HELP-Settings, Charging Issues and other Stuff

[Garret]

I'm having charging problems.  Batteries are new 4-6V, Rolls S-550 (428 Ah) from reputable company.  Batteries voltages good when received (6.2 +/- 0.1 V/cell).  Small off-grid system up and running for about 3 weeks.  Not much sun, so using Magnum 4024 PAE built-in charger in addition to Classic 150/Solar input to charge and top off batteries.  Using SOC on Magnum charger for AC-connect and charge; range 65%-100% SOC.  So batteries never discharged more than 45% and never allowed to remain discharged more that 24 hrs.  Loads are light and carefully monitored using Midnite Local App.

Custom battery settings on Inv/charger and Classic 150 (according to Rolls @ 0-16C):
Bulk/Absorb=30V
Float=26.3V
Eq=31.5V (middle of range)

Decided to check specific gravity on all cells and keep a log to monitor battery health.  At 100% SOC (according to ME-ARC remote monitor with BTS as well as WB-Jr on classic), SG=1.22 +/- 0.01 variation between cells.  Should be 1.255-1.275 at 100% SOC.  Never added water, levels good.  Decided to perform EQ charge (temp compensated) after reaching float charge stage (according to Rolls specs, 2-3 hrs duration).  After required rest period, retested SG=1.22 +/- 0.01?  Performed another EQ.  Still 1.22 +/- 0.01.  Using (temp compensated) refractometer calibrated with distilled water. 

As a research chemist, I wanted to verify accuracy of SG readings.  So, removed an aliquot from pilot test cell and measured SG the old fashioned way.  Weighed a carefully measured volume (from an analytical grade burette) of the electrolyte soln. into a tared container on an analytical balance.  Used a volume large enough to reduce error (~25 ml). The calculated SG at 20C was 1.2075.  Good agreement with refractometer.  This would indicate barely a 75% SOC.

I am thinking the batteries are not getting enough absorb charge time and/or voltage.  During absorb, the batteries are only accepting between 3 and 12 amps.  I set the absorb time on Classic to 4:18 (T=0.42*C20/I).

I am also unclear on what settings I should use for temp compensation, charge voltages and voltage limits on the Classic?  The default ranges appeared too low/narrow and appear to be set to a range to protect AGM batteries(?)

Don't want to trash my batteries this early in the game.  At this point, I'm looking for any good advice on how I might correct my SG problem and some good default numerical parameters for the Classic that would perform well for my simple system.

Please help.

 

[Westbranch]

I am thinking the batteries are not getting enough absorb charge time and/or voltage.

you've got it!  Rolls have a  reputation for a long break in period and needing longer Absorb times to start, from readings, up to a year...

do a search here and over on the NAWS board, lots of posts...http://forum.solar-electric.com/search.php?do=getnew&contenttype=vBForum_Post

[Bob D]

I had a similar situation with a new battery this year ( not Rolls, but a Crown industrial battery;  the issues are the same).
I had really good dialogue with a tech person at the battery supplier, and eventually worked out the following:
-yes, the batteries are never getting charged.  The absorb time is too short.  I use the end amps setting on the Classic, when solar is available.    My battery is 790AH; an end amp reading of 3 amps was recommended; I am using 4 A. In my case it takes 6-7 hours of absorb to reach the end amps - I determined that by collecting data from the Classic, and also by setting the magnum time high enough that it stayed in absorb ( off the generator) until I reached the end amps.
-once in a while, (and at least once early in the life) an extended equalize is recommended; run the equalize with a generator until the SG's stop rising.  It will likely take a few hours. In my case, the target SG is 1.285 and it took longer than I expected to get there. I think the Rolls manual suggests something similar. You need to check the SG's about once an hour, and run until they stop for about an hour.  Now you know the battery SOC is 100%, and the SOC on the Magnum (I have the same inverter) will now have a proper starting point.  I assume you have the battery monitoring kit.
-You will find you need to run the absorb for longer than you might think - the end amps feature of the WB Jr is really a good approach.
-the battery tech guy was very clear that chronic undercharging is going to have a much more deleterious effect that almost anything else; he really encouraged me to sort of hammer on the battery fairly hard to make sure it was charged fully.

Hope that is of some help.

[Vic]

Hi Garret,

Welcome to the Forum.

Your stated Absorb voltage should be high enough to charge these batteries.  Yes you do need a longer Absorb time.

Surrette also recommends an initial charge,  then a Commissioning EQ.  You are doing this,  but,  this EQ may take hours and hours.

It would not hurt to get a glass Hydrometer from an Auto Parts store as a convenient check on your Refractometer.

YES,  IMO,  in the Classic Limits menu,  the Temperature Compensation range seems much too narrow,  however,  in my case - set the EQ voltage to Vabs,  and think that the upper comp limit is set at Veq.  You will probably want to set this upper Limit to several volts above your Veq,  unless your inverter will not tolerate this high a setting.

The customary setting for Surrette Floodeds (and most other FLAs)  for battery temp compensation is -- 5 mV/C.  (this is the per-cell setting - the Classic knows your battery V,  and will scale it to that correct number of cells).

Batteries are not completely 'Finished' when shipped from the manufacturer.  They need from about 20 - 50  Cycles (or even more)  for them to settle in,  and have the plates fully-Formed.  Until then,  you just need to try to get them fully charged at least several times per week if possible IMO.

And,  6.2 Volts as the incoming initial reading is not so well charged,  although,  as you know  the Resting voltage is affected by battery temperature at about that -- 5 mV/cell.

FWIW,   Opinions ...   More later,   Good Luck,    Vic

[Garret]

Thanks for the quick responses Bob and Vic.  Don't get me wrong, so far I am pleased with performance and the batteries seem to handle the loads.  Just seek to understand what I have observed which so far has been contrary to expectations.  I don't want to go down the wrong road as so many others have in regard to battery maintenance and end up with 4-123 lb plastic covered lead blocks.

The customary setting for Surrette Floodeds (and most other FLAs)  for battery temp compensation is -- 5 mV/C.  (this is the per-cell setting - the Classic knows your battery V,  and will scale it to that correct number of cells).

This was the default, but can't seem to find the reference voltage anywhere on Rolls site.  The default on my Classic was set at 10C.  The amount of compensation appears to be affected by this setting.

Batteries are not completely 'Finished' when shipped from the manufacturer.  They need from about 20 - 50  Cycles (or even more)  for them to settle in,  and have the plates fully-Formed.

I understand the concept of electrolytic formation where the PbO2 is "formed" in situ and becomes the positive plates (cathode).  That doesn't explain the unusually low H2SO4 concentration (SG) after extended charging and EQ.  It would appear to me that someone either commissioned these batteries with dilute electrolyte, or watered them excessively.

I am using the WBJr ending amps setting to end the absorb charge cycle.  I set this at 1 amp (no particular reason, other than to extend the absorb time).  During absorb the batteries will not accept very much current which leads me to believe something is really wrong.  During bulk cycle, they will accept as much as I can give them (in a test using the magnum charger set at 100%, they would accept over 85 amperes, initially).  For bulk, I set the charge current limit to 40% (0.4*105(max output=42 amperes) which is about 10% of C20 rated capacity of 428 ampere-hrs).  The only way to increase the charging current to anything greater than 5 amps in absorb is to set the absorb voltage to >32 V, which is essentially an EQ.

I will try an extended equalization with periodic SG monitoring.  If SG doesn't increase I'm thinking the only solution is to empty the batteries and refill with the proper electrolyte which I will prepare myself.

Thanks You again

[boB]

Lots of good info here.

"The default on my Classic was set at 10C. "

Good thing you found that.   That was an oversight on the VMM (factory default reset) from the MNGP/Remote.
Hopefully you set it back to 25.0 C.

If the battery temperature is low and the temperature compensation wants to make it higher than the LIMIT temp-
comp maximum voltage, yes, definitely raise that up higher.  Those limits are actually for VRLA batteries to make sure
they don't go too high or too low.   Those default limits should really be wider.  But you can just widen those out.

It is true that these batteries have a bit of a break in time to form the plates.  If you do call Rolls/Surrette,
ask for Steve Higgins.

boB

[Vic]

Hi Garret,

First,  (just saw boB's post)  the correct reference temperature for Surrette FLAs IS 25 C.

IMO,  the plates continue to Form with cycles to 60,  or so percent SOC -- just use them as you typically would.  This could require several months.

However,   it would be good for these batteries to get a FULL CHARGE to rated SG of 1.265 nominal.  This might require a charge and a L O N G  EQ  at about 63.3 or so volts,  temperature compensated.  Might take 6 or 10,  or even 12 hours.  This could be done over a period of two or three days (or so).

In your first post you mentioned that the Surrette batteries were from a "reputable company".  Were you referring to Surrette,  or a battery dealer?

IMO,  you should NOT do anything with the electrolyte in your batteries -  definitely do NOT DUMP it.  There is no reason to suspect that there is anything wrong with it.  Your initial voltage reading seems a bit low to me,  especially if the battery temperatures were cool.

IF the batteries were from a dealer,  they may have been sitting of some time,  off charge.  You should do your Commissioning EQ,  UNTIL THE SGs rise to the nominal of 1.265.  The batteries may get a bit warm after hours and hours of EQ.   Your Refractometer may have temperature compensation.  If the batteries become fairly warm,  the SG readings that are not compensated will appear a bit low.

You might want to assure yourself that the Absorb and EQ voltages at the battery terminals are very close to the reading on the Classic for battery voltage.  At higher currents, there will be some voltage drop in wiring,  and the Classic has an offset that can help compensate for the cable resistance.

One question for you;
When you EQ the batteries,  just after the Classic has gone to Float,  what is the initial current when your EQ target voltage has been reached?  This is when the Classic begins to say EQ,  as opposed to EQ MPPT?  And during the EQ,  do you smell an aroma of Sulphur?

You are a very technical person,  and you are paying careful attention to your new battery bank.  This is uncommon,  and very good.   But FLAs are very forgiving,  and they are patient with us while we get into rhythm with them.

Thanks,  and FWIW,  I have been very happy with the Surrette batteries here -- 9.5 years and counting ...  knock,  knock.

AND,  as boB mentioned,  Surrette gives very good Tech Support via e-mail and telephone.  Steve Higgins,  Surrette Tech Manager is very knowledgable --  he spent a number of years at Outback Power in Tech Support.

Vic

[Garret]

All great comments/suggestions folks.  Much appreciated.  For more details regarding my system please refer to my intro post:http://midniteforum.com/index.php?topic=2301.msg21716#msg21716.

My reference voltage was set to 10C in all the aforementioned operations/troubleshooting.  Will set to 25C immediately.  This would certainly increase the voltages especially since the batteries (according to sensor in proximity) are now hovering around -9C.

Vic, were you implying possibility of sulfation from a prolonged storage at dealer?  I purchased my batteries as a package including the cabling from Wholesalesolar.  Couldn't find a date of manufacture unless it is encoded in serial.

I did not anticipate having enough daylight remaining after reaching float to do an EQ from the C150, so this was accomplished from the Magnum charger (utility power).  So can't report on anything related to the charge controller with regard to currents and/or voltages.  With an EQ voltage of 31.5 V, the current settled in at only about 5 amps, however the batteries were evolving gas quite vigorously.  There was a slight smell of hydrogen sulfide in addition to what I would expect from sulfuric acid aerosol.

When your mentioned 63+ volts for EQ, assuming you were referring to a 48 V system.

The battery cabling is 10 ft of doubled 4-0, so voltage drop would be fairly minimal I would suspect.

Had to attach a screenshot of my 1650 W panel array generating 1663 W.  Actually the recorded peak was 1687 W.  How is that possible in Pennsylvania in Jan?  Miracles do happen.

Cheers,

Garret

[vtmaps]

I did not anticipate having enough daylight remaining after reaching float to do an EQ from the C150, so this was accomplished from the Magnum charger (utility power).

For those of us without utility power we get up early to bulk up the batteries with the generator, and then let the solar charge controller do absorb and equalization.   Bulk charging uses the generator at near its rated output, when it is most fuel efficient.  It's usually an inefficient use of a generator to run at low output during absorb and eq. 

I hadn't thought about it before, but your timing makes sense if you're on the grid... solar for bulk and grid for absorb & eq after dark.

Had to attach a screenshot of my 1650 W panel array generating 1663 W.  Actually the recorded peak was 1687 W.  How is that possible in Pennsylvania in Jan?  Miracles do happen.

Quite normal in cold weather.  Your panels put out higher voltage at cold temperatures.  Cold temperatures are often associated with clear, dry atmosphere which allows full insolation. 

--vtMaps

[Bob D]

Garret, I concur with Vic about the electrolyte - it is unlikely it is a problem.
fwiw, I had all the same concerns you are expressing when I installed my new battery, but, as I mentioned in my other post, once I talked it through with Crown, I ran a commissioning equalization, which took many hours spread over two days, all is OK.
As a point of reference, it took about 5 hours to move from 1.24 to 1.272, then another 4 hours to move from 1.272 to 1.290.  In my case, the target is 1.285; there is a small difference in SG cell to cell and so I ran until the lowest cell reached 1.283, at which point the cell I used for reference was at 1.290.
At the 4 hour ( 10 total) mark, the SG's stopped rising and I stopped the EQ after another 2 hours.
During EQ, the current dropped into the 5-10 amp range with my somewhat larger battery, so the 5 amps you mention sounds about right.

[Garret]

Bob D:

As a point of reference, it took about 5 hours to move from 1.24 to 1.272, then another 4 hours to move from 1.272 to 1.290.

My whole point here is that I have done 2-3 hour EQs (after reaching float) at 31.5 V (plus temp. comp.) which is in the middle of the range specified by Rolls (30.96-32.04V) and the SG went from 1.22 to 1.22 (no change).  That's what seems so suspicious to me.

On the bright side of things, at least the inter-cell SG range does not vary more than 0.01.  I will try 1 more extended EQ, as I was directed (and watch SG and water levels).  If that doesn't work, I'll go with Vic and boB's other suggestion to run normally expecting the batteries to settle in after a period.

You Folks are very helpful and responsive.

[Vic]

Hi Garret,

Thanks for the pic,   and added info.

Sorry about the wrong EQ voltage,  forgot to scale it.  Almost all systems that we deal with around here are 48 V,   unfortunately,   think in 48 V,  not V per cell.

Good that you found the 10C reference temp,  and changed it.  Hope that you can confirm the charge/EQ voltages at the battery terminals,  verses those indicated by the Classic and the Inverter/Charger.

Also,  almost all of my Surrette experience is with batteries made almost 10 years ago.  They may have changed the composition of the plates slightly,  or the Forming process,  that may change the way the newer batteries behave on Commissioning.

Also,  would not hurt to contact Surrette Tech support.   They do a very good job supporting their customers.   Also,  you might contact your battery dealer with your wonderments about these new batteries.

And so on.    Vic

[Vic]

Hi Garret,

Surrette has always used 25 C as the reference point for Capacity,  although,  now do not find it in the Data Sheet for any of the batteries that just looked up.  They do make a reference to 80 F as the reference temp for SG measurements.

There appears to be a new version of the Rolls Battery Manual,  this is probably the one that you are using.

The Absorb time formula seems useful only for Constant Current battery charging -- often used for Industrial batteries.  But since Renewable Energy (RE) systems do not use this approach,  this formula yields times that are usually not so appropriate,  IMO.

In your initial post,  you mentione,   "Loads are light and carefully monitored using Midnite Local App".   So do not understand quite how your battery bank reaches 65%  SOC that you mentioned.   Believe that you need to purposely Cycle your batteries to 50 or 60% SOC a number of times.  Rather than using an SOC display on the Magnum or even the Classic,  why not use the AH that are removed from the battery on the WBjr Status display -  this is probably on the LA,  as well.

Technically,  the actual AH Capacity of your S-550 is 5% less than the 428 AH that you note,  as 428 is the 20-hour C if the batteries used 1.280 SG electrolyte,  which they do not -- 1.265 is the correct target for almost all Surrette batteries,  as you know.

Just a drive-by post.   Good Luck,   Vic

[Garret]

Vic,

Never noticed the fine print at the bottom of the A-hr capacity table.  Should have been wearing my cheater glasses.

I think that is pretty sleazy of Rolls and explains why competitors L16s of roughly equal dimensions and weight are 400 A-hr.

The Absorb time formula seems useful only for Constant Current battery charging -- often used for Industrial batteries.  But since Renewable Energy (RE) systems do not use this approach,  this formula yields times that are usually not so appropriate,  IMO.

What amount of absorb time would be appropriate?

n your initial post,  you mentione,   "Loads are light and carefully monitored using Midnite Local App".   So do not understand quite how your battery bank reaches 65%  SOC that you mentioned.   Believe that you need to purposely Cycle your batteries to 50 or 60% SOC a number of times.  Rather than using an SOC display on the Magnum or even the Classic,  why not use the AH that are removed from the battery on the WBjr Status display -  this is probably on the LA,  as well.

I was using SOC on Magnum which was consistently a few point higher than Classic/WBJr.  So that I wouldn't risk an inadvertently excessively discharging, I used the AC auto connect/charge feature on the Magnum that is activated by reaching the lower SOC setting.  I guess I could simply offset this SOC setting by the difference from the WBJr, since I believe the implication here is that the WBJr is the more accurate of the two.

You might want to assure yourself that the Absorb and EQ voltages at the battery terminals are very close to the reading on the Classic for battery voltage.  At higher currents, there will be some voltage drop in wiring,  and the Classic has an offset that can help compensate for the cable resistance.

I had an opportunity to monitor the voltage readings at the battery terminals during absorb using the Classic.  The Classic was reporting 29.8 V and I measured 29.69 V at the terminals (Fluke 87 III true RMS).  Conversely, when I started an EQ from the Magnum it reported 31.6 V (there is no 31.5 V setting, my target) and at the terminals, 31.55 V.  The BMK meter voltage on the Magnum remote has 2-place precision and showed 31.59 V. -  considerable more accurate than the Classic (not intentionally bashing the Classic).  Interestingly, at the same time, the WBJr was giving me the same 31.6 V.  Voltage drop doesn't appear to be an issue with double 4-0 x 12 ft cabling. 

UPDATE:
Last evening I performed 2-consecutive EQs of 4 hours duration using the magnum set at 31.6 V.  There is no option on the magnum remote to temperature compensate the EQ voltage as there is on the Classic and the maximum setting being 32 V.  After the first EQ, SG had increased from 1.22 to 1.24 in the pilot cell.  After the second EQ, SG=1.26 (almost there!).  I think changing the reference temperature (which bumped absorb voltage on the Classic) had a very positive effect.  It would seem to me that although the SOC in WBJr was telling me 100% on reaching float, I wasn't getting a complete absorb stage due to inadequate voltage.  The temperature compensated absorb voltage (with my cold batteries) on the Classic display was 31.8 V, essentially an EQ.

At this point I am convinced this whole thing was nothing more than another case of pilot error (I should know as a pilot).  I am also confident in your earlier assessment that the batteries require a conditioning EQ and cycling to realize their full capacity.

Appreciate all the help.

Until my next complaint,

Garret

[Garret]

Vic,

I attached an interesting paper on stratification in FLA batteries you may find interesting.

If you have already seen this, disregard.

Garret