Absorb times...

Started by alyaz, April 10, 2019, 08:31:19 PM

Previous topic - Next topic

alyaz

Thought I kinda understood ‘absorb times’ but started questioning myself when explaining it to a friend today.  So is my reasoning faulty here?  This is how I explained it to him...

He completed the ‘Rolls FLA absorb equation’ and came up with 7 hours.  He only has a 1500 watt array and I don’t recall how many a/hrs his battery bank is.

Anyway, I told him that 7 hours is basically what his battery bank requires about every 7-10 days after he has been cycling it and it has not received a full charge.

So that got me to thinking, most people set their charge controllers up with that ‘magic’ number.  In his case he doesn’t use ‘end amps’ so his controller would be set up to run 7 hours of absorb every day - depending of course on how much solar he is pulling in.    During the summer it is feasible he could get 7 hours of absorb time on his battery bank every day for a period of time.  So that can’t be good can it? 

In my case, before I set up end amps with the Whizbang Jr, my controllers were set the same way (with the old battery bank).  Generally, my absorb times (using end amps) are about 2 - 3 hours now before I get to float.

So am I missing something here, or are absorb times using a ‘timer’ (without end amps) generally going to be excessive during the summer months and just overcharging and boiling off the batteries?

Just trying to understand this so I can actually speak intelligently about it, if that is possible.  Thanks.
3.3 kW solar. (1.5 kW Helios - MN Classic 150 CC; .8 kW CSun - MN Classic 150 CC; 1 kW Sharp - MN  Classic 250 CC); Magnum PAE-4024 Inverter, AGS, ARC, BMK. Whizbang Jr., Midnite Solar battery monitor, 5 Midnite SPD’s. 24V Rolls 4KS21P - 1105 a/hr FLA.  10 kW Perkins Diesel.

ClassicCrazy

#1
yeah 7 hours would most likely be way too long and bubbling  the batteries way too much -- might be harming them.
2 hours for absorb time  would be a better number to start with and see how it goes. 

If your friend has such a big battery pack and not enough PV to sufficiently charge it that isn't a good thing. Even if that was the case it would probably be in bulk a lot longer and then when it got to absorb it would start its count down then.

Whizbangs and ending amps take better care of the batteries when set up correctly because it will compensate for different charging conditions .

Larry
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

mike90045

The best Absorb system I've come across, is measure the time required to BULK the batteries, and use that time for Absorb.

That sort of covers your array size, nightly depth of discharge and costs nothing.
http://tinyurl.com/LMR-Solar

Classic 200| 2Kw PV, 160Voc | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph )| Listeroid 6/1, st5 gen head | XW6048 inverter/chgr | midnight ePanel & 4 SPDs | 48V, 800A NiFe battery bank | MS-TS-MPPT60 w/3Kw PV

Vic

Quote from: mike90045 on April 10, 2019, 11:00:40 PM
The best Absorb system I've come across, is measure the time required to BULK the batteries, and use that time for Absorb.

That sort of covers your array size, nightly depth of discharge and costs nothing.

In decades gone-by,   the venerable OB MX-60 MPPT CC would use this Bulk time = Absorb time,  if no EA,  and no Absorb time was set  --  at least with early FW versions.

The early MN Classics also had this option.   I was one those who was begging boB to remove this option,  as it  often would result in considerably longer Abs times than necessary.   This approach is quite variable,   depending upon WX patterns,  elevation/azimuth of PV array,   view of horizon,  etc.

On the systems in use here,   Bulk begins before Sunrise.   In the Summer,   Bulk begins around 5:30 AM,   and Absorb is often not reached,  until about 10 - 11 AM.   This could easily dictate an Absorb time of five, or so hours.   This would be about double the typical required Absorb time.

If one has  morning clouds/fog,   the Bulk time could be further-extended,   causing even un-necessarily-longer Absorb times.

Wb EA  is about the best approach that I've found.   Even that  has  some variables that seem to make it not quite perfect.

Absorb time for the main bank here is usually between 2:15,   and 2:30 hours.   If Bulk time was used,  it would usually be more than twice that.

That Surrette Equation for Absorb time seems to yield  poor results,  for off-grid systems.   Believe that it is for Grid-Charged systems,   with Constant Current Bulk.   RE-charged (solar charged)  batteries  do not have the advantage of prescribed,   controlled Bulk charge current (IMO).

Just my opinions,   YMMV.    Vic
Off Grid - Sys 1: 2ea SW+ 5548, Surrette 4KS25 1280 AH, 5.25 KW PV, Classic 150,WB, Beta Barcelona, Beta KID
Sys 2: SW+ 5548s, 4KS25s, 5.88 KW PV, 2 ea. Classic 150, WB, HB CC-needs remote Monitoring/Control, site=remote.
 MN Bkrs/Bxs/Combiners. Thanks MN for Great Products/Svc/Support&This Forum!!

boB

Quote from: Vic on April 11, 2019, 11:53:41 PM

In decades gone-by,   the venerable OB MX-60 MPPT CC would use this Bulk time = Absorb time,  if no EA,  and no Absorb time was set  --  at least with early FW versions.

The early MN Classics also had this option.   I was one those who was begging boB to remove this option,  as it  often would result in considerably longer Abs times than necessary.   This approach is quite variable,   depending upon WX patterns,  elevation/azimuth of PV array,   view of horizon,  etc.


Absorb time = Bulk time   (minimum I think it was ?)

wasn't terrible but it was the wrong thing to do, in hindsight...

The charge current in bulk might be limited for whatever reason like, maybe it was cloudy for most of the morning and then the sun comes out ?

Just have it absorb long enough time to keep sulfation to a minimum.  Don't even absorb every single day maybe to keep water use down and possibly increasing the life of the batteries while being able to use more of the day's solar energy for other things like pumping water or selling back or whatever.

Similar thing with using ending amps to end absorb all the time.  If EA somehow keeps the battery from getting a nice long-ish absorb cycle once in a while, then maybe the batteries will sulfate sooner ?

Just when you think a product has the best way of doing something.....


K7IQ 🌛  He/She/Me

Vic

#5
Quote from: boB on April 12, 2019, 12:39:29 AM

Absorb time = Bulk time   (minimum I think it was ?)

wasn't terrible but it was the wrong thing to do, in hindsight...

The charge current in bulk might be limited for whatever reason like, maybe it was cloudy for most of the morning and then the sun comes out ?

Just when you think a product has the best way of doing something.....

boB,

Was not trying to speak negatively about either the MX-60,  or the Classic.   Each of these CCs were and still are revolutionary.   SO many useful functions built into each of them.   No  other designers/manufacturers even try to come close to creating such useful products to so many users.

Think that the  Magnum PT-100 CC copied from this earlier work   ...

It seems that there is no real perfect solution for setting the needed Absorb time.   Just too many variables involved.

There are times when the EA value that was fine for the previous discharge/charge cycle  will simply not be reached in a reasonable amount of time,   etc.   It is for this reason that we Skip Days,   use Wb EA,   but EQ a bit more frequently,  than we might,   if we used  a fixed (and too long) Absorb times.

As you have noted,   it is much better to use Absorb times that are a bit too long,   verses ones that are generally too short   ...   thank goodness,   Flooded LA batteries are fairly forgiving.

Please DO keep up the great work,   boB.    We all are counting on you and the MidNite crew,   to continue the string of terrific new products.

Realize that all of the functions that are built into MN products take more development time,  debug time,   Support time,  etc.   And all of this makes your products  the absolute best available,   and may have somewhat smaller profit margins for MN,  than do products from others,  who are not "swinging for the stars",   as MidNite is doing.

Nothing is perfect,   but you guys sure are working hard at making your products far closer to perfection than any others on this planet  ...   IMO.

People like me,   were begging,   and begging you to ship the Classic.   We simply could not wait to get the Classic into our hands!   Had one of the very first Classics made  --  shipped in December 2010,   IIRC.

Thanks,   and was not trying to be snarky.     73   GL,     Vic
Off Grid - Sys 1: 2ea SW+ 5548, Surrette 4KS25 1280 AH, 5.25 KW PV, Classic 150,WB, Beta Barcelona, Beta KID
Sys 2: SW+ 5548s, 4KS25s, 5.88 KW PV, 2 ea. Classic 150, WB, HB CC-needs remote Monitoring/Control, site=remote.
 MN Bkrs/Bxs/Combiners. Thanks MN for Great Products/Svc/Support&This Forum!!

SolarMusher

Quote from: boB on April 12, 2019, 12:39:29 AM
Quote from: Vic on April 11, 2019, 11:53:41 PM

In decades gone-by,   the venerable OB MX-60 MPPT CC would use this Bulk time = Absorb time,  if no EA,  and no Absorb time was set  --  at least with early FW versions.

The early MN Classics also had this option.   I was one those who was begging boB to remove this option,  as it  often would result in considerably longer Abs times than necessary.   This approach is quite variable,   depending upon WX patterns,  elevation/azimuth of PV array,   view of horizon,  etc.


Absorb time = Bulk time   (minimum I think it was ?)

wasn't terrible but it was the wrong thing to do, in hindsight...

The charge current in bulk might be limited for whatever reason like, maybe it was cloudy for most of the morning and then the sun comes out ?

Just have it absorb long enough time to keep sulfation to a minimum.  Don't even absorb every single day maybe to keep water use down and possibly increasing the life of the batteries while being able to use more of the day's solar energy for other things like pumping water or selling back or whatever.

Similar thing with using ending amps to end absorb all the time.  If EA somehow keeps the battery from getting a nice long-ish absorb cycle once in a while, then maybe the batteries will sulfate sooner ?

Just when you think a product has the best way of doing something.....
Bob, I feel your pain, lol.
But..., why not add a third factor to end absorb. Overcharge percentage from Wbjr for example.
A+
Erik
Off Grid with 4kw PV | 2x Classic 200/WBjr | 2x Outback VFX3648 Epanel | 3x SPD300 + 1x Schneider HEPD80 | Hub + Mate + PSX-240 | Volthium 400Ah/51.2V LFP battery bank + Trimetric | 1500 watts AC water heater | Kubota 11kw GL diesel generator

Vic

Hi Erik,   boB,   others,

It would seem,   that having a Rate-Of-Change in WbJr current would also be quite useful.   We have used EA,  forever on our off-grid systems (using CC output current EA,  prior to the advent of the WbJr).    But,   now,  it seems that the ideal WB EA value changes,   depending upon the DOD of the battery in the previous discharge.   This can mean,   that an EA setting that worked fine on one day,  cannot be reached on the following,   or subsequent full charge day.

Also,  some Sealed battery manufacturers spec a Rate-Of-Change for termination of Absorb.   But for these batteries,   the recommended EA current change/time  would be in the EA Jitter of the WB.

Would also like to see a reduction in EA Jitter.    This Jitter can also make the difference in reaching a meaningful Absorb time,   and having the  Classic  hang on until the Absorb Timer,   Times-Out.

Some of the Jitter seen here is from Noise riding on the battery voltage,   and is dependent upon the amount of Classic output current,   and Inverter's current from the battery bank.   The inverters were installed in 2005,   and the Classic dates from 2011.   Perhaps the Capacitors on the battery terminals of these units,   have been worked-to-death   ...

The batteries here are in their 14th year  --  single string of Surrette 1280 Ah batteries @ 48 V.   In Absorb the battery impedance is probably fairly high,  and perhaps ageing batteries are also getting into higher Z territory.     one might guess ...

Anyway,   know that the Classic and WB,  and LA  are mature products,   and to MN,  and ALL of the rest of we users,   NEW PRODUCTS are the life-blood of all tech companies,   and so on.

FWIW,   Thanks,    Vic
Off Grid - Sys 1: 2ea SW+ 5548, Surrette 4KS25 1280 AH, 5.25 KW PV, Classic 150,WB, Beta Barcelona, Beta KID
Sys 2: SW+ 5548s, 4KS25s, 5.88 KW PV, 2 ea. Classic 150, WB, HB CC-needs remote Monitoring/Control, site=remote.
 MN Bkrs/Bxs/Combiners. Thanks MN for Great Products/Svc/Support&This Forum!!

SolarMusher

Quote from: Vic on April 22, 2019, 02:10:04 PM
Hi Erik,   boB,   others,

It would seem,   that having a Rate-Of-Change in WbJr current would also be quite useful.   We have used EA,  forever on our off-grid systems (using CC output current EA,  prior to the advent of the WbJr).    But,   now,  it seems that the ideal WB EA value changes,   depending upon the DOD of the battery in the previous discharge.   This can mean,   that an EA setting that worked fine on one day,  cannot be reached on the following,   or subsequent full charge day.

Also,  some Sealed battery manufacturers spec a Rate-Of-Change for termination of Absorb.   But for these batteries,   the recommended EA current change/time  would be in the EA Jitter of the WB.

Would also like to see a reduction in EA Jitter.    This Jitter can also make the difference in reaching a meaningful Absorb time,   and having the  Classic  hang on until the Absorb Timer,   Times-Out.

Some of the Jitter seen here is from Noise riding on the battery voltage,   and is dependent upon the amount of Classic output current,   and Inverter's current from the battery bank.   The inverters were installed in 2005,   and the Classic dates from 2011.   Perhaps the Capacitors on the battery terminals of these units,   have been worked-to-death   ...

The batteries here are in their 14th year  --  single string of Surrette 1280 Ah batteries @ 48 V.   In Absorb the battery impedance is probably fairly high,  and perhaps ageing batteries are also getting into higher Z territory.     one might guess ...

Anyway,   know that the Classic and WB,  and LA  are mature products,   and to MN,  and ALL of the rest of we users,   NEW PRODUCTS are the life-blood of all tech companies,   and so on.

FWIW,   Thanks,    Vic
Hey Vic,
I agree with what you say and especially about the jitter that is a bit high and could be improved to get the most accurate absorb time possible.
I hope that Midnite will never see the Classic controller as a "mature product" and will keep it under constant development. I hate this word, it makes me think about Morningstar products with their 10 years old dip switches (and still working)...
We should start a thread something like "suggestions for a better Classic", lol!
A+
Erik
Off Grid with 4kw PV | 2x Classic 200/WBjr | 2x Outback VFX3648 Epanel | 3x SPD300 + 1x Schneider HEPD80 | Hub + Mate + PSX-240 | Volthium 400Ah/51.2V LFP battery bank + Trimetric | 1500 watts AC water heater | Kubota 11kw GL diesel generator

boB

Quote from: SolarMusher on April 22, 2019, 04:48:03 PM
Quote from: Vic on April 22, 2019, 02:10:04 PM


Would also like to see a reduction in EA Jitter.    This Jitter can also make the difference in reaching a meaningful Absorb time,   and having the  Classic  hang on until the Absorb Timer,   Times-Out.

Some of the Jitter seen here is from Noise riding on the battery voltage,   and is dependent upon the amount of Classic output current,   and Inverter's current from the battery bank.  ....
....

Hey Vic,
I agree with what you say and especially about the jitter that is a bit high and could be improved to get the most accurate absorb time possible.
......

I think I know what is causing the WB Jr. jitter....   Maybe...

While regulating in Absorb, the regulation, since it is stepped, makes the current jump above the EA value and thus resetting the timer.

If there were just a minimum amount of time required before the timer was reset, might just fix it ?

boB
K7IQ 🌛  He/She/Me

SolarMusher

Quote from: boB on April 22, 2019, 11:41:55 PM
Quote from: SolarMusher on April 22, 2019, 04:48:03 PM
Quote from: Vic on April 22, 2019, 02:10:04 PM


Would also like to see a reduction in EA Jitter.    This Jitter can also make the difference in reaching a meaningful Absorb time,   and having the  Classic  hang on until the Absorb Timer,   Times-Out.

Some of the Jitter seen here is from Noise riding on the battery voltage,   and is dependent upon the amount of Classic output current,   and Inverter's current from the battery bank.  ....
....

Hey Vic,
I agree with what you say and especially about the jitter that is a bit high and could be improved to get the most accurate absorb time possible.
......

I think I know what is causing the WB Jr. jitter....   Maybe...

While regulating in Absorb, the regulation, since it is stepped, makes the current jump above the EA value and thus resetting the timer.

If there were just a minimum amount of time required before the timer was reset, might just fix it ?

boB
Hey boB,
Not sure that a shortest time would change anything, any load from inverters or battery diversion will only make it reset sooner.
I agree that nothing is perfect but for FLA, I would personally prefer an overcharge factor rather than EA value, to me it would be more predictable.
Just my opinion,
A+
Erik
Off Grid with 4kw PV | 2x Classic 200/WBjr | 2x Outback VFX3648 Epanel | 3x SPD300 + 1x Schneider HEPD80 | Hub + Mate + PSX-240 | Volthium 400Ah/51.2V LFP battery bank + Trimetric | 1500 watts AC water heater | Kubota 11kw GL diesel generator

boB


Yeah, a new rate of change algorithm is going to be more of a project than just "debouncing" the current noise to help reduce EA timer reset.

Of course, with communications and a raspberri pi or similar, some smart cookie could do almost anything imaginable.

So, won't happen this week but the ideas here will be considered in the future.

Input is really appreciated !
boB
K7IQ 🌛  He/She/Me

ClassicCrazy

#12
It would be a good project to use Raspberry Pi running Node Red.  I have been working with Node Red a bit - still getting it figured out but have it talking to the Classic via IP no problem. From there you can pull in input and output blocks or write functions and have it display or graph or email - sky seems the limit  using Node Red on what you can do with it. 

Larry
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

TimBandTech

Great thread guys.

I am bumping into the same ground.
In the past I've encouraged the look at absorption as a mini equalization.
Rolls Surrette's latest battery manual has raised the absorption voltage to a whopping 60 Volts for PSOC systems. That's Partial State Of Charge which does cover PV systems. Even for generally floating batteries they have listed a very high 58.8 Volt absorbtion. http://www.rollsbattery.com/wp-content/uploads/2018/01/Rolls_Battery_Manual.pdf ; see lucky page 13.

I have just refurbished an XW based system which had a Tristar diversion system possibly ruining the charge control algorithm; XW has no current shunt battery monitoring abilities and no ability to work cooperatively with such a diversion charge controller. After research I landed on MS Classic's Waste Not Hi PWM mode which is the technically correct solution, though few seem to understand why. The level of detail even in discussing the absorption voltage is beyond the standard home owner who wishes to own a PV offgrid system. Yet their system cost is tied into the battery bank life.

The system has 24 SURRETTE 2-KS-33PS 2V 2500AH 100HR RATE batteries charged by roughly 7kW of PV in three arrays through Midnite Classics with Waste Not High diversion driving 3kW of heating element at 48VDC through three SSR circuits. Two 6kW XW inverters power the house, including a manually rationed induction stove and 5kW clothes dryer. On a sunny afternoon in float mode why not cook a turkey or get some fresh laundry?

Using RS's new specs has possibly caused the CO detectors to go off in this ICF house. Possibly Hydrogen Sulfide, which is toxic. I have no hard proof of this but there is a correlation to sunny days. Also On the Whizbang Junior I've dropped the efficiency to 60% based on the accumulated figures. The system is consuming water much more and so this would explain the poor efficiency, which essentially appears as 'internal resistance' as I see it. It appears that RS is mistaken to publish these aggressive figures, but this system was scraping by for so long in an undercharged state I felt that this might be just what it needed. Could End Amps be enough to improve the charge system at these high absorption voltages? Without it I suspect the efficiency of the system will suffer badly. For now I've backed off of the high absorption voltage down to a not so sneezly 59.0 Volts but with an end amps at 2.5 Amps, which is difficult to verify given that I am not onsite all the time. The math computation on absorb time comes out around six hours. I've set it for five hours. 
   Stepping out away from the tech a bit: the sensibilities that are  being discussed on this thread can only be achieved by those who essentially are running battery charge algorithms in their head; seeing hydrogen bubbles offgasing, and so forth; not to mention a lack of long term practical experience for most of us. New battery technology will be welcomed, but this lead acid technology will be around for another twenty years or so. In a PV system everything is statistical so we will always have to settle for good enough. That said the proper level of technology should not require reconfiguration; not even on a seasonal basis. Equalization is enough of an issue. When somebody comes in from offsite to tweak a system and then leaves again without verification of the changes; like me on a cloudy day setting up End Amps; it can't be good. Anybody trying to kiss lead acid batteries is mistaken imo. The best systems are operated by technically apt homeowners, but the room for error is remarkable.
24 SURRETTE 2-KS-33PS 2V 2500AH @48VDC, 7kW STC PV  (three arrays) 3 Midnite Classics, one with Waste Not High diversion driving 3kW of heating element at 48VDC through three SSR circuits, Two 6kW XW inverters

dgd

Wow TimBandTech, I never realised dealing with a real power bank of batteries could be so technical in fine tuning the exact time to terminate Absorb.
I looked at this with some muted interest when the WBjr first appeared for the Classic CC.
After realising it could become a major consumer of my available think time I didn’t change anything and just left my 1025ah 24v FLA  with the slightly higher Absorb 29v (instead of Enersun recommended 28.4v), and fixed Absorb 2 hour timer. Don’t bother with EQ either.
21 years and the Enersun bank is still bubbling away :)
Looks like the original recommendation of 2 hour default ABS time from cc40 to fm60 to classic150 charge controllers was pretty good. I always wondered if the same tech person did all these controllers ;)
Dgd
Classic 250, 150,  20 140w, 6 250w PVs, 2Kw turbine, MN ac Clipper, Epanel/MNdc, Trace SW3024E (1997), Century 1050Ah 24V FLA (1999). Arduino power monitoring and web server.  Off grid since 4/2000
West Auckland, New Zealand