Batteries never charge 100%

2»

Comments

  • lkruper
    lkruper Solar Expert Posts: 115 ✭✭
    BB. wrote: »
    OK, Rolls is saying that the absorb time is 0.42 * Capacity / charging current.

    So, a 50% SOC battery at 20% rate of charge would be:
    • (0.80 - 0.50 state of charge) / 0.10 rate of charge = 3 hours of bulk charge
    • 0.42 * 800 AH / 80 Amp charging = 4.2 hours Absorb
    • Total charge time = 3+4.2=7.2 hours of total charge time

    At the end of an estimated 7.2 hours, the charge controller should return to float charging state (i.e., just high enough voltage from the charge controller to keep the battery from charging or discharging).

    AGM are still a form of Lead Acid Battery... So, I am not sure that the 0.42C/I factor is any shorter.

    Here is Rolls' AGM charging FAQ:

    http://support.rollsbattery.com/support/solutions/articles/4345-agm-charging



    And they list total charging time (bulk+absorb) as <=16 hours.

    Hopefully somebody here with AGMs can give some experience about their Absorb Time with AGMs.

    -Bill

    Except the 0.42 factor from Rolls applies to their FLA batteries, not their AGM batteries. The Concorde AGM user manual gives time of charge to be AH charged / amps + 2 hours. The Rolls manual does not give a way to estimate total charge time, but it looks from their chart like the absorb is roughly the same as the bulk in terms of time. Their charge rate is lower 0.25C and upper 0.35C. So a 225AH Rolls AGM battery could be charged at 78.75 ah. At 50% DOD from 50% to 80% would take one hour. Another hour for absorb would be two hours. The Concorde way of calculating would be three hours total.

    Real world experience from someone would be very nice indeed.

  • BB.
    BB. Super Moderators, Administrators Posts: 33,433 admin
    Sorry--I have clarified that the 0.42 calc are for their FLA batteries.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Absorb time varys from battery tech, brand and model. While those formulas might be a guide ( assuming the right rule for the right battery), in real life, time is not that good a way to manage it. End amps is built into the better controllers, and most battery manufactures will specify a figure for the float transtion. For a tpyical AGM that will be 0.5% of the rated battery capacity , eg EA of 2A for a 400Ah battery. My bank will absorb for about 1-2 hours on a average charge cycle (90% SOC), but take 3.5 hours for a occasion where the bank was at 70% and getting on for 5 hours at 50% SOC. A backstop time is set but it isnt used, just the end amps. Lead acid banks are kinda painful in many ways, but kinda cool in other ways. That they like being held at 100% SOC just happens to match our desire for maximum power availabilty. That their long absorb taper charge occurs in the afternoon matchs the decline of irradiance post solar noon.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Time to revisit this I guess in regards to AGM. It really isn't that hard for cyclic solar. (Don't want to read? For cyclic apps, set BULK=ABSORB=FLOAT=14.4 (or 14.7 for pure lead) - done!)

    Often confused with gel, agm's are not THAT sensitive to voltage - but with some caveats.

    They have a wide range of voltages, and the *best* range is between 14.4v, and 15.0v. Here comes the caveat:

    If you run close to 15v, you had *better* be well temperature compensated - preferably with a probe on the battery itself.
    If you run at the low end of 14.4v, you had better not have any voltage drop from the cc to the battery.
    14.7v is a common "cya" middle ground recommendation - especially if temp comp or voltage drop is an issue.

    Note - to stay within the manufacturer's warrantee, stick with THEIR voltage range, even if experience shows different, high or low. Ie, I'll NEVER run below 14.4v personally, nor any higher than 14.8v. But that's me.

    The major problem that ALL manufacturers clearly state is that UNDERcharging an agm is bad. But what does this mean from an end-amp standpoint?

    It does NOT mean .05 or even .01C in absorb before switching to float. What it really means is much smaller like .001 to .007C. I might have left off a leading zero! Using C/20 for an end-amp scenario is a trigger for an ac-charger application that relies on a long float to finish the job. We don't have the luxury of float with solar to REALLY recharge an agm of good quality.

    Most documentation does not consider we are running in a solar-application, and thus a FLOAT of 6-8 hours or more is usually necessary to FINISH the 100% charge - otherwise under cyclic applications, the battery walks down in capacity over time. But we don't have that luxury unless you finish absorb an hour after sunrise - in which case your bank is totally oversized, and you may be damaging the battery with very shallow cycles. (sulfate clumping)

    Some conventional AGM's can only accept up to about 0.3C MAX inrush current. Exceeding this tends to create hot-spots within and eventual thermal runaway - so don't exceed that with conventional. Pure-lead (Odyssey/Enersys/Outback/Optima and so forth) have no such limitation. Their internal resistance is even lower than a conventional agm, and thus can take very high charge currents without going thermal-runaway as long as the CV absorb is tight! But let's get real - we aren't talking EV's from the last decade here. Most of us will never exceed even .5C for a bank designed to have any decent autonomy and pure-leads have no problem with that at all. Pure lead is great for areas of low solar-insolation, where one may want to save fuel with a genny, etc. Covered elsewhere.

    Our problem is consistent undercharge by dropping to float too soon. Either by time, or the fact that .01C end-amps is not good enough!

    So toss out the concept of float in a cyclic solar application. All we have left is an extended absorb, preferably near the high end (of whatever the manufacturer states is ok) to get the job done in as short a time as possible.

    Basically, all you need is BULK=ABSORB=FLOAT=14.4v (or even better 14.7 if the manufacturer allows - temp compensated!)

    Consider that Enersys (Odyssey / Outback / Enersys SBS agm's) have an "ideal" IUU (cc/cv/cv) that consists of an 8 hour absorb at 14.7v!

    Most of us will never have the luxury of an 8 hour absorb in a typical solar day. So it is best to just strive to jam it in the fastest you can with a *temperature compensated* absorb as long as you can daily, preferably higher up the scale. Forget floating.

    Again, this is for cyclic daily discharges of a reasonable depth, like 20% or more. Don't do this if you aren't cycling, but are in a standby application, where just hanging your hat at 13.5v might be just fine.

  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    If you can't get the batteries to a full charge every few weeks (because the time it takes is more than sun hours), then you can get an isolated ground, 120VAC to  12V charger and rotate it among individual batteries. 

    I am available for custom hardware/firmware development

  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    PNJ's thoughts two posts above are interesting. On a sunny summer day with lite overnite loads, our bank will start absorb around 10am, end absorb about noon (0.05C EA), then float til 7pm, where it will flatline at about 0.01C. That i'm comfortable with. On a different day, heavier overnite loads, down to 80%DOD, cloudy winter day, absorb starts at 1pm, ends at 4pm, floats for about an hour, reaching little below 0.05C.

    The one thing preventing me setting abosrb=float is that having 12x 2V cells in series, later in absorb the individual cell voltages can get quite high., the higher cells im talking. Which reminds me i need to log cell voltages again...

    The plan there is so that when the first cells start to die, ill know whether they were the weak cells that suffered routine undercharging, or the strong cells that suffered routine over charging.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    edited October 2015 #38
    You should check the SG of all of the cells, then do a top balance of all of them either with an extended Equalize or by charging them individually or in groups of three similar cells until they are closely matched in State Of Charge (SOC). From there you can see whether they track well or drift out of SOC matching.
    SMA SB 3000, old BP panels.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Theyre AGM. But shouldnt the long afternoon floats ammount to a top balance? Its not unusual to get 7 hours of float this time of year.

    My understanding is that cell variance is characterized is in terms of both capacity and internal resistance, the top balance will only address the former, not the latter. Which is what i assume is happening here. They get out of balance every cycle.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar