APC Back-UPS 1500 erratic behavior

[Curious.George]

Re: APC Back-UPS 1500 erratic behavior

Measured voltage on the pack, it was 30.8V charged.

Did you measure this with the batteries still connected to the UPS? Or, did you lift one of the legs from the charger so you were actually seeing the batteries' voltage? (ideally, you'd put a brief load on the battery pack to remove any surface charge before measuring).

Float charging a battery should leave you with ~2.3 volts per cell (6 cells in a nominal 12V battery means 13.8V per battery or 27.6V for the entire battery pack). Even a fast charge would typically only be about 2.4 - 2.45V/cell (or 28.2 - 29.4V for the pack). Are you using a DMM that is reasonably calibrated?

It is not uncommon for UPS's to overcharge their battery packs. The batteries you removed reflect the common result of such a "cooking". You should check your UPS to see what sort of float voltage it impresses on the battery pack.

I have each of my UPSs set up to log mains voltage/current, load voltage/current, battery voltage, temperature, etc. regularly (once a minute) and dump these log files onto one of my servers. This lets me see if something wonky is happening in the UPSs before things go off the rails (I have a little script that watches the log files and alerts me of any trends that seem odd). I tend to see ~13.6V reported for each 12V "battery" in a battery pack (again, some packs have 4 such batteries so ~55V for the pack).

Another Q: can 2 batteries at different discharge level be charged together by the UPS as a pack? Like one is at 14V, and another at 6V, but surprisingly keeps charging? How in this case the charge levels between batteries, and how the UPS decides that the pack get fully charged? How generally the UPS measures a battery charge level?

Ideally, any battery wants its cells to be in approximately the same state when being charged. (note a battery is a group of cells; a battery pack is a group of batteries; so, a battery pack is a battery is a group of cells!) Otherwise, one cell may become overcharged while another is undercharged. This eventually manifests as a faulty/inadequate battery.

But, battery management systems (BMS) are an expensive addition to what is typically a commodity product (UPS). OTOH, you'll note that an electric car invests a fair bit in its BMS as the battery is something that has significant value and premature failure would tarnish the car maker's image in the market!

Note that even many of the larger UPSs don't go beyond treating a battery pack as just a monolithic "big battery". E.g., several of my UPSs use a 48V battery pack but still see it as a "2 terminal device" (i.e., like a single "battery")

Bottom line, you want to put two batteries of roughly equivalent "condition" into the UPS (i.e., never replace just one!)

More Q: can I use together a 3-year old good battery and a brand new one as a pack? If not recommended - why?

See above. Batteries are cheap (a 12V 7.2AHr Panny battery is ~$30; a pair of them for $60; other brands are less expensive). Treat it as "the cost of insurance" amortized over 3-4 years (typ). What is it worth (to you) if the power goes out (or even just GLITCHES!) while you're doing something? Will you remember everything that you have done, recently, in order to recreate it (from some yet-to-be-discovered "last saved state")?

[sambul83]

Re: APC Back-UPS 1500 erratic behavior

Sorry, I didn't realize what you mean "under load". I'll hook the lamp directly to each battery contacts and see, if any of them drops voltage.

They were at 6V each yesterday. By today they slowly recharged by the UPS to 14V+ each, and the pack seems to hold the charge for now. I realize, both may have some short cells despite charging OK.

[budm]

Re: APC Back-UPS 1500 erratic behavior

"Checked voltages as 14.3 and 14.8 on batteries under charge, and 14.1 / 14.5 disconnected." But you indicate 6V, but now the 6V bad battery shows 14.5V disconnected?
If one battery has 14V and connected to another battery (6V) in series, so when you measure the total Voltages (not connected to the UPS) it should then show around 20V with no load.
When I indicate loading the battery during test it meas you connect the 12V battery to the load resistor while having the DC Volt meter connected.

[sambul83]

Re: APC Back-UPS 1500 erratic behavior

I partially discharged the battery pack under 200W lamp load. It was showing 28min backup for that load, and discharging very slowly. Checked voltages as 14.3 and 14.8 on batteries under charge, and 14.1 / 14.5 disconnected. It looks like they may not fully recovered (yet) given somewhat lower voltages, but work OK. I wonder if they dropped to 6V due to a "lack of training" short?

The other Back-UPS that caused the pack sudden discharge seems to work OK with another battery pack, the 3-year pack voltage stays at 30.3-30.8 in small discharge / charge.

With this, its probably better to wait and see if anything happen again rather than replace the batteries just a week after receiving them? May be the other UPS will fail again at mains voltage spark with older battery pack?

[budm]

Re: APC Back-UPS 1500 erratic behavior

"And by connecting 14V battery to 6V battery in series, I assumed the impact on the charger will be soften, the pack voltage higher, and the 14V battery might limit current run through the 6V battery" If you read page 10 you will see why that will not be the case.

[sambul83]

Re: APC Back-UPS 1500 erratic behavior

Thanks for the article. I'm not insisting on anything, just have limited knowledge of the subject (i.e. silly, stupid). My reservation at replacing the brand new battery pack is APC at some point would block my warranty unless there is understanding why such discharge happen. Also, since the UPS accepts batteries as a 2-pack, I didn't try to charge only one battery - is it possible and how? And by connecting 14V battery to 6V battery in series, I assumed the impact on the charger will be soften, the pack voltage higher, and the 14V battery might limit current run through the 6V battery. Right now voltages on each charged and removed battery are measured at 14.8V, will put back and check connected voltages shortly.

The location of that UPS is now far from heat sources. Its unclear why some battery cells in both batteries in the pack were subjected to thermal runaway if that's the case? Something caused them both to discharge to 6V each. Without understanding what was it, do you think its rational to ask APC for a replacement pack?

[budm]

Re: APC Back-UPS 1500 erratic behavior

Read page 10!

[budm]

Re: APC Back-UPS 1500 erratic behavior

1) Testing battery by checking Voltage only is not valid, you need to load test with battery.
2) Check the Voltage on the two terminals of each one of those battery when connected to the UPS while it is being charged an tell us what you get.
3) Did the UPS do the self test to find out if the batteries has the capacity to run?
4) I would not put that Battery with 6V in to that UPS and try to charge it because you can damage the charger circuit or damage the good battery that is connected to that bad battery since you can over charge the good battery.
5) I do not understand why you insisting on trying to charge bad battery. Are you trying to damage the charger?

[sambul83]

Re: APC Back-UPS 1500 erratic behavior

Your 12V battery that shows only 6V is bad

I suspected this is the case, as both brand new batteries in the pack suddenly discharged to 6V each at no load. Do you think both got short cells? May be its because the batteries were fully charged without gradual training?

However, now I recharged both in the UPS by joining each with a good 14V battery. I just run PowerChute self-test, and it shows expected 26min run at 25% load. I know actual runtime may differ, and will check it shortly by full discharge. If the batteries have short cells, they will discharge at once under load - correct? There may be the case they discharged slowly over a few days without the UPS updating charge level on display due to some UPS component intermittent failure (read ccteng's post at that thread bottom).

I can ask APC to replace them under warranty, but the cause of this failure remains unknown, and the chance is the replacement batteries will act the same unless we can find the cause.

[budm]

Re: APC Back-UPS 1500 erratic behavior

Your 12V battery that shows only 6V is bad, period, it probably has shorted cells so forget about trying to revive it.
UPS did not allow the battery to discharge down to 6V, the battery itself went bad.
Batteries are connected in simple series circuit so it does not matter about first or second.
Just do the search on 'series connection batteries'.
Learn about batteries here:
http://batteryuniversity.com/

[sambul83]

Re: APC Back-UPS 1500 erratic behavior

I also noticed some of Back-UPS problems are artificial and APC marketing driven. For example, the brand new battery pack was not charging to full capacity, and also not discharging much, but instead showing as empty within 1 min until I updated Battery Replacement Date in PowerChute software. Then it shown a higher charge at once, and the pack was able to fully discharge at test load instead of shutting down the PC in 1 min. It seems to hint the charge info is stored on the EEPROM, and the UPS in fact controls battery charge & discharge cycle based on replacement date in PowerChute rather then its actual condition. Some UPS are able to adjust charge value to show actual charge during startup test, and some don't. So basically APC could make a good battery look bad after 3 years of service to prompt replacement.

[sambul83]

Re: APC Back-UPS 1500 erratic behavior

When you did the load test on these batteries how long did they last like a new battery would if so

These are new 12V 9Ah batteries, so initial load test at 200Wt 2 lamps load lasted like 15-20 min at factory charge, then I did a couple of full charge & discharge cycles for battery training. They lasted an average time, not immediate discharge, don't remember exactly, but longer than factory charge. How can a brand new never used at full load battery get "tired"?

I wonder how the Back-UPS allowed the batteries to discharge to 6V at no load? Does it hint that the UPS has some intermittent problem? Can a spike in mains cause that, if some load was attached to Surge Protection outlets? But these outlets kept working OK without a battery pack. I had somewhat similar issue with another Back-UPS earlier, where it didn't charge the battery pack due to bad main Cap, so it gradually discharged without load. I was able to fix it with great support from forum members.

The UPS couldn't start with 2 batteries at 6V each, so I joined one battery at 14V and another at 6V to charge it. As you said it got quite heated during the charge over 10 hours, but slowly raised to 14V, but the charged good battery in the pack stayed cooler and at the initial voltage probably since I put it the last. Then I did the same to charge another 6V battery. Will do load test to full discharge on that pack shortly. What would be normal UPS controlled discharge voltage for such batteries - 10-12V?

My Q is: does it make any difference in what order I connect 2 batteries at different voltage in series to UPS in such case, i.e. 6V first and 14V second, or wise versa? And how to identify, what "first" means - is that the battery hooked to red Power wire?

Another Q: why the Back-UPS allowed the batteries to discharge to 6V at no load?

Relevant Q: I've periodic video spikes with loud click sound on the monitor hooked with PC to that Back-UPS Surge Protection ports. What can cause these loud spikes - voltage spikes in the mains, bad UPS, too cold or bad GPU board, or bad video cable hookup? How to figure it out, and can these spikes damage the monitor? I noticed the spikes slowly vanish as the GPU gets warmer, and at times when I reseat the HDMI cable.

[sambul83]

Re: APC Back-UPS 1500 erratic behavior

This is not a good idea

In this case, how I can charge just one battery with the UPS, since it takes 2 batteries as a set, and now they are at different voltage?

Can someone suggest a good but rather not expensive standalone charger for Back-UPS type batteries? I've some Power Tool 12V chargers for Li-Ion batteries, I assume they won't do it?

[sam_sam_sam]

Re: APC Back-UPS 1500 erratic behavior

I disassembled the fully discharged pack, and put its batteries to another UPS, so they're slowly charging, at 12V now after 6 hours.

If this was on the two new batteries and the voltage on both batteries were the same exact voltage then disregard this post

When you did the load test on these batteries how long did they last like a new battery would if so ( I will explain later)

If not then your battery or batteries are tired and need to be replaced

I be very concerned about the battery that was at 6 volts this is a red flag that something is wrong with it even if it changes up to 14.00 volts when you do the load test it tell you what going if you load test until the battery voltage goes to 10 volts how long does it do this test

[sam_sam_sam]

Re: APC Back-UPS 1500 erratic behavior

Another Q: can 2 batteries at different discharge level be charged together by the UPS as a pack? Like one is at 14V, and another at 6V, but surprisingly keeps charging?

This is not a good idea here why
The battery that is 14 volts may do one of things either will keep charging or the voltage will keep going

The 6 volt battery will either keep charging until it gets to 13 volts or ack like very high ohm resistor and not charge or ack like a short and become very warm or very hot

More Q: can I use together a 3-year old good battery and a brand new one as a pack? If not recommended - why?

This is just not a good idea for some of the same reasons that was given above

[sambul83]

I recently had to replace a battery pack on APC Back-UPS 1500, since it stopped charging after 3 years of service close to a heat source, and upon inspection one battery get cracked and swelled on one side. I got a brand new APC battery pack, reset the battery date in PowerChute, then Brain Dead the UPS, fully discharged and charged the new pack a couple of times to reset the microchip and train the battery. Measured voltage on the pack, it was 30.8V charged.

Two days later there was heard a pop after a relay kick and 2 beeps every two seconds started. Back-UPS was showing various errors at each reset, like F05 bad charger, next time F06 relay weld, etc. I took out the pack, and suddenly its voltage dropped to 12.6V, i.e 6.3V per battery as measured. The UPS was still showing full charge. During these 2 days no load was attached to the battery outlets, only a small load to Surge Protection outlets, and there was no surge in mains that I noticed.

I put another good battery pack into the UPS, and it restarted seemingly normal operation with all tests passed, and discharge & charge cycle under a test load 30%. I disassembled the fully discharged pack, and put its batteries to another UPS, so they're slowly charging, at 12V now after 6 hours. It seems everything works as it should: the UPS works again with another pack, and the discharged batteries are being charged again and pass the start test in another UPS.

My Q is: what happen? Why fully charged new batteries get fully discharged below limit in 2 days, while no load was attached to them? If they're short, why they recharged again in a different UPS? May be the UPS caused a short to batteries, or some component gives intermittent fault? What should I do now?

Another Q: can 2 batteries at different discharge level be charged together by the UPS as a pack? Like one is at 14V, and another at 6V, but surprisingly keeps charging? How in this case the charge levels between batteries, and how the UPS decides that the pack get fully charged? How generally the UPS measures a battery charge level?

More Q: can I use together a 3-year old good battery and a brand new one as a pack? If not recommended - why?