Actual Charging voltage SMARTUPS 1000

[Anonymous user]

Hello
I pulled a 6 year old SMARTUPS 1000 rack mount from service and replaced it with a much smaller unit (customer was wayyyy over sold initially)
The old unit had a battery fault light on it; I ordered a replacement battery pack for it. After it arrived, I attempted to replace it and was unable to get the old battery pack out due to VERY swollen batteries. I ended up drilling rivets and cussing a bit and finally removed the swollen battery pack from inside the unit instead of sliding it out. Anyway, at the moment, the ups is not under load, not connected to a computer at all and it has been plugged in and charging for 2 or three days. At the battery pack disconnect plug I am seeing 27.68 volts DC.
My question is this... is that voltage too high (is this even any indicarton of the actual charging voltage anyway?) and if it is too high is there any way to adjust the charging voltage/amperage?
When this unit was in service, it saw very infrequent use; the hydro here is pretty reliable all in all really. The environment may be less that ideal in the summer time; not very climate controlled.
I just don't want to bake my brand new expensive battery pack like the old one very obviously was.
I am going to connect it to a computer with powerchute installed and see what kind of information I can glean from that.
Any help would be greatly appreciated!

Thanks
Rusty

[NoMoreJuice_apc]

I know this is an old thread, but nevertheless, it shows that something CAN be done which was declared as undoable.

My background on this issue:
We have transmission equipment at ~ 3000m geographical altitude, backed by 700 XL units, at approx. 10% load. The minimum run-time requirement is 24H after a power outage. 24 hours power failures are not uncommon on the slopes of a Volcano.
Initially, we had ALL of the SLA and AGM type external batteries failing, as little as 6 months after installation, no matter what capacity in AH we were throwing at it. It was not a matter of capacity, it was a matter of very little time until failure, and that was a very expensive experience.
I even chopped up some of the failed AGM batteries to find the cause of their premature demise, and they were all completely dried out, no trace of humidity was left in the cells, which are supposed to contain a gel of sulfuric acid. (Safety note: Don't do that at home without proper knowledge. Sulfuric acid is Hazardous Material).

My conclusion is and remains: If the charge (float) Voltage is too high, then all SLA and AGM batteries will vent through the pressure valve. This is by design. An altitude of 3000m aggravates the situation severely because of the lower air pressue, meaning that the VRLA battery is venting much earlier compared to sea level.

Now for APC's pre-set charge Voltage:
CSB batteries is a primary supplier for APC concerning batteries.

CSB batteries, at:
http://www.csb-battery.com/english/03_support/01_support_detail.php?fid=1&id=8&page=2&action=page

- recommends: 2.275V/cell for stand-by float charging. That is: 27.3V on a 24V nominal battery string, at 25 degrees Celsius.

All APC units with 24V nominal battery Voltage I had hands on charge at least to 27.87V, looking at the UPS management card, some up to 28.01V. My Voltmeter tells me 27.7V. And yes, I have taken temperature into account.

So my next conclusion is of course: The APC UPS that I work with (700 XL, 1400, 1500 with 24V batteries) notoriously apply a too high float voltage on the batteries. 0.4 Volt does not sound like much, but in the long run it can and will be damaging, and the allowable tolerances are narrow.

For our transmission equipment, I decided to go for wet cell types. That helped a lot,
however, a float charge of 26.4V and an equalization charge every 3 months would be more adequate for this type of batteries. With APC's pre-set of 27.8 Volts on a 24V string, they are beyond the gassing voltage, and require constant replenishment of water.

Bottom line:
It all cooks down to: how to adjust the float charge voltage on APC UPS.
APC is not divulging schematics, and the sound from APC technical support is: "Can't be done, we are right, that is the factory default, that is an unsupported feature, that is the factory recommended setting" etcetc. (I'm not quoting, but that is the tune).
While APC deviates from the recommendations from their own suppliers, SIC!

After more than two years of hunting for schematics for the 1400/1500 "class" control board, I found it and there is a solution for the float charge voltage. To modify it, you will need some basic electronic skills and skills with handling a soldering iron.

Needless to say that any modification of an APC UPS will void the warranty, bear that in mind before you go to work, and everything you do, you do at your own risk.

The charge circuit is hard-wired and controlled by IC14. This IC is close to the fan
connector and marked APC2020B. The battery voltage is measured via a voltage divider of 100K for the upper and 22K1 for the lower resistor. These are R118 and R119 and the resulting voltage is fed to pin 13 on IC14. Notably is that the resulting voltage on pin 13 is exactly 5.0136V at an input Voltage of 27.7V when you calculate the voltage divider, and assuming that the divider has practically no load. Apparently, this voltage is compared to a 5V reference in IC14.

The interesting resistor is R118. Recalculating the voltage divider, we need about 95K for R118 in order to get 5V at a feed of 26.5V. Practically, you will need a 95K resistor and a 5K spindle trimpot to be able to regulate between ~ 26.5 up to 27.7 volts. Being short on the exact value, I took 3x32K resistors in series with the trimpot and replaced R118 with that neat arrangement, and I was lucky that due to tolerances the resulting resistor was exactly 100K with the trimpot at max.

Now, you don't have to wait for hours for the battery voltage to stabilize in order to adjust the float charge voltage. Connect the ups, wait until self test has finished and disconnect the batteries. Then measure across the battery terminals at the UPS.

This information applies to the Smart UPS 700XL and 1400/1500 types. Most likely some others, like the 1000, too.

[EZ_Button_apc · Accepted solution]

About 27.5 is what you should be seeing on the battery charge. If the evironment is going to continue to be hot like you said in the summer, in the end you're probably going to run into the same or a similar situation.

[Anonymous user]

That's normal. I would like it just a wee bit lower, but you can't adjust it. On 80s and early 90s APC UPSs, you could tweak a pot and change the float voltage, but on newer units, I believe it's in part of factory calibration constants within the EEPROM.

Message was edited by: KVAr

[Micro-Masters]

Thank you VERY much for your detailed information. I'm wanting to put LiFePO4 batteries in, and this is just the information I was looking for.