High Specific Gravity Unexplained

Ako · June 2018

Mu Rolls S605 6v batteries are now 18 months old and doing ok except that i keep getting very high SG readings which i cant explaine and theres no logical reason they should be above normal , i have followed everything from the Rolls manual and Rolls staff . The SG i am seeing is around 1.30 and thats at 33c so when the increased temperature is taken into account its even higher . They should be 1.265 at 25c .

Rolls are aware that i am using a Morningstar TS60 PWM charge controller and advised me the correct voltage should be 29.76 at 25c which i have maintained throughout and the only variations have been when the temperature has been higher or lower than 25c and i check the Remote temperature is making correct calculation with target voltage and use a thermometer to check RTS is accurate .

The Manual states L16 type batteries require a higher charge voltage and that charge voltage is extremely important . I have just received a reply from Rolls saying that the high SG is bad for my batteries and suggested i reduce the charge voltage by 0.4 or 0.6 or reduce the adsorption time which contradicts everything Rolls specify about charge voltages .

The TS60 PWM controller can only reduce or increase the adsorption time by adjusting the point it goes into float using a method where the controller looks at the pulse charge going ( duty cycle ) in and when it drops below 30% for an hour it goes into float which is at most once a month . I have now had to increase the 30% to 50% and reduce the hour to 30 minutes in order to have it go into float .

My concern now is that the adjustment to get the controller to go into float seems drasticly more than just tweaking and i am reluctant to reduce the charge voltage as it goes against everything i have read with the exception of one advisory in a short emailed reply to my inquiry . My voltage after charging has finished is also high at around 25.6 which Rolls told me was just surface charge bit it takes as much as 600 watts/hours over several hours to reduce it to 25.2 to me thats seems a lot more than a surface charge .

Any observations or advice would be very much appreciated as to how to reduce the SG or if i should reduce charging voltage but most importantly any ideas on how the SG could have got so high as i cant find anything through Google and knowing how it became high could assist me in avoiding it happening again ..

BB. · June 2018

Your battery bank is fairly large for that size solar array... A 605 AH battery bank should be 5% to 13% rate of charge with 10%+ ideal (especially for larger "industrial" type batteries).

1,050 Watt array * 0.77 panel+controller deratings * 1/29.0 volts charging = ~28 amps charging

Ideally, a 10% rate of charge would be ~60.5 amps.

You can get stratification in your battery bank. Typically, the heavy SG electrolyte migrates to the bottom of the cell, and the light SG migrates to the top (which would cause a "low SG reading", not high like yours...????). Ideally, once a month you want to equalize the bank and get some significant gassing going to "mix" the cells (especially larger/taller cells).

Are you EQ'ing with ~2.5% to 5% rate of charge once a month?

The other possibility, are your electrolyte levels "low" and are the plates nearly exposed? (gas out hydrogen/oxygen, the remaining electrolyte SG will rise)?

Are you filling battery with distilled/deionized water or with sulfuric acid/electrolyte (filling with pure water is what should be done).

When "refilling" the cells before charging. Only fill 1/2 full. Between gassing (trapped gasses) and rising temperature (electrolyte expanding), filling to 100% full can push out electrolyte onto the top of the battery. Only "finish filling" cells after complete charge (battery warm and no further charging/equalization).

Sorry for all of the basic questions--Just what I can think of that "could" be happening.

-Bill

softdown · June 2018

Your SG gauge may be wrong as well.

Estragon · June 2018

A couple of thoughts... if the SG is consistently high, is it possible the batteries were filled with a higher concentration of acid to begin with? Batteries are sometimes filled with different concentrations depending on the intended use (eg. traction, RE-off-grid, etc.). Another possibility is, as @softdown notes, your hydrometer may be wrong. I have one that's difficult to read, and tends to read low, and another that's easier to read, and tends to read higher. Both are prone to sticking, especially if not well rinsed between readings.

For the charge controller settings, I'd see what the batteries are doing in absorb. Normally, I'd expect some gentle bubbling, and a small rise in temperature (maybe 5°C or so). More vigorous bubbling and/or more heating would tell me the absorb voltage is too high.

Are there loads on the system that might account for the PWM duty cycle? The 30% for an hour thing may be assuming all the current is going into the battery, but if some of that's going to loads, it can be misleading.

Ako · June 2018

Thank you all for your suggestions and will answer them in no particular order .

The Hydrometer is a good quality one and checked and gives correct readings .

I am getting between 33 and 44 amps input , averaging 160 to 200 amp/hours a day , that hasnt changed much except its slightly higher i the winter as its still sunny but panels are cooler but daylight is shorter but overall input for the day remains within the same range .

Until recently i was Equalizing every 30 days however when i told Rolls this they asked WHY , told them because your instructions say to do so , they agreed the voltage and duration ( 2.5 hours at 31.2 v ) was correct but informed me that their batteries should ONLY be equalized when there was a variation between cells of 0.25 - 0.3 . The later manuals say this .

Around mid day normally goes from Bulk to Asorb and usually had 100 - 110 Apm/Hrs gone in by then .

The Batteries were filled before delivers with 1.265 .

I top up only ever with distilled water .

There are small loads occasionally during the day which is why they didnt often go into float but i understand from Morningstar this is common with the TS60 PWM and no a cause for concern .

The batteries gently bubble during charging both in bulk and asorb as i understood they should once the cells go above a certain voltage which is also when i notice bubbling so thats also correct , i think possibly thats why Rolls say equalizing is no longer nescersays as a voltage close to 30v produces enough bubbling daily .

The fluid levels are correct and due to this high SG problem i check them daily , Rolls batteries also have a reasonable large fluid capacity over the cells .

Everything has been as it has been sinse i bought the batteries nes 18 months ago and the high SG problem had been occasional only but now its become constant .

I am reluctant to reduce the voltage and go against everything Rolls have published especially as they themselves keep saying how important voltages are and following their instructions .
.

BB. · June 2018

It sounds like you are doing everything right. And almost all of the Rolls issues have been low SG readings that take hours, and sometimes days of EQ to bring up the low/weak cells.

Another thing that can cause "high SG readings" is when the small bubles cling to the float (gently tap hydrometer to knock bubbles off).

But this problem would have been apparent quite a while ago--Not just recent.

You are also rinsing the hydrometer with distilled water a few times before putting away (leaving electrolyte in hydrometer leaves a sticky film build up over time). Again, I don't think this is the issue--Just dotting i's and crossing t's.

-Bill

Ako · June 2018

Thanks Bill , all i can think of doing now is leaving it as it is for a few days and hope for the best . I noticed after it had spent several hours yesterday in FLOAT it went to absorb later today and with 150 Amp/Hrs gone in , 40% more than normal . Gave the Hydrometer a through clean just to eliminate it , I always shake t when its loaded to make sure its not sticking but where it is the reading would be lower anyway .

mcgivor · June 2018

How is the water usage? It seems strange the SG can increase, theoretically it should never rise above what the initial fill was, but I've noticed the same phenomenon, higher than when new, I choose to ignore it, figuring it's better than having lower SG, which would indicate the sulphur is adhered to the plates, rather than in solution. My water usage is low, about 1 cm. every 3 months after a year, was less in the first few months.Voltage wise for charging, bulk 30.0V one hour into absorbtion, then 29.6V for the remainder of absorption 2 hours, or 2% current of battery capacity, whichever occurs first then float 27V, all temperature compensated.About the sane daily usage 160Ah temperatures ~33°C.

Ako · June 2018

Water is higher than yours but then its summer and batteries are 30 to 34c everyday and Bulk charging 29.76 ( before temperature compensated ) for 4 or 5 hours starting at a fraction of an amp going up to around 44 amps reaching around 100 amps midday then absorption which has been 4 - 5 or even 6+ hours before returning to Bulk later in the day when input voltage drops below 29.76 . The float voltage is the same as yours 27volts .

Now i have reduced the time i Adsorption i am hoping i see a decrease in SG after a few days , if not then i have no choice but to lower the charge voltage by about half a volt , dont know what effect it will have on the longevity of the batteries .

mcgivor · June 2018

Just curious, you mentioned 100A around midday, how is that possible, or was that a 10A typo. The absorption is really not a true absorption in the sence of other controllers, the 30 percent for an hour is really a reduced rate, an average of amps 30% of the time in pulses, so 10A would be 3A averaged. My retired system which is strictly used for intermittent water pumping duties, uses a TS60 PWM set at 28.8V, the water consumption is dreadful, possibly due to aging batteries, SG is 1.300, higher than its supposed to be, I turn off the solar every other day to save water, which leads me to think that perhaps if you are shallow discharging the voltage settings are perhaps too aggressive, it's all about finding that ballance point. Perhaps I'm wrong but it is an interesting subject none the less. It seems there are no definitive answers, even from battery manufacturers, you would think that with over a 100 years of LA technology there would be some, i guess it comes down to each individual system, wish I knew all the answers, but I don't.

Ako · June 2018

I meant the total amps that have gone into the batteries around midday according to the TS60 is about 100 Amp/Hrs . My inverter shows around 3000 Watt/Hours consumption in 24 hours , half of this is daylight hours and the other half Night time so just using from the batteries leaving a voltage in the mornings before i start getting any charge of around 24.5 to 24.8
The bulk charge is everything coming in from the pannels ( 0 slowly increasing to between 33 Amps or 44 Amps, ) once into Absorption it slowly drops over the hours to around 7 Amps
I have been doing 1 minute logs daily and so its easy for me to keep a check of everything and everything checks out .

Its almost 6pm now and still in Asorption with just 10 Amps coming in and the SG is 1.30 at 35c , didnt go into float at all today and have so far had 220 Amps input which will end up around 230 - 240 Amp/Hours when it goes into Night Mode .

mcgivor · June 2018

Our consumption, my main system, is close, generally about 4KW passes through mine, albeit with a slightly lower battery capacity 450Ah, morning voltage is 24.6-24.8V before sunrise, load dependant, which leads me to think we are either both ballanced, if there is such a thing, or completely wrong. Mainly I watch what the electrolyte is doing, light gassing bulk/absorption, so far so good, voltages, although helpful are not a reliable source of information, but the higher SG thing is still a bit of a mystery

Vic · June 2018

Hi Ako,

I, too, would suspect the accuracy of your Hydrometer ... how did you check its accuracy?

What brand is your Hydro?
Have you ALWAYS completely rinsed the Hydro, three or more times, within an hour, or so of its use?
Do you ever see any air bubbles on the float or outer tube/case?

Some cheapo plastic Hydros can have the float eroded from acid, and inadequate rinsing.

Here is a Link to the Surrete article on measuring SG:
http://support.rollsbattery.com/support/solutions/articles/4347-measuring-specific-gravity

You may be doing things correctly, but, just in case.

Personally have never liked it, that the TS-60 CC had no Absorb Time setting. If it were possible, you might set your Vflt down to about 26.2 V, or so ...

Later, FWIW, Vic

Ako · June 2018

My hydrometers are both Trojan ones , the see through glass section is 21cm long with the overall length being 50cm and were quite expensive , I have tested them with loaned ones from a local Varta battery distributor .

I always give them a good rinse in distilled water after use and after drawing fluid i shake the Hydrometer to make sure the float is as high as it will go and that there are no air bubbles left in the tube .

These Rolls batteries are a replacement for the 12 Trojan 600 Amp/Hr 2.2 volt cells batteries that lasted me 10 years on the same system , my consumption has changed very little in all that time and the only change was the charge controller , before i was using another model of Morningstar controllers which i sold after over 14 years continual use which is why i decided to go for the same manufacturer , Panels , inverters are the still original ones ,. . . ,

jonr · June 2018

AFAIK, the average SG can't go above the initial fill (where would the sulfur come from?), except with stratification or evaporation or contamination - excessive charging can't cause it and less charging shouldn't be used to fix it. An incorrect initial fill seems highly unlikely.

You could buy or mix up some battery acid to calibrate your hydrometer. And measure your battery just after EQ to reduce stratification. Maybe use this post EQ SG value as your new 100% charge SG.

mcgivor · June 2018

My understanding is as @jonr states, but there has to be a logical explanation why readings of SG higher than initial fill density are observed. As stated I too have observed this, with well used batteries, which got me thinking that perhaps the electrolyte has become polluted by partials of lead oxide in solution which increases the density thereby giving false results. Came across this article
https://www.engineersedge.com/battery/specific_gravity_battery.htm
which, while not addressing the particular subject we're discussing, has some interesting points, one of which is the statement about waiting 72 hours on float, after an EQ, to get an accurate measurement. This rest period may allow suspended partials to settle, something not practical in a cycled off grid application. Given the off grid charge rate is quite aggressive in order to recharge within a limited window of opportunity, the daily gassing may never allow full settlement of suspended particles.

That's my theroy behind the mystery, any comments or other ideas ?

Ako · June 2018

Thought for a moment i had found a possible cause however i only found i became even more confusing . " As batteries near full capacity the charge current drops. End Amps or Return Amps generally refers to the lowest amount of current (Amps) running from the charger when the batteries have reached full capacity and are no longer accepting a charge. Some charges will measure the actual current output. If the charge current drops to the End Amps or Return Amps set point, this will trigger the charger to shut off. This setting is typically 2%-3% of the 20 Hr AH rating (C20) of the battery bank. " , on ,my batteries that would be 9.38 Amps and they daily go down to just under 7 volts but by then its late in the day and they go back int Bulk , on the rare occasion they do go into float which is controlled by the TS60 PWM the Duty Cycle is drops down to around 4 - 5 Amps which would indicate the Duty Cycle needs adjusting so that there is less time spent in Absorption as Rolls suggested .

Rolls were not given the absorption End Amps , i only gave them the time in absorption which brings up another anomaly , their method or calculating the Absorption time which matches my time " Where : T = 0.42 x C /I T = ABSORPTION CHARGE TIME C = 20 hr RATED CAPACITY (total battery bank) I = Charging Current (Amps) (recommended 10% to 20% of C20 discharge rate) 0.42 = ( 20%/50%) + 5% (5% is added due to losses) " my 20c rate is 468 so .42 x 468/33 = 5.96 hours which matches mine perfectly but according to Rolls thats to long .

I decided that as the End Amps were the objective and the absorption calculations were derived to achieve that so i opted for adjusting the TS60 Duty Cycle to achieve the required results . ie: End amps taking priority over Absorption time .

Have to wait now and see how that works out and the End Amps when they go into Float .

mcgivor · June 2018

Even if the charging is excessive, this would have no connection with the SG being above what it should be, case in point, when performing an equalization the guide to determine when to terminate is when SG no longer rises for a given period of time, in other words all the sulphur is in solution. Since there is only X amount, from initial fill, then that's what it is, there is no way to increase this, unless more is added, so there has to be another explanation.

Ako · June 2018

What about the time of year and panel temperature , during cooler months 20c ambient temperatures i was getting between 42 and 51 Amps coming on off my panels , now the weather is much warmer the panels are much hotter to the touch Im getting no more than 32-33 Amps in Bulk .

Rolls batteries hold a large volume of fluid above the plates , the difference between being 1cm over the plates to being 3cm over the plates could also be a factor as the dilution rate could be enough to efffect the SG enough to be noticeable .

Could either of those things be a factor ?

mcgivor · June 2018

Definitely would make a difference, the volume below the top of the plates is largely occupied by the plates themselves, unlike above where there is only electrolyte.

Ako · June 2018

The problem with these Rolls batteries is that its very difficult even with a torch to see the fluid level , it has to be between 6mm and 13mm , can see the fluid level but not the relative distance between it and the point the measurement is taken from . Have been playing about with the clear part of a ball point pen with a wooden match inside it , there is 36mm distance between the ideal fluid level and the top of the case so the idea is if lowered into the fluid to a mark on the plastic level with the top of the battery casing then a 5cm match would float up inside the plastic tube to an established mark above the battery casing . I have only tried it in a glass of water , IT DIDNT WORK , the surface tension inside the plastic tube offered to much resistance to allow the match to float freely so im now looking for some hollow small plastic balls to pile on top of each other inside as a replacement for the match . Im expecting the same problem so looking for a larger diameter tube with larger and therefor more buoyant hollow plastic balls . It might be i have to devise a float that will sit on the water with something to slide up inside the tube so i can overcome the sticking problem . Originally i had the plastic tube fitting tightly down the filler tube of a funnel and drilled holes around the bottom of the cone , in theory the funnel sat on the battery so gave a fixed known point and i could pour water in so it would enter the battery through the small holes around the plastic tube within the battery opening and watch the match float up inside the plastic tube so i knew when to stop pouring , all that worked a treat , it was only the float that didnt . Do i sound mad , will know if i ever solve the problem .

Now someone will tell me to get a self watering system or that theres a product i can buy that does the job . I bought a container for the purpose of filling that cuts off when the fluid reaches the nozzle , unfortunately was a waste of money , worked on batteries where the flui level was 22mm below the top and no way to modify or adjust it so a lower point .

mcgivor · June 2018

For what it's worth, being the batteries are warm, which is what I have 30°C plus for the most part, I fill until the electrolyte touches the neck, that way all cells are at least close to equal, always when float is reached to ensure no further temperature rise. Never had any problems using this method, could be a problem in cool climates with big swings in temperature. Yes it is difficult to see, let alone gauge the level but quite easy to see when the neck breaks surface tension.

Ako · June 2018

Thanks for that McGivor , mine are above 30c now even at night , daytime 35c so i have been tempted to fill to the lower part of the protrusions also on the assumption that the only reason for Rolls giving a lower level was so that the fluid wouldn't be lost when it expanded . If thats the only reason then i dont see any harm in filing to that level if i cant come up with a simple way to determining it while filling . While i was typing this i had a Eureka moment so gonna try something else .

mcgivor · June 2018

That was my deduction, rarely do I ever see battery temperatures below 25°C and even with the fill to the neck method, there has been no expansion of any concern, keep us posted on results, we are all learning from others experiences.Good luck.

Ako · June 2018

I will but i think Rolls have given up after telling me that high SG is BAD for the batteries their only suggestion was to lessen the Absorption time or reduce the voltage , they couldent even give me a possible reason for the problem . My email reply was that reducing voltage goes against all the instructions in the manual after saying how important voltages were for each phase of charging . Dont think they have an answer for that .

Looking at the fact sheet it stated theres 57mm reserve fluid above the plates , that is a lot however it shows 12kg difference between dry and wet weight , thats 4 kg of fluid a cell so 4 ltrs per cell .

Just learnt something interesting to me although everyone else probably already knew . I understood the Duty Cycle had to remain at or below the threshold for the set time and if it went above the clock started again from Zero , im sure i even read that somewhere however, it doesn't , i have just looked at my log as my batteries just went into float after just a few minutes below the threshold so i looked back and counted the minutes and its a cumulative total since the last 24 hour cycle , on 3 occasions today mine have dropped below the threshold and for a short time only , its the cumulative total of them all that triggers Float , checked against other days they have Floated and its the same .

Tomorrow i plan to remove all the batteries , clean all the terminals as some have a build up and reverse the order they are conected to the system .

mcgivor · June 2018

Telling you that high SG is bad defies logic, if it's not possible to increase above the fill SG, what dose the amount of absorption have to do with the equation? My suggestion would be to go above the heads of who you're dealing with, perhaps to hesd office in Canada. My situation is without any tech support, but you are dealing with a reputable company, ask a direct question, can the SG rise above the fill specification due to over charging, my guess is the answer will be no.

jonr · June 2018

Even more importantly, the cure for too high SG isn't less charging - that will just leave lead sulfate on the plates that will eventually become the stable crystalline form (ie, the way to ruin a battery). Charge until SG won't rise any more. Roll's writes "insure no further rise in specific gravity".

I see that Roll's writes "For standard batteries, electrolyte levels should be 1/8” below the bottom of the vent well about 1/2” to 3/4” above the top of the plates. "

I would imagine that filling with anything other than distilled water could cause SG problems.

Ako · June 2018

Jonr , thank you for your input , its appreciated however the batteries are the L16 type and the correct level given is 6mm to 13mm ( quarter to half an inch ) below the vent protrusions inside the cell and 57mm ( over 2 inches ) above the top of the plates , there is an illustration to make the explanation easier to understand in the manual . The part you read referring to " insure no further rise in specific gravity " in with reference to equalizing charge only . I would never fill with anything other than distilled water as you point out topping up with anything else would cause problems with the SG .

Estragon · June 2018

For measuring acid above the plates, you might want to try just using the empty plastic pen tube itself. Mark it with relevant levels, put it in a cell until it touches plate top, put your thumb over the open top end, then raise it while keeping the top plugged. The acid should stay in the tube to read the level (same idea as a lab pipette). Just take your thumb off the end to let the acid drain back into the cell after reading.

Ako · June 2018

Thats a thought but i was hoping to find a one action method . The tube and thumb suggestion would be quicker and as accurate as continually punting a funnel in then pouring the water and lifting the funnel out to check and possibly repeat the process two or more time for each cell .

Think i have now found a method that is both accurate and only involves one action removing the need to check .

Vic · June 2018

Just a couple of things ...

Generally, even for L-16s, there should be no real issue in filling each cell just to the bottom of the Vent Well (fill tube), particularly, if the batteries are at, or near their ultimate operating temperature. The difference in SG readings between electrolyte being just barely above the plates, and having been filled with Distilled Water to the bottom of the vent well should only be a few SG points.

There can be one additional reason for having accurate SG readings somewhat above that of the fill electrolyte. In addition to a slight SG rise caused by low electrolyte level, is routine charging an FLA battery, that has only been lightly cycled -- ie, not below about 90% SOC. But based upon what Ako has staged above, it did seem, that these batteries had been cycled below 90% SOC on most charge cycles.

I have used TS-60 CCs, and personally, I do not respect the method that they use to end Absorb. Duty-Cycle would be an indication of the total demand of the system, so, an indication on the CC output current, but, it aint NO Shunt!

Using the Morningstar MS View, or whatever it is called might allow one to somewhat customize the charge voltage settings, but, still, sure would recommend a friendlier CC for this system.

You might try measuring the SGs in the early AM, to see what is the correlation 'twix those readings and those taken later in the day.

Have tried cleaning a very good (Freas) glass Hydro, that some had NEVER rinsed. It was just NOT possible to remove the sticky acid residue, using mildly-abrasive cleaners, and repeated rinsing ... it always read high.

Yes, do believe that eroded plate particles floating in the electrolyte, can cause erroneous SG readings. Waiting an hour or two, after entering Float, for these particles to settle out, can help accuracy, if the CC is still in Float at the time of measurement.

Would suggest getting two new, all-glass Hydros to use, as a sanity check.

This is a long DIscussion, this might have been touched on, but one does wonder about a bad batch of Distilled Water, having, possibly contaminated the batteries.

And, so on, FWIW, Vic

High Specific Gravity Unexplained

Comments

Categories