New Battery Shelf Life - How much is too much - and what is APC's policy?

[Anonymous user]

I just got a 550 UPS from Amazon.   When I plugged it in, it showed 62% capacity.    After it charged to 100% and the plug is pulled from the wall to go into UPS power it immediately drops to 75% and then under load it decreases at a slow pace downwards.       Tech support verified that the battery would not last as long as it should.

I asked tech support if having a new battery and ups come in at 62% was acceptable and he said APC does not ship batteries fully charged.  After doing research on this,  industry websites report that if a battery sits without being charged it immediately starts to degrade.  The term is sulphation.    Question for someone from APC.  Is is really true that you would intentionally ship a battery that is not fully charged?   If so, can you explain why?

Second, I asked the tech support person if the battery sat too long on the shelf at Amazon since the testing was done in September and I received it in December.    It was actually about 3 months and 5 days.    He said a battery could sit for 6 months on a shelf in an uncharged state without affecting the battery.   And yet I find on the net a guideline that if a battery is left on the shelf for three months it should be sent back.   The tech did say at the end of testing that the battery may have sat on the shelf too long.  Question:  What is acceptable shelf life for these batteries?

Finally,  could a battery damaged in this sort of fashion be rejuvenated by being put on a smart charger.   From looking at the time it took to charge (See my previous post on the BE550G)   it appears that the charge rate is a slow charge at a constant rate up until 100%.  A smart three phase charger will ramp up to almost full and then drop down to a trickle to maintain.

[BillP]

What I know is that units in our own stock are re-charged at a minimum at least every 6 months. If it has been a few months in storage or on a shelf at a distributor after it has left APC warehouses, then it is feasible it could drop a certain percentage. The long and short is that if a battery is not charging properly or is not operating properly (such as not providing the proper amount of runtime), then it would be replaced under warranty assuming it was just purchased new from an authorized reseller (like Amazon). It usually takes about 3 months average from manufacturing to reach distribution (based on where it is manufactured, travel time to that country/location, time it takes to get to you as the user, etc) so I think your experience is normal and how it should work.

I don't think batteries are intentionally ship discharged but rather it is normal that they discharge within the few months they are not plugged in waiting for someone to adopt them 

Our information available on this topic is here:

http://www.apc.com/support/index?page=content&country=ITB〈=en&locale=en_US&id=FA156516

http://www.apc.com/support/index?page=content&country=ITB〈=en&locale=en_US&id=FA157446

I think the major harm is leaving them at like 0% for an extended period of time.

[Anonymous user]

 I could not see where to start a new thread so I will respond to this topic I started.

I have a 1300G which is described as having 1300 VA.   The chart on the description says it will handle about 100W for one hour.    The way I understand VA is that it is the same as watts.    The UPS apparently has two internal batteries which from the picture on the web site are 7AH each.   That to me says there is a total of 14 AH, which I though was

the same as watts and VA.     So my question is, where am I going wrong in my understanding.   How can two 7AH batteries which are 14 watt hour capacities power 100W for one hour?

[Terry_Kennedy_apc]

Two issues here:

VA does not equal Watts for complex loads such as computer power supplies, though sometimes people refer to them interchangably. The discrepancy between them is called the power factor.

Next, AH is a unit of charge (think of it as "how much electricity can you fit in a battery") not a unit of energy. So there's no conversion from VA or Watts to AH. Kind of like one describing how much water fits in your bathtub, and the other describing how rapidly it empties out when you remove the drain stopper.

This is a bit of a simplified explanation, but I hope it helps.

[Anonymous user]

Thanks.  So if we ignore the power factor and consider this to be a simple resistive load with no capacitance, then  how does one derive 1300VA from two 7AH batteries?

### Edit ###

Looks like reverse engineering this, the way it was done is to take 1300 av / 14 ah = 93 which with

inverter efficiency of 93% would mean that the 14 ah is being multipled by the AC voltage.

Is that correct?

###

[voidstar_apc]

Don't forget the units...

1300 va / 14 ah = 93 v/h

Though I'm not sure of any physically meaningful intepretations of volts/hour.

More seriously, 1300VA is not about the batteries, it's about the UPS design. It's a limitation of the UPS. Going over 780W or 1300VA will overload the BR1300G. So if your load is resistive, the 780W limit applies, and if your load has a power factor <<1, you'll hit the 1300VA limit first.

The power/runtime curves on APC's website is measured in a lab, not derived from amp-hours. If you look at the curve, runtime goes up exponentially as load goes down, because the slower you discharge a battery, the more capacity it has.

[Anonymous user]

Thanks for that explanation.   So how do we determine how long two 7AH batteries will power a 60w load at 110v?  The APC curve shows that it can power that for about one hour.   If AH is equivalant to watts with the proper load, I would expect 14/60 of an hour at the very most if we ignore the inverter losses, etc.  So obviously I don't understand this at all.

[Terry_Kennedy_apc]

The answer is "it depends". 7AH means "7 Amps for 1 hour", but that does not imply 1 Amp for 7 hours". The relationship is more complex than that, since there's physical chemistry at work. The graphs on various battery manufacturers' spec sheets show the expected behaviour for a new battery. Since these batteries are all Sealed Lead Acid (SLA) regardless of manufacturer, the graphs will be similar.

APC presumably takes that data, combines it with either the calculated or measured efficiency of the electronics in the UPS at various load percentages, de-rates the results to take into account battery aging during the warranty period and uses the resulting data to produce the graphs and the runtime calculator.

Here's a back-of-the-envelope calculation: 7 Amps x 12 Volts = 84 Watts or 84VA (this is DC, so there's no power factor). We have 2 batteries, so this is either 14 Amps x 12 Volts (parallel) or 7 Amps x 24 Volts (series). That gives us 168 Watts for 1 hour if the batteries were connected directly to a corresponding 12V or 24V DC load. We don't have a DC load, however - the UPS has to convert that back into 120V AC (for domestic units; anything from 100V to 240V for international models). Assuming (I have no idea if this is anywhere close to reality) an 80% efficiency on the inverter, and perhaps de-rating another 20% for aging effects on the battery, and we end up with 107 Watts for 1 hour. The inverter efficiency calculation probably has three components - a fixed loss regardless of load (overhead), a variable loss depending on the load, and a variable loss depending on the battery voltage (which drops over the period of time the UPS is on battery as they discharge). But that would require a bigger envelope, and we don't know the actual numbers anyway, so we won't consider it.

In any event, 60W for 1 hour sounds reasonable given the above calculations.

[Anonymous user · Accepted solution]

I think I see where I was going wrong.   I was looking at the 7AH and thinking it was watts, but that is AV.  Thank you very much for your time and also communcations skills!