US L16HCXC 420amp batteries in 24V bank

TexSun

Our 1st bank for off-the-grid solar contained Trojan J250's (20). These lasted from 2001 to summer '09 and then began to fail slowly one by one; some are still good. I thought 8yr life indicated we have learned proper care. But our 2nd bank, US L16HCXC 6v 420amp (16) purchased in summer'09 before US's new RE L16's for solar was announced, is failing rapidly. One cell of one battery in a series set fails, causing overcharging of another battery of same set, discovered quickly due inconsistency of readings on the TriMetric meter. System: Trace 4800 24v inverter, 3.1kW solar array, (2) 60amp charge controllers. We reacted in a panic by replacing 3 of the batteries with the same model and another one has failed.

Time for stop-loss on wrong battery for application? US Battery tech didn't seem to speak "solarese" and we need seasoned advice.

BB.

Re: US L16HCXC 420amp batteries in 24V bank

Sorry for the reasons of your first posting on the forum.

Hmmm... Sounds like you treated the first bank "right"--So, did you have any indications of things going wrong with the second bank (possibly even when the first bank was failing--I.e., did the first bank actually fail for cause)? What charge controllers do you currently have? Are you using remote battery temperature sensing?

Charging/discharging voltage? Water usage (higher/lower)? Any problems with over heated cable ends/battery terminals? What are your charging set points, what specific gravity do you measure, how deeply (voltage/specific gravity) do you discharge the battery bank? Does the battery bank get hot (during summer/heavy loading-charging)?

Are the banks wired with "equal" cable lengths for each string (see this web page for battery wiring)?

If you do not have one, I would suggest getting a DC Current Clamp type meter. Use the meter to check the charging and discharging current between all four (?) parallel strings and see that they are properly sharing current. This meter is fairly inexpensive and is "good enough" for debugging a DC battery system.

Anyway--That is where I would start. Hope you get the problem(s) under control. Batteries ain't cheap.

-Bill

vtmaps

Re: US L16HCXC 420amp batteries in 24V bank

Like Bill, I have more questions than answers at this point. Let me add one more to his reading list:
http://forum.solar-electric.com/showthread.php?14674
--vtMaps

niel

Re: US L16HCXC 420amp batteries in 24V bank

the technician shouldn't actually care the charge source was solar as long as the charge source delivered the power properly and it sounds like you do have it right or the previous bank would not have lasted. go over your settings again informing us of the layout and settings you have and then recheck your wiring. if all is as it should be then try talking to them again and if they shrug their shoulders again then ask to talk to somebody higher up who does know the score. if they all claim ignorance then you bought the wrong battery make.

mike95490

Re: US L16HCXC 420amp batteries in 24V bank

Your 2nd bank, [ But our 2nd bank, US L16HCXC 6v 420amp ] is it a series bank, or are they series with parallel ??

TexSun

Re: US L16HCXC 420amp batteries in 24V bank

Sorry for the reasons of your first posting on the forum.

Hmmm... Sounds like you treated the first bank "right"--So, did you have any indications of things going wrong with the second bank (possibly even when the first bank was failing--I.e., did the first bank actually fail for cause)?

1st bank: Indication of gradual failure was on meter reading of indivual batteries and hydrometer readings plus uneven electrolyte levels (failing cell not "using"). Due to the 5 sets of 6V for the 24V system in the 1st bank, we occasionally moved batteries around within the bank. But these were 72 lb.; the L16s weigh more than I do.

What charge controllers do you currently have? Are you using remote battery temperature sensing?

Original 60amp controller was Trace (before 1st buyout); second, added with addition of more panels, was 60amp Xantrex (before 2nd buyout). Solar panel input to each controller is balanced within 0.2 amps. By sight, the circuit boards of these controllers are identical, but it's very hard to exactly match the settings of two devices with age differences (2001 & 2007).

Charging/discharging voltage? Water usage (higher/lower)? Any problems with over heated cable ends/battery terminals? What are your charging set points, what specific gravity do you measure, how deeply (voltage/specific gravity) do you discharge the battery bank? Does the battery bank get hot (during summer/heavy loading-charging)?

Bulk - 28.2, Float 27.6 US Battery tech said they sh/be 30 and 27. This area had 105 days of 100 or + degrees F. days, drought and blowing dust. Water was rationed and pressure here wasn't sufficient to wash off panels. Normal charge on sunny day is in 90amp range. Can't be specific about discharge as freezer & refrigerator cycle more often in hot weather, fans also used. Year-round is satellite internet, nighttime satellite TV, freezer & refrigerator which shows 9 to 11 amp dischage on TriMetric meter. Voltage falls from 29V to 25.2V under these loads when charge stops (both the J250s & L16s). If it falls to 24.3 before 2am, generator is run. When only load is battery room DC light, a 24.3V reading rises to 24.6V. (Disconnecting the system long enough to get readings at 72 degrees F. would result in food spoilage if it could be done - impossible from April to October.)

Yes, batteries do get hot - this is Texas.

Watering schedule during the first 2-1/4 years was 30 days ~1-1/4 gal. for the 16 6V L16s - schedule changed to 2 weeks after first cell failed. Failing cells overcharging others.

Are the banks wired with "equal" cable lengths for each string (see this web page for battery wiring)?

Cables are from 6-1/2" between those in the series set (short as possible on side-by-side negative-to-positive arrangement) and 27" between paralleled sets (end-to-end of sets). Cables from inverter enter the box from the front at mid-point and are each 7-ft. I do not understand the diagrams on this site unless the voltage application is the same as that of one battery or unless that box represents a series set. If you have access to a Trace 4800 watt inveter manual, our arrangement is that shown Techanical Information section. Our bank is not cross-wired; should I try to cross-wire without blocking access to the cells, cross-wiring cables would have to be long enough to fasten to the back of the battery box. If that would help to balance the charge-discharge, I will try. Please advise.

If you do not have one, I would suggest getting a DC Current Clamp type meter. Use the meter to check the charging and discharging current between all four (?) parallel strings and see that they are properly sharing current. This meter is fairly inexpensive and is "good enough" for debugging a DC battery system.

Link did not work but found others that did. I'd need a good gas mask to move the clamps around while charging and opening and closing the lid.

Anyway--That is where I would start. Hope you get the problem(s) under control. Batteries ain't cheap.

Thank you for your help and suggestions. Spoke to another experienced solar off-the-grider and no one has any long-term experience with the US RE batteries as only on the market about 3 yrs. In that time US has changed the wording on the L16HCXCs on their site which I feel speaks more to the scrubber, sweeper, forklift, etc. applications wherein equipment to used until discharged, then recharged offline.

vtmaps

Re: US L16HCXC 420amp batteries in 24V bank

TexSun, heat kills! You got 8 years from your first bank.... I think that's very good considering the heat down there.
Both your first bank (5 strings) and your second bank (4 strings) are what I call 'Massively Parallel'. When you changed from 5 strings to 4 strings you obviously changed your wiring, but you indicate (in your 2nd post) that you don't understand the wiring diagrams at smartguage.com. Getting the wiring right is critical in any parallel system. Can you submit a schematic or a photo of your wiring to this forum for us to evaluate? Read (or reread): http://forum.solar-electric.com/showthread.php?14674

You did not answer BB.'s question about whether you have a battery temperature sensor. At high temperatures the charging voltages must be reduced or you will 'cook' your batteries. That is true even with one string of batteries. And even if you do have temperature compensation it is difficult to get all parallel strings charged properly because if one string is just a bit warmer than the others (inevitable) it will draw more current than the other strings which will make it warmer which will make it draw more current which will make it warmer which will make it draw more current, etc.

The current clamp meter will help you figure out whether you have equal current flowing in each string, whether because of wiring problems, different string temperatures, or battery failure. It is best to take the measurements while bulk charging.

Assuming you get the wiring and temp compensation under control, you might want to think of ways to keep the batteries cooler. I'm in Vermont and don't have to deal with that issue, but there have been several discussions on this forum... I recall reading about swamp coolers, battery sheds in earthen berms, a water cooling system (circulate water underground to cool it and then into the battery box, and then back underground).

Hope this helps, and let us know what you find diagnostically with your clamp ammeter.
--vtMaps

SolaRevolution

Re: US L16HCXC 420amp batteries in 24V bank

Here is a link to the US Battery charging instructions PDF:
http://www.usbattery.com/usb_images/charging_instruction_2011_2.pdf


Your new battery bank is very different from your old one.

Things to consider:

With 4 parallel, 24v strings of 420ah batteries you have a 20hr rated capacity of 1680ah.
Your old ones were 250ah x 5 strings for a 20hr rated capacity of 1250ah.

Your 3.1kw solar array operating at 80% efficiency of STC would produce <100amps (< 6% C20 rate) of charging current for your new battaries.
This would have been 8% of C20 for your old batteries.

Having 4 parallel strings is a bit of a "performance limiting design factor" so your batteries are probably not getting the full advantage of the charging capacity that you have.

Let me go out on a limb here. My opinion, for what it's worth and from my experience as a service tech for many old systems, is that Trojans are "easier", that is, they seem to be more forgiving of less-than-ideal charging sources than some other "high quality" batteries. I was just at a job which had some 9 year old Trojan T-105 "golf cart" batteries which had never been equalized and had been working "satisfactorily" (client's claim) until the recent rains. Usually if you get more than 5 years out of "golf cart" batteries you are doing quite well.

Alex

niel

Re: US L16HCXC 420amp batteries in 24V bank

furthering vtmaps mentioning of how heat kills that this is a high possibility of what is going on here as it seems to fit. initially you were charging at what most perceived to be too low of a bulk charge voltage. this may have saved the first batteries if you weren't using a bts as the lower voltage prevented damage to the strings. i do agree the charge voltage should normally be higher, but that is generally at a battery temp of 77 degrees f and when ambient temps are higher the batteries can't dissipate heat any lower than that of ambient making the battery temps most likely far higher than ambient. i probably overlooked it in your answer, but if there i'll bet you had a much higher need of watering as i believe they may have been cooked due to no temp compensation or improper compensation when you upped the voltage. also exposing the plates to the air will lead the batteries downhill too so proper watering is essential.

i'm not saying this is a certainty or only cause for failure as we aren't right there to be sure you didn't overlook something else, but i'm leaning the batteries did cook because the low voltage you were feeding them prior did not cause battery failure due to undercharging as evidenced by them lasting this long.

BB.

Re: US L16HCXC 420amp batteries in 24V bank

Everyone else has pretty much replied what I would have said... Your battery voltage charging set point seems to be a bit on the low side but you are using a significant amount of water which would indicate higher voltages.

The float voltage is a bit high, could lead to higher water usage--But if you are cycling the batteries daily, it should not be a big issue.

Have you measured the actual charging/float voltages with an accurate volt meter--Especially when the bank is hot?

How often/how many minutes/hours do you equalize the bank?

I would not "cross wire" a battery bank. If everything is working correctly, cross wiring will not help anything. If there is a problem (open/shorted cell, failing cell), cross wiring makes it more difficult to find and diagnose a bad cell/battery.

Do you have good ventilation for your battery bank? Hydrogen gas and electrolyte "spray" can cause problems. Are the cells kept "too full"? Is electrolyte spilling out on the top of the battery bank?

It would be nice to have more solar panels for your larger bank.

The fact that your first bank lasted pretty well, it sounds like your basic system is OK.

-Bill

Photowhit

Re: US L16HCXC 420amp batteries in 24V bank

I think one thing that hasn't been discussed is if the batteries have ever been equalized?

Trojans don't recomend equalizing(unless batteries are out of balance) but most recomend equalizing every month to 6 months. I read the US Battery link provided by someone else enough to see the have an Equalizing charge every 30 days for 1-3 hours.

Also you migh be sure about the terms you are using with your battery tech, 28.8 is what the data sheet says (+ or - .05v per cell) and Equalizing at 30.6V (+ or - .05v per cell) I have a 800A 24 volt battery and charge at 29V float at 27.4(Ithink) I equalize at 31.2or4.

As others have stated/suggested/implied you may not be fully charging your battery bank. Doing an Equalizing charge would be the first thing to do as batteries get out of balance (SG reading vary between cells)

Another questions would be did you reset/reenter your Trimetric with your new battery capacity?

TexSun

Re: US L16HCXC 420amp batteries in 24V bank

HEAT
Originally there was one vent point in floor under inverter (coolest air is from under house - house is on piers) - bulk setting calculated compensating for summer heat using Trojan specs. When the automatic 3-speed inverter fan switched to highest speed one day, another larger vent in the floor was added nearest the battery box. We experimented with adding a 12v muffin fan on top of floor vents - then with an input fan at the side entry to the battery box - then an inline fan in the vent stack to the roof which exits the batterybox at the opposite end.
Results in inverse order: The inline fan in the vent stack actually inhibited air flow when not used requiring power 24/7; convection produced as much air flow when metal roof was hot (determined by using thin paper across the battery box inflow point to indicate suction). With inline fan moved, temperaturemeasured with/without the muffin fan on top of 2nd floor vent - no difference as same convection pulled air into battery box. These were the J250s, 1250 amp hr bank,1.9kW array. Reduced the bulk setting using the difference in the temperature compensation readings on the inverter meter.
With addition of panels to 3.1kW, bank was actually 10 yrs. old and beginning to fail very slowly. Still have 16 good ones. Built a box outside for them with lines from a 30 amp 24V battery charger mounted in power room intended as backup. Replaced with the 16 US L16HCXC 420s (1680 amp hrs.) only because nearest distributor, 200 mi trip, had dropped the Trojan line, and added the 2nd Xantrex C60 charge controller. Added a vent hole in the wall of power room above the battery box - left uncovered March-December. First watered the batteries to the split ring like the J250s and had indication around 1 cell that that was too high and changed the watering level to 1/4" below the bottom of the ring. For 3 summers there were no problems - able to run a high efficiency mini-split air conditioner to sunset without the $propane generator.
Sudden cell failure started in November 2011 after heat of summer passed and that's when water usage jumped on all cells. Normally I would have increased the bulk setting for winter but didn't because of the excess water usage.
Equalization for the J250's was to 31v for 2 hrs. ~30 days but TriMetric meter now displays 30.7v as highest point.
Panels mounted on roof both north/south slope - each panel on 4 aluminum rails drilled in three side positions for tilts to be within 7-8 degrees - north row at top mounted high enough on rails for no shadow at bottom of panel from ridge. This permits max air circulation. Angles changed 3 times per yr. but one slant change can be skipped and still be within the 15 degrees. Metal roof has continuous ridge vent; max temperature in attic has been 125 degrees. First 13 panels were MSX-120's, actually two 60 watt mounted in one frame which can be wired for 12 or 24V and one side was wired for 24V when lightning destroyed the other. Mounted 12 as soon as roof was finished June 21-25 2001 & could use power tools to finish the house - other panel was used on the 12'x16' tool shed for small DC refrigerator, lights and DC fan. When BP bought Solarex, this panel was discontinued - added three BP 130 watts after moving in 2004. 1.92kW array, 1250 amp hr. bank, 1.5 ratio, none of it same age. Added 10 Mitsubishi 125W (chosen by the frame size to match rails/structs spacing to avoid ridges in metal roof), these 16 L16s, the second C60 amp charge controller, and the new panels all within 3 weeks in May-June 2009 - 3.1kW, 1680 amp hr. bank, ratio 1.8. Using 12 of those batteries, 1260 amp hr. is ratio 1.8.
Have been reading the replies and looking up system components used trying to determine ratios of arrays/banks. During that time, many visits to the power room checking water/state-of-battery voltage and a third factor has now surfaced - the newer Xantrex C60 charge controller was going from normal digital display to display failure and light indicator of "Load Disconnected" which sh/not happen when jumpered as a charge controller, then back to normal again. Replaced the jumper but on day 3 it stopped returning to normal so is now disconnected - only half the array is charging a 1260 amp bank right now. (4 batteries are disconnected pending decision on the original stop-loss question). Don't know if this controller malfunction was a source of problem or there's a problem between controller and panels or if just defective - may know when replaced with the new C60 en route; disassembling/reassembling/testing. Wires from panel pairs enter the attic to two centered combiner boxes which each have lightning arrestors, ground wire from each runs straight down through a double wall to common equipment grounding for entire system. Moving the wires from one combiner box to the other is easily done to distribute flow to controllers but won't be done pending arrival of the new charge controller this week. All wires are on the shortest route; the attic midpoint combiner boxes are almost directly over the power room.
Removing 4 batteries from the box would provide room to arrange the other 12 and wire differently but still 3 strings. There is no practical way to place a bank outside of the house. The power room has common walls with the utility area and closet with insulation between but is kept much cooler than any outside structure could be. An AC evaporative cooler doesn't work well in humid climates - we had a small one for that 12x16 structure - any water circulation device requires extra loads.
Bill wrote, "It would be nice to have more solar panels for your larger bank."
Bank too large for array? That would be great news! We would rather reduce the bank to match the panels.
Given the verbose explanation of conditions, what amp hr. bank would you recommend?
Sure do appreciate all the replies.

Photowhit

Re: US L16HCXC 420amp batteries in 24V bank

Sizing a battery bank to the array is largely a matter of choice and lifestyle, but minimum charging is normally 5% of the battery capacity, Trojan recomends 10%. Using PWM C60 at an input voltage of @35 volt and array of 3.1Kw on paper will max out at @88 Amps X 20 = @1760amp with the heat an MPPT wouldn't do you a lot to improve this. Typically panels will only produce @90% of label current but your in the minimum ballpark.

Were you running an AC before the battery change? perhaps you've been undercharging the batteries for 3 years, Even the minimal draw of a mini split is @350 watts, after inverter ineffiecentcies, your taking 500+ watts off your charging abilities (lifestyle stuff). The multiple strings might asentuate(?) the problems.

I'll reread your earlier post later in the day...

If you use Trojan's guide lines @ 880 Amp hour battery. Also are the old batteries also tied to the array? if not how are you keeping them topped off.

stephendv

Re: US L16HCXC 420amp batteries in 24V bank

How long is your absorb time set for? I ask, because I borrowed someone's C40 charger and it had no programmable absorb time, so it was stuck at 1 hour- which is too short for any decent sized bank. Can you change your absorb time to 2-3 hours and then measure the SG once it hits float and see whether they're really fully charged?

TexSun

Re: US L16HCXC 420amp batteries in 24V bank

Photowhit wrote: »

Sizing a battery bank to the array is largely a matter of choice and lifestyle, but minimum charging is normally 5% of the battery capacity, Trojan recomends 10%. Using PWM C60 at an input voltage of @35 volt and array of 3.1Kw on paper will max out at @88 Amps X 20 = @1760amp with the heat an MPPT wouldn't do you a lot to improve this. Typically panels will only produce @90% of label current but your in the minimum ballpark.

Were you running an AC before the battery change? perhaps you've been undercharging the batteries for 3 years, Even the minimal draw of a mini split is @350 watts, after inverter ineffiecentcies, your taking 500+ watts off your charging abilities (lifestyle stuff). The multiple strings might asentuate(?) the problems.

I'll reread your earlier post later in the day...

If you use Trojan's guide lines @ 880 Amp hour battery. Also are the old batteries also tied to the array? if not how are you keeping them topped off.

The highest voltage setting on a Trace or Xantrex C60 charge controller is 30V. Temp compensation on inverter shows a difference of 0.4V (night) to 1.6V (~3.00pm) on a 80 degree F. day.

Yes, we've run two window ACs with both the Trogan J250s and these US L16s - these were alternated and turned on/off according to the amps readings. Also a 700W microwave oven up to 12 mintes a day; 1440W vacuum cleaner when the sun is available up to 5-7 minutes occasionally; toaster (1-1/2 minutes); biggest user is freezer which uses 3amps DC at 24V when running, 13amp DC when starting . Can list amps each application uses as we watch the meters but that isn't our situation here - no usages have changed. A major difference is the addition of the 10 120W panels.

No, I built a housing just outside the power room for 8 of the old J250s still good and used a 30V battery charger to charge them when we had a string of sunny days once every ~30 days just to have a backup. Haven't needed; and now haven't been charged since trouble started in Nov'11.

I now believe heat is the factor - the same battery box is being used and the taller L16's leave much less room between the top of the box and the batteries than those J250s. According to all prior experience, those 80 degree days shouldn't result in a 1.6V between actual & compensated. Five have dead cells, two more are fading fast and can't be charged above 5.8amps.

It shouldn't be difficult to increase the highth of that box - if that doesn't do it...

TexSun

Re: US L16HCXC 420amp batteries in 24V bank

stephendv wrote: »

How long is your absorb time set for? I ask, because I borrowed someone's C40 charger and it had no programmable absorb time, so it was stuck at 1 hour- which is too short for any decent sized bank. Can you change your absorb time to 2-3 hours and then measure the SG once it hits float and see whether they're really fully charged?

Charger doesn't have a timer - changes from bulk, to absorb, to float based on bank voltage.

TexSun

Re: US L16HCXC 420amp batteries in 24V bank

Think I'm getting posts out of order here -

Yes, equalizing was done ~days days at 31 for 2 hrs. Can't be done now - of remaining batteries, only three will charge over 6V - others are 5.9V to 5.0V.

Photowhit

Re: US L16HCXC 420amp batteries in 24V bank

TexSun wrote: »

The highest voltage setting on a Trace or Xantrex C60 charge controller is 30V. Temp compensation on inverter shows a difference of 0.4V (night) to 1.6V (~3.00pm) on a 80 degree F. day

It does have an automatic and manual equalizsing feature. Have you ever equalized this battery bank? it is standard procedure for these batteries, check the link someone provided. If you have dead cells this bank is likely toast, but the first defence when cells get out of balance is to do an equalizing charge.

Have you read the battery FAQ's?

Equalizing can be done as a regular maintainance or, like Trojan recomends for ther batteries, through checking the SG(specific gravity) of each cell and monitoring the balance between cells until the difference is minimal. I'm guessing that it has to do with Trojans specific matting between cells. Most tall cells like/need equalizing to help mix the electrolite, IMHO.

Photowhit

Re: US L16HCXC 420amp batteries in 24V bank

TexSun wrote: »

Think I'm getting posts out of order here -

Yes, equalizing was done ~days days at 31 for 2 hrs. Can't be done now - of remaining batteries, only three will charge over 6V - others are 5.9V to 5.0V.

I was doing some updating of my post, if they will not charge above 6V it's time to start over again... Did you regularly equalize this set?

Did your daughter turn 12 and discover guys and blow driers?

Sorry trying to make light of a bad situation.

You may be correct that heat helped damage the batteries, and perhaps Trojans are better protected, You migh consider a berm shelter for the batteries as ground temps 3-4 feet under ground are cooler. I'll do a berm shelter once I move my battery to it's new home.

For your next battery you might consider a forklift battery, it beats dealing with multiple strings, but does tend to want more care. Water miser caps can help with the need to water cells so often.