General > General Technical Chat

Noisy power from typical backup generators

<< < (11/19) > >>

Renate:

--- Quote from: Johnny B Good on March 03, 2021, 04:08:43 pm ---...before the voltage dropped to the cut out point at 40.4v.
--- End quote ---
Wow, you run the batteries hard, 10.1 V cutoff?
I used to see things pointing straight down as soon as I hit 11.5 V.
I guess that I'm a wimp, under 200 W load I like to keep them above 12 V.

james_s:

--- Quote from: Renate on March 03, 2021, 05:11:31 pm ---
--- Quote from: Johnny B Good on March 03, 2021, 04:08:43 pm ---...before the voltage dropped to the cut out point at 40.4v.
--- End quote ---
Wow, you run the batteries hard, 10.1 V cutoff?
I used to see things pointing straight down as soon as I hit 11.5 V.
I guess that I'm a wimp, under 200 W load I like to keep them above 12 V.

--- End quote ---

Well that's determined by the UPS, at least the little ones I have don't have any way of configuring that as far as I know. I try to fire up the generator quickly enough that they get drained very little but occasionally the power goes out when I'm not home and then if it stays out long enough it will drain them all the way down like that. Setting a higher cutoff reduces the run time significantly so I'm not surprised a UPS defaults to such a low voltage.

Johnny B Good:
@Renate

 I only run the battery right down to the UPS's cut off point when testing a new battery assembly to verify the expected autonomy, otherwise, with SLA batteries, it's best to avoid discharging them at all. A first time deep discharge of a new set of SLAs seems to be beneficial in doubling up their effective capacity (it seems to be more a halving of their internal resistance) going by my first encounter with this effect.

 The first time I replaced the worn out pair of 7AH SLAs in the Emerson 30 with brand new fully charged SLAs, I tested its run time with a 60W incandescent lamp load and only got 25 minutes.  The next day, with the battery fully recharged, I saw a run time of 80 minutes!

 It would seem a good idea to subject brand new fully charged SLAs to a 1C discharge test to properly condition them which, rather conveniently after a second run, provides a benchmark autonomy figure by which to compare against similar annual test runs. Otherwise it's best to avoid such deep DoD events with SLAs - LFP otoh, can be safely taken to an 80% DoD without harm until their original capacity has dropped to 80% several thousand cycles later.

 Standard practice with LFP is to limit their charging to 90% of their maximum capacity (3.2v per cell) and their discharge to no less than 10% (around 2.25 to, preferably, 2.5 volt per cell), hence the typical 80% of their full capacity that's normally used to determine their optimal total WH's worth of autonomy.

 Unlike LA battery technology, LFP offers an almost unchanging (and very predictable) WH capacity with varying discharge rates due to their much flatter discharge voltage curve and almost non-existent Peukert effect, as well as discharge/charge cycles counted in thousands rather than the tens of cycles of such abuse in the LA case. Even when you've hit, say, the 2000 cycle limit with an LFP battery, that just means you've only lost 20% of its original capacity leaving you with another 2000 cycles before you lose another 20%.

 That wear out characteristic is what makes the acquisition of secondhand LFPs such an attractive proposition. A used 120AH LFP reduced to 80% capacity is the equivalent of a brand new 96AH LFP which, if obtained through a reputable specialist battery dealer, will be a minimum and likely to have more than a 100AH's worth of remaining capacity for typically less than half the price of a new 100AH LFP with the bonus of the BMS being rated to handle 120A rather than the 100A limit of a new 100AH battery pack.

 That set of 7AH SLAs I'd bought to test the compatibility between my ancient SmartUPS2000 and the Parkside inverter genset will be my last purchase of SLAs. My next UPS battery will definitely be an LFP battery pack. I've had a gut full of the concept of the expensive SLA battery pack as "A Consumable" with a service life at best (2.25v per cell float charging voltage) of 6 years tops.

NiHaoMike:

--- Quote from: Johnny B Good on March 04, 2021, 02:18:43 pm ---Standard practice with LFP is to limit their charging to 90% of their maximum capacity (3.2v per cell) and their discharge to no less than 10% (around 2.25 to, preferably, 2.5 volt per cell), hence the typical 80% of their full capacity that's normally used to determine their optimal total WH's worth of autonomy.

--- End quote ---
LiFePO4 is normally charged to 3.6V/cell with 3.45V/cell being a good derating, you might be confusing it with LTO.

Johnny B Good:
@NiHaoMike,

 Thanks for that reminder. I was being a little lax in referring to the "Nominal 4 cell battery pack" 12.8 voltage equivalent to the 12 volt SLA it replaces. It's been a few weeks since I last researched LFP battery usage for UPS, Solar power and off grid, so am a little hazy on the finer details.

 Now you mention it though, ISTR seeing figures of 3.6v (and even 3.65v for a complete 100AH 16 cell battery pack I'd tried bidding on back then). Also, ISTR that battery pack, which went for 10 times my upper bid limit, had a low voltage cut off set to 40v. I suspect its BMS had been tuned for use with a 48v powered UPS to allow it to remain in control of the low battery shutdown process rather than chance  'having the rug pulled out from underneath it'.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod