[mythtv-users] Power outages and UPSs

[Michael T. Dean]

On 07/28/2007 05:00 PM, f-myth-users at media.mit.edu wrote:
>     > Date: Sat, 28 Jul 2007 11:37:49 -0400
>     > From: "Michael T. Dean" <mtdean at thirdcontact.com>
>
>     > So, almost always a float charge with the occasional (manual) topping
>     > charge,
>
> And what about the constant-current phase first, aka bulk charge?
> Who's doing that?  Or is it your intent to take many times as long as
> necessary to recover from an extended power outage which discharges
> your battery?  [And see PS for a totally different approach.]
>
> [There's also the issue that your load probably consumes many times
> the current of the float for the battery, so getting the Vreg correct
> on both legs of that will be interesting, and voltage drops down wires
> become noticeable as well.]
>
> 	   >  as described in the link I posted.
>
> Which?  Where?

The one that described an approach in an article on the 'net that Lionel
said could be done better with a DC-DC PSU.  The one you obviously
haven't read because you're trying to convince /me/ that "my ideas"
(which are not my ideas, but are straight out of the article) are
wrong.  Perhaps you should write to Dan and set him straight (or at
least have him remove his blasphemy from the 'net).
http://www.gossamer-threads.com/lists/mythtv/users/281185#281185

>   You posted several, but I didn't see that.  In any
> event, if all you're going to do is trickle, then you'll have very
> poor recovery from an extended outage,

And I'm saying this probably isn't a problem because my power is almost
always available.  I have infrequent short outages (less than 5
minutes--usually just blinks) less than 10 times per year.  Occasionally
I'll have longer outages (perhaps once per year or once per 18 months). 
The longest outage I've had that wasn't due to weather (but due to an
idiot drunken driver) was 1 hour.  If that were to drain the batteries,
they'd probably have almost a year to recharge. :)  (Once in 13 years I
had a 36-hour outage from a hurricane, but that's not the problem I'm
trying to solve.)

>  and you don't want to bulk
> charge at the topping voltage (excessive current -> gassing).
>
>     > >  and
>     > > undercharge it as well.  Attempting to recharge only via float can
>     > > take days,
>
>     > Not usually a problem with big enough batteries.
>
> Are you saying "I don't expect to -ever- discharge my battery"?
>   

I'm saying that I don't expect to completely discharge my battery often
and, therefore, it won't often have less power than it needs to ride out
the small (and infrequent) outages.  And, since--more than anything--the
whole reason for the battery backup is to get me through power outages
during the one or two week trips I often take so I can get home and do
what I need to do to set things up for the next trip.

I'm really not aiming for five 9's uptime or anything.  Downtime is
never a problem for my systems--as long as they recover after the fact. 
I just want a solution that will ride through power blinks (and perhaps
the occasional couple-of-minute-long outage) and that will allow my
system to reboot when power returns.

> If that's not what you're saying, then the problem will only get worse
> as cell size increases, because your trickle-charge current will
> increase with it.  [I'm not sure if the intent is "shut down at
> the first sign of trouble and don't come back up until the mains
> are back" or "run as long as possible and -then- shut down if the
> mains haven't returned."]
>
> Basically, what I'm saying in this increasingly off-topic thread is,
> "Some of us are electrical engineers who have designed high-rate
> lead-acid chargers, and some of us aren't."  A good charger that
> minimizes your time discharged (and vulnerable to the next outage)
> and which also maximizes your battery lifetime is a lot more
> complicated than a benchtop supply that's just set to a single
> voltage and forgotten---but it's still not rocket science.*

Again, it's not my science.

>   It
> takes some good power engineering for the high-A paths and (these
> days) a -lot- of what used to be an annoying analog computer is
> now available in IC form to do all kinds of great charge regulation,
> e.g., the UC2906, as well as many others from Maxim et al.
>
> [* When charging a single 12V lead-acid at a few A, that is.  But easy
> commercial solutions don't really exist for 72V AGM strings at many
> hundreds of A, though, especially given other numerous constraints,
> which is where I came in.]
>
>     > Interesting.  I'll keep this in mind.  Looks like the
>     > appropriately-sized commercial chargers are around the same cost as
>     > plain bench power supplies.
>
> And they'll perform better.  Consider also what's going to happen with
> your load/charge regulation if you have a power supply that's not
> designed with sufficient output current to recharge a discharged
> battery.  The battery will drag your charge rail down sufficiently
> that your -load- (e.g., the computer) may not work until the battery
> has been able to recharge---which, at trickle rate, will take days.
> Stiffening Vreg on the charge rail is going to take a fair current
> output capacity; going into current-limit on your benchtop supply
> means your -load- may be current-limited as well as your battery,
> and that's unlikely to do what you want... :)
>
> Thus, since you're using a nonbypassable online UPS (e.g., your load
> -must- run from the battery rail, not from the AC mains), your load
> is hosed until your battery resistance rises.  And if you oversize
> a bench supply so it can simultaneously bulk-charge the battery and
> supply your load, then (a) it has a risk of recharging the battery too
> fast unless you've got everything tweaked just right [don't add or
> subtract hardware!] and (b) bulk-charging at the trickle-charge or
> topping voltages may gas your battery.
>
>     > Yeah, but considering my consumer UPS's have caused more downtime* on my
>     > systems than running without them (since they'd just come back--a little
>     > worse for the wear--when power is restored), I've unplugged them all
>
> I don't understand this sentence.  Is the problem here that you get an
> extended outage, the UPS runs flat, the power comes back, the CPU comes
> up, and then you get an immediate outage again when there isn't enough
> UPS runtime to do a controlled shutdown?  (If so, and you don't want to
> spend the extra money on a SmartUPS or equivalent with its "hold down"
> logic, there are very simple circuits (amounting to a pair of one-shots)
> that can say "if AC fails for more than x minutes, then don't turn the
> power relay back on until y minutes after it's back".  Not as good as
> something that's actually monitoring the state of charge of the battery, 
> but not horrendous either, and very cheap.)
>
> Oh wait.  "they'd" is your CPUs, right, not the UPSes?
>
>     > (and am running directly off the mains).  Basically, cheap "smarts" are
>     > worse than "dumb."  At the least, I'll be going for a very large APC
>     > SmartUps (SUA1500 or more) if I go with an OTS UPS.
>
>     > * The UPS doesn't restore power after an outage, so my system is down
>     > the 4 days while I'm traveling rather than the 10 minutes the power was out.
>     > * The UPS decides that the battery isn't performing as expected, so it
>     > cuts power--while there's still power to the mains--causing an outage
>     > that wouldn't have existed without.
>     > * The circuit breaker in the UPS triggers and the /only/ way to reset
>     > the UPS is manually by holding the power button for 4 seconds or something.
>
> WTF?  All of these sound like broken or misconfigured UPSes.  I've
> used a couple dozen midrange consumer APC units (Back-UPS Pro &
> ordinary Back-UPS; SmartUPS, and also much older ones like the 1200VX
> and others) for 15+ years and I've never seen any of these misbehaviors.
> I assume other vendors' UPSes are similar or we'd hear lots of screaming.
> (a) & (b) are just dead wrong;

Both of my APC Back-Ups did that.  Perhaps they were broken, but after
just over a year, I felt they shouldn't be (and I refuse to keep buying
new  garbage to replace the old).  I don't know if the problem was
buying consumer-grade UPS's, but I'm convinced that consumer-grade UPS's
aren't what I want.  I may spend several hundred dollars on some
SmartUps UPS's, but the smarter my previous UPS's have been the more
they've failed (i.e. they think they're smarter than they are).

>  (c) sounds like you're overloading the UPS;
> the only time I've seen (c) is when some numbnut plugged a laser printer 
> into a UPS, which managed to both pop its breaker -and- permanently blow
> up its inverter (!).
>   

I'm pretty sure that my 8-port gigabit switch (only thing plugged into
that UPS) wasn't the culprit (technically, I also had a 5-port 100Mb
switch plugged into the surge-only, non-battery-backup side of it,
too).  I'd expect it was actually a power surge that the UPS couldn't
handle.

> Also, how old are (were) your batteries again?
>
>     > Basically, I travel enough that these things cause days of outages
>     > because no one is around to reset the UPS/hit the power button on the
>     > computer.  When using the UPS's, I would have days or weeks of downtime
>     > (generally about every 4th trip--and, it seems, usually losing the
>     > system on day 1 or day 2 of travel).  Since removing all 7 UPS's from my
>     > 7 systems, my downtimes have only been in the minutes.
>
> What brands/models of UPSes are you using?  
>
>     > >   And you can only
>     > > get rid of the always-on "inefficiency" of a typical switching power
>     > > supply (already about as efficient as any supply you'll see) if you
>     > > can supply exactly the required voltage from your batteries,
>
>     > I'm not trying to improve efficiency, I'm trying to find a workable
>     > solution that uses the batteries I choose.  However, the dual-conversion
>     > design introduced a completely unnecessary inefficiency, so...
>
> My inefficiency comment wasn't aimed at you, but at someone else in
> the thread who bemoaned those "inefficient" CPU switchmode PSUs, which
> are likely more efficient than anything else in the entire DC powerchain.
>   

Who was telling me that Dan's solution from the article I posted would
have been inefficient...

> P.S.  If your currents are low enough (doubtful), and you only want to
> ride out a few seconds of interruption (not enough for a controlled
> shutdown, but to get around flickering) then another solution which
> bypasses all the complexities of battery charging regimes would be
> to use some supercapacitors on your DC rail.  You'll get more usable
> capacity out of 'em (due to the exponential in the discharge curve)
> if you can either make a high-V string (20-30V) and use a buck
> converter to generate your 12V (or whatever), or make a low-V string
> (e.g., 5, 7.5, or some other multiple of the basic 2.5V rating of most
> supercaps) and a boost converter that can supply your 12V (or whatever)
> until the whole set falls below your boost's min Vin.  So compute the
> load in W, remember that caps store 0.5 * C V^2 in joules, and do the
> math to compute the F required for a given t; who knows, if you need
> few enough wattseconds aka J, it could work---and it will last "forever",
> rather than decaying in a few years the way batteries do no matter how
> well they're treated.  If you're willing to spend a couple hundred bucks
> for that AGM battery you were talking about, you could get a few thousand
> F of supercap and that might be good enough.  Send me the VA of what you'd
> like to power and I'll spend a couple minutes computing whether it's even
> -possible- to survive more than 5s of flicker without going broke...:)
I'll keep that in mind.  I'm still in the research phase (as I have been
for about 2 years--the second of which I've been operating without any
UPS's).


Also, I'm anxiously awaiting your, "Better than Dan's DIY UPS" article.  :)

Mike