Author Topic: UPC Batteries - fail vs no-fail  (Read 1123 times)

0 Members and 1 Guest are viewing this topic.

Offline Lorenzo_1Topic starter

  • Regular Contributor
  • *
  • Posts: 218
  • Country: au
UPC Batteries - fail vs no-fail
« on: August 07, 2022, 02:56:31 am »
Thought members might be interested in battery failure in APC UPS SURT1000XLI.  Had a reserve battery pack plugged into rear battery port for lasy year or so.  UPS started alarming - bad battery. Pulled main and reserve pack out and found three dead batteries in reserve pack.  All were the same brand and all had swollen and burst open (pics attached).  The fourth battery in this pack was a Yuasa of the type I usually buy and it was in seemingly fine shape and fully charged.   Figured failure likely not due to overcharging given the internal pack and the Yuasa battery in this external pack seemingly suffered no ill effects.  The Yuasa batteries might well be newer than the others however - can't easily track that down. Could this failure be caused by mixing battery types?  Attached datasheets for both in case of interest.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9016
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: UPC Batteries - fail vs no-fail
« Reply #1 on: August 07, 2022, 03:07:55 am »
Just poor quality batteries failing early. The Yuasa is a better brand so it lasted longer, could probably still get some more use out of it in some other application.

Probably best to replace with LiFePO4 nowadays, you'll have to figure out the wiring configuration to get the correct voltage and add a properly rated BMS.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline edpalmer42

  • Super Contributor
  • ***
  • Posts: 2271
  • Country: ca
Re: UPC Batteries - fail vs no-fail
« Reply #2 on: August 07, 2022, 04:21:50 am »
Mixing brands or ages of batteries can cause problems if the batteries are in series.  Parallel connections are okay.

When the batteries are in series, the lowest capacity battery reaches full charge first.  The voltage starts to rise, which should reduce the current.  But the other batteries aren't fully charged so their voltage is still low.  The result is a high voltage across the smallest battery which causes it to bubble and lose water, even though it's supposedly a 'sealed' battery.  It dies a horrible death and the problem sometimes dominos to the next smallest battery until the UPS finally realizes that something's gone horribly wrong.

Note that in your pictures, the dead battery is shown as 7Ah, possibly 7.0 Ah based on the part number, while the Yuasa battery is 7.2 Ah.  Could be marketing or different measurement methods or it could be that the Yuasa battery has a slightly higher capacity.

Some APC UPSs are known to have too high a charging voltage.  They want to recharge the battery ASAP in case there's a second failure so they overcharge the batteries.  The batteries recharge quickly, but it shortens their lifetime.
 

Offline J-R

  • Frequent Contributor
  • **
  • Posts: 975
  • Country: us
Re: UPC Batteries - fail vs no-fail
« Reply #3 on: August 07, 2022, 04:28:27 am »
Previous poster beat me to it, but agreed that it's critical to have identical batteries in each series string, but somewhat less important that they match when in parallel, although there are some extreme cases where that isn't recommended.

SLA batteries can have drastically variable life spans depending on environmental and use conditions.  I have some UPS units with batteries over 10 years old.  A lot of UPS designs honestly overwork the battery by allowing it to support high loads and discharging as well as charging too quickly.  1C max is a reasonable rule of thumb for discharging and 0.1C max for charging.

There are plenty of bad battery photos out there, but attached is one from a UPS I did recently that was a bit of work.
 

Offline Lorenzo_1Topic starter

  • Regular Contributor
  • *
  • Posts: 218
  • Country: au
Re: UPC Batteries - fail vs no-fail
« Reply #4 on: August 07, 2022, 07:35:49 am »
Thanks all for the useful advice - I think I'll stick with Yuasa and make sure I don't mix types next time round.
 

Offline wraper

  • Supporter
  • ****
  • Posts: 16863
  • Country: lv
Re: UPC Batteries - fail vs no-fail
« Reply #5 on: August 07, 2022, 07:53:04 am »
CSB is as good as Yuasa and usually can be purchased significantly cheaper. They often come from factory in expensive high quality UPS such as high power Eaton and Riello. APC tends to use garbage batteries and capacitors even for expensive models. Also it's worth considering not only brand but battery model as well. Often you can get a model with two times longer rated lifetime for like 20% more. Long life models have more lead in them therefore they are heavier.
EDIT: suggest this one in particular https://www.csb-battery.com.tw/upfiles/dow01601472210.pdf
« Last Edit: August 07, 2022, 08:36:43 am by wraper »
 
The following users thanked this post: Per Hansson

Offline Lorenzo_1Topic starter

  • Regular Contributor
  • *
  • Posts: 218
  • Country: au
Re: UPC Batteries - fail vs no-fail
« Reply #6 on: August 08, 2022, 06:36:51 am »
Thanks for the tip - these are available in Oz.  Will give them a try.  Not sure they'll prove much cheaper than Yuasa after GST and freight. But certainly competitive.
 

Offline wraper

  • Supporter
  • ****
  • Posts: 16863
  • Country: lv
Re: UPC Batteries - fail vs no-fail
« Reply #7 on: August 08, 2022, 08:16:11 am »
Direct equivalent of that Yuasa is GP1272F2, not long life GPL1272F2 which is of course more expensive than GP series.
 

Offline CaptDon

  • Super Contributor
  • ***
  • Posts: 1740
  • Country: is
Re: UPC Batteries - fail vs no-fail
« Reply #8 on: August 08, 2022, 12:39:15 pm »
I have a unit here connected to my bench. it has a pair of 12vdc 12ah gel-cells that are never balanced because they are in series. Both batteries are same make/model/age/capacity. One is at 13.7vdc the other at 14.2vdc and that is with a 680 ohm 'equalizing' resistor across each battery to help balance the charge current per battery (about 20ma. through the resistors). Without the equalizing resistors that I installed there was a much greater difference in charge voltage, more like 13.5vdc vs. 14.4vdc or even worse. Of course if I wanted to power down the unit I would have to disconnect the resistors so I installed a Jones plug so I could unplug the resistors. It also makes for a handy place to measure the individual battery voltage under charge and load conditions. I usually test run my bench UPS for about 5 minutes once a week.
Collector and repairer of vintage and not so vintage electronic gadgets and test equipment. What's the difference between a pizza and a musician? A pizza can feed a family of four!! Classically trained guitarist. Sound engineer.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf