actually *true* (i.e., how do you know that those 2 cycles
were 9% below nominal voltage?)
 
> >> This was a major pain for a number of reasons.  It could
> >> only be reset via manual intervention which made remote
> >> action impossible.
> >
> >That sounds like an overload, then.  Note that most UPS's
> 
> I suppose, but then why hasn't it occured again since I
> moved back to Tucson?  My output wiring is almost identical.

Dunno.  No doubt has *something* to do with the mains
where you are now vs. then.  That's the annoying thing about
most of these products -- you can't look at the code and
see what the device is *trying* to do.  The manufacturers
act as if this is some sort of super-science; some special
technology that only *they* could understand and don't want
to disclose for fear their competitors will benefit from it.
(in reality, there is usually very little "trade secret"
information in these devices... but, it's *their* IP so they
can decide whether or not to disclose it  :<  )  Having
access to that information would help you figure out why
its doing what it is doing, etc.

(e.g., having access to the code in an engine control unit
would make it a *lot* easier to figure out why it is
reporting some particular error code, etc.)

> Hmmm... based on what you wrote about the two types,
> I'm pretty sure my
> relatively inexpensive $140 UPS is of the standby variety. 

It's an easier way of doing things.  Otherwise, they have
to build an AC->DC power supply that is capable of supplying
X+15% Watts (for your X Watt load) in addition to the
charger, monitoring circuit and X Watt DC->AC inverter.

> If that is the case, then perhaps during some bizarre power
> event on the mains, my load will see that event for at
> least 15ms.  Then, during that time span, this "event"
> causes one of my devices to suddenly draw lots of power,
> thus kicking the UPS into overdraw shutoff.

<shrug>  One can only hypothesize.  :-(  The power supplies
in most newer electronic devices are pretty hard on the mains.
Older supplies (linear) started up "soft".  Your load (inside
the device) sat behind a transformer so the current drawn from
the mains was reduced by the windings ratio of the transformer
(e.g., if the transformer delivered "12V" to the circuitry
inside the device, then the actual current drawn from the
mains was 1/10th what your device was using -- blech, I said this
poorly  :< ).  Newer power supplies convert the AC mains directly
to DC (i.e., about 170 volts DC) and then chop that up (duty
cycle control) to get your "12V".  But, as a result, when
your device first turns on, it presents a dead short to the
AC mains as that 170VDC power supply charges up "instantly".

When I built my first (HiFi) amplifier, I deliberately added
big resistors in the power supply so that the initial turn-on
wouldn't blow the circuit breakers.  It was a linear supply
but quite large so it exhibited this same sort of behaviour.

> It's a possible explantion, anyway.  I guess the real
> solution is, if I ever move back to a place with variable/lousy
> power and/or ancient wiring, I should
> spend the extra money and purchase an online UPS. 

<shrug>  Or, live with the occasional "blip".

> Afterall, having a UPS which only protects me from outages 
> and brownouts when those constitute only
> about 1/3 of my power problems isn't doing much.

<shrug>  It depends on how much of a problem those power
events are causing you.  I once tried using a CVT (constant
voltage transformer) as a power conditioner.  It worked
pretty good but it was very heavy and I didn't like the "hum".
And, of course, it really was just a (great!) line filter
and didn't really give me any protection from brownouts, etc.
(but, it could almost survive a direct lightning strike  :> )

> >If you *really* are concerned about power, you can buy
> >tools that will watch the AC line and log anomalies
> >(brownouts, blackouts, dropped cycles, spikes, etc.).
> >But, $$$$
> 
> I can get at least some of this information from my current
> UPS by using NUT.

Ah, OK.  I guess it depends on how much the UPS is willing
to report.

> There is an add-on graphing/stats package for it.  I
> attempted to set it up when I first got the UPS, but never 
> followed through.  The only event which truly gets
> reported now is switching to and from battery. But with decent
> graphs I could at least deduce some trends.

I'm not sure of that.  Power anomalies can be very isolated.
It's not likely that you'll see, for example, the AC mains
voltage slowly drooping over the course of a few minutes
just prior to a brown-out, etc.

There are a lot of rules governing the supply of mains power.
E.g., how tightly the frequency has to be controlled, voltage
variations, etc. -- for obvious reasons.  :>  And, its just
too easy for people (i.e., businesses who could suffer losses
as a result of the utility failing to meet these needs... said
businesses having lawyers on staff just looking for a reason
to litigate  :> ) to be able to *watch* what the mains look
like (contrast this, for example, with agriculture where it
is pretty hard to guarantee that the product you are buying
from the farmer has been grown in a manner that won't harm you)

> However, power on the East side
> of Tucson is so stable, there's not much point in doing
> that here.  :)

Yes -- though we had a "blip" the other night.  Just long
enough to trip the UPS's (and crash those machines that were
not connected to them!).

> >Note that the output of many UPS's isn't very "clean".
> >If you get neighbors complaining about RF interference,
> >this is one quick fix to try (eliminating the long cord).
> 
> That wasn't a problem in the old house.  It had actual
> thick plaster walls.
> Until I moved in I didn't know that plaster is
> remarkably good at absorbing RF energy, wanted or unwanted.

Really?  I wasn't aware of that!  (however, I do know that
plaster is a good fire control mechanism.  I think because
there is moisture trapped in the plaster?)
 
> >[1]  This ignores the effects of "surge supressors" in the UPS
> >Personally, I think these "supressors" are a complete waste!
> >They'll only handle trivial problems -- problems that the
> >power supply in most electronic devices will already handle!
> 
> Are they really that bad?  Certainly this was one of the
> things I wanted when I bought the UPS.  My PC is still
> plugged into a surge protector/power strip
> as a matter of convenience, but I assumed the UPS would do
> at least as good a job as that, if not better.

<frown>  Well, it's better than *nothing*.  But, if you look
at their physical size (and compare them with their counterparts
present *inside* your PC's power supply, etc.) and "do the
math", you can see that they can't really store much energy
(the purpose of these is to *store* the transient/surge and
dissipate/release it thereafter).

I think they probably provide some measure of RFI/EMI control
that is of more practical benefit than their "surge protection".
(i.e., they help cut down on the interference that your equipment
*generates* on the AC line).

<shrug>  I don't know what the actual cost-benefit analysis looks
like.  And, since they are "free" (i.e., it's hard to find a
UPS that *doesn't* have them), they are better than nothing.
I just don't know how much practical benefit they have vs.
their added product cost.  :-/

--don