Ok, I know this seems like a beat to death subject, but I've been earnestly studying the phenomena and have some more input in regard to the topic...
This 'desulfation' concept as it's been getting viewed I think may be getting a bad rap from improper application. Here's my hypothesis as to why:
First, the surrounding concepts (for those not up to speed)... If you've been around lead-acid for numerous years, and aren't up for the refresher course (or just a good read), then skip to Just above Phase "5".
It's known by the 'hard timers' that the lead-acid chemistry really has 4 stages of charging, whether we choose to accommodate for them or not. They are:
1 - Recovery. This is a very current limited charge rate that when applied properly, prevents (further) damage to a deeply ( < ~60% SoC) discharged cell. Partly due to 'soft' sulfation, there is less active surface area on the plates, which can lead to the areas that are
more conductive 'charging' before the rest of the plate.
This can have 1 of 2 effects, and usually a combination. A, loss of active plate material due to premature bubbling knocking some of the soft sulfate off of the plate, and two, further weakening of an already fragile structure susceptible to warping due to higher current with less mechanical support, again resulting in lost plate material. Worst case scenarios can even lead to the active material becoming lodged between plates and shorting the cell.
The recovery phase prevents as much damage as possible by severely limiting the physical activity that can occur within the cell. This can take 12 - 24 hours at preferably a MAXIMUM of C/20 rate, preferably much lower.
The end of this phase is triggered by the cell(s) reaching a predetermined voltage. The problem with this phase is, it is probably one of the most critical, and the specifics appear to be dependent on battery age, as well as many other factors (DoD, Temperature, Number of times it has gone too far, etc). A lot for an automatic charger to keep track of, nevermind do.
2.05V/cell is a good general rule of thumb benchmark to end the recovery phase.
Side note: Flooded cells and SLA's seem to have differing needs here. Still looking into that.
2 - Bulk. This is the phase where the, you guessed it, bulk of the charge is replenished into the cell. Higher current is forced into the cell to cram as much energy as possible in as short of a period of time as possible. This phase for the most part could be considered to happen on average between 60 and 80% SoC.
Below 60%, the chances of cell damage due to weakened cell structures increases, and above 80%, charge efficiency goes down. Continuing to push high current after about 80% can result in damage to the positive plate structure by oxidizing the grid, a difficult (if not impossible) situation to reverse, much like 'hard' sulfation. Voltages above ~2.667V/cell won't gain much in terms of charge absorption, and just liberates heat, explosive gas, causes plate warping, and even internal cell shorting (the combination of which can lead to explosion).
This is the mode that most RE batteries see for the duration of their lifetime. RE is a 'take what you can get when you can get it' kinda thing, and doesn't naturally conform to the chemical and physical needs of a battery very well at all.
For most, by design, this means the system is (hopefully) operating within about 20% or so of total battery capacity, no matter what, with a WYSIWYG approach to charging. If the sun's out, charge. If the wind blows, charge. If your wife wants to shop, charge. Oh wait... sorry...
The end of this phase is triggered by the cell(s) reaching a predetermined voltage. Typically between 2.4 and 2.5V/cell.
More on this later...
3 - Absorption. This is where the current is once again reduced, and an upper voltage limit is introduced. Again, usually between 2.4 and 2.5V/cell. The difference is that during absorption, the cell is allowed to remain at this voltage for a period, rather than triggering the next phase. The so-called 'Smart' chargers have, to varying degrees, managed to successfully merge Bulk and Absorption together. I personally prefer them to be two separate events, but this is not always feasible, particularly with RE.
During this phase, the charge rate is reduced (to somewhere between C/5 and C/10 usually) to allow the remaining parts of the plates to 'catch up' to the parts that were charged during the Bulk phase. With reduced current comes reduced heat production and gassing, factors in service life and maintenance schedules of the cell, respectively.
This period can take as long as the absorption phase, and so is usually just overlooked (as mentioned in Bulk) in the interest of getting as much energy crammed into the cell as possible, as soon as possible.
'Skipping' this phase ultimately means that a cell never really reaches a full charge, and over time (months to years depending on the user and/or charger's 'tolerance' factor), this eventually leads to some irreversible ('hard') sulfation. For the RE crowd (particularly with solar), this is a concession: A necessary evil in tradeoff, give up some service life for 'the now'; usable capacity.
The end of this phase is ideally triggered by the cell crossing a minimum current threshold, indicating that the cell is nearing full (> 95% SoC), and if the cell's capacity is known to an automatic charger, the charger may also be set up to end this phase after a predetermined amount of time at a given rate.
4 - Finishing, or 'Float'. This phase is characterized by a much reduced charge rate, typically < C/50, at a reduced voltage as well. Typically, depending on exact chemistry (ie FLA vs SLA), the finishing voltage will be between 2.3 and 2.35V/cell. This low charge rate allows the cell to absorb the very top band of it's capacity (above 95% SoC), and is rarely seen in RE use.
The goal of the finishing phase serves 2 purposes; final absorption (as mentioned) and prevention of self discharge, all while minimizing gassing (which of course increases maintenance intervals).
There is no official 'end' to this phase, although the cell can be taken off charge and stored for some period of time (generally less than 6 months) following this phase, but will need to be topped again at regular intervals, usually beginning with the Absorption phase.
Now then, there really are 2 other phases that should be in play here, at least on an 'official' basis, and I believe that this is the key. One of them (equalization) is now in fairly widespread use for stationary lead-acid systems, the other I believe to have been misapplied to an extent, resulting in it's bad rap. They are:
5 - Equalization. This concept is a bit newer than the other four, and was never really brought into the mix as an officially named mainstream practice until much more recently than the age of the chemistry itself. It does not apply to single cell systems (not really found in 'nature').
The purpose of equalization is to bring all of the cells in a
battery to equal levels by providing a controlled overcharge for a predetermined period of time at a set rate, based on the battery's capacity. During equalization, all of the cells experience vigorous bubbling, which helps stir the electrolyte to avoid striation (condition where the electrolyte separates into a stronger layer of mostly acid, and a weaker layer of mostly water), which can cause problems with cell balance during the life of the battery.
All cells ideally reach 100% SoC, restoring some lost capacity due to minor cell differences that are unavoidable. Cell imbalances have a tendency to snowball, with a weak cell reaching full and overcharging before stronger cells, as well as becoming more deeply discharged than stronger cells each cycle. Each direction causes more and more damage to the already weak cell, eventually resulting in cell failure. This can mean the failure of an entire battery, when the individual cells are not readily available for replacement.
Equalization attempts to prevent this by giving all of the cells the best chance of keeping their full capacity.
The end of an equalization can be triggered by a few things, usually time, temperature, and when done manually, over several cycles based on CTV (charging terminal voltage). The voltage should be kept below 2.667V/cell at
all times. Equalizing chargers are usually set up to deal with this process automatically, and usually only require the user to monitor the temperature of the battery, and discontinue the equalization based on experience and discretion.
NEVER attempt to equalize a sealed battery (SLA/VRLA). You will do more harm than good.
6 - Desulfation. Ahh, yes, the meat and potatoes. Still with me? Good, this is what you came for.
This has to be one of the most controversial topics surrounding battery maintenance and usage in it's entire history. Some say it works wonders, some say it doesn't work at all, some say it makes the battery even worse than when they started.
I agree.
You're probably wondering why I just went through typing all of that just to end up here. Good question, and here's the answer:
I believe that the process of desulfation actually
is valid, however there are a lot of caveats to it, and many factors that play into it's success.
1 - The age of the battery. As I mentioned in phase 1 (recovery), lost plate material is just that. Lost. You can't get it out of the 'dust pan' and back onto the plates no matter what you do. This material represents permanent lost capacity. This phenomena is indistinguishable from sulfation in terms of external electrical observation, and is often confused for it. Performing desulfation on a battery in this condition is as futile as trying to move a mountain with a tonka truck.
2 - Hard sulfation. There are generally two 'accepted' types of sulfation, hard and soft. Soft sulfation is (varyingly) reversible, hard is next to impossible to undo. Sulfation starts out as a natural part of discharging a cell, and is reversed (for the most part) when the cell is recharged. Hard sulfation results when the sulfate crystals begin to grow on the 'seed' provided by the soft sulfation. Hard sulfate crystals are much less conductive, and will not pass a current anywhere
near as well as soft sulfate does, and therefore is more or less a permanent condition. Generally, hard sulfation sets in when a cell is 'left for dead', providing all the conditions necessary for the growth of these unwanted crystals.
3 - Frequency dependent. This is somewhat pure speculation, so I'm sure I'll get spanked pretty bad for this one by the 'it works' crowd, but there's deep logic for why I do not believe that high frequency pulsing has any real effect on a cell. In fact it may actually be part of the 'makes the problem worse' observation when combined with 'Timing' (mentioned later).
The common belief is that high current, high
frequency (>25KHz) pulsing is the way to go about removing sulfation. I only partially agree with this. In fact, having tried it 8 different ways to sunday, on several different batteries in varying SoC, age, currents, and frequencies, there is only
one combination that seems to work at all. More on this later.
The problem with such high frequency is that the pulses are so short, they are attenuated readily by even the smallest chunk of plate, as well as no time to allow the cell to 'recover' from the pulse. If the energy in the pulse is not 'seen' by the cell, it will have no effect.
I have found that frequencies less than 20Hz work best.
4 - Duty Cycle dependent. Taking heed in the frequency dependency, the duty cycle at lower frequencies also plays a role. Depending on a few factors, the pulse width changes the outcome somewhat, although honestly, it is not entirely clear exactly where the 'butter zone' is. My rough guess is between 2 and 5%, but may be capacity dependent as well. Still looking into this.
5 - Rate dependent. The pulse current makes a big difference in the results of the tests, and is roughly proportional to the rated capacity of the cell. In truth, this is only partly correct; the actual capacity plays more of a role than the rated capacity does. The rated capacity gives an approximation of where one can begin, but should only be used as a rough guide. Around 1C seems to work rather well. Below this level, the results fall off, and while it may be working, I did not have the patience to find out. Above 1C (and even
at 1C can be very impractical (not to mention dangerous?) on a larger battery bank. Personally, I haven't performed these tests on anything bigger than a 12V 36AH battery. My 8D didn't do much more than just kinda assert gravity when I ran these tests on it (not enough current, for sure).
Not sure what to tell you there... although I suppose the determined will find a way should anyone wish to try and replicate my results... Think I'll leave that one up to the people with the experience with very large batteries. I wouldn't want to begin to tell you how much current your 8000 pound forklift battery would need to pass in order to successfully perform this... if you get what I'm saying.
5 - Direction of pulse. It appears to me that the pulse needs to be of the discharge variety, rather than charging. This is kinda a mixed bag, but the tendency appears to be more effect from discharge pulses than charge pulses. May eventually look into doing both, but for now, this is where I'm calling it.
6 - Timing. This is last, but
definitely not least. I think this is the key to the whole thing. Desulfation should be considered a
preventative routine, not a recovery routine. Retaining plate material in it's usable state is
much easier than getting it back from the 'other side'.
One of the keys where I really believe this helped was when I recently accidentally left an amplifier on, and it drained one of my nicer sets of 18AH SLAs (36AH total) down to just 7.5V OTV (YIKES!!!).. I don't know what it was loaded, but just hoped that it wasn't as bad as I feared, then the news of the OTV set reality in quick. I needed to bring this thing back to life, and carefully, so I threw my very own recipe at it, and once I was done, I actually noticed an
increase in capacity compared to what it was before! YAY!
THE PROCEDURE USED IN THE TESTS, not just the recovery of the overdrained SLAs:
The most aggressive effect has been observed with the following procedure
at room temperature, with adjustments based on age/SoC, etc:
*** CHECK YOUR WATER FIRST if you're doing this on an FLA! ***
A - Slow discharge (< C/20, lower based on age) down to 10.5V LTV (loaded terminal voltage). A little higher might be better for older batteries, say 11V.
B - Release for 1 hour.
C - Begin Phase 1 charging, as described above.
D - Release for 4 hours.
E - Begin Phase 2 charging, as described above.
F - Release for 24 hours. This allows the battery's core temperature to stabilize and become even throughout the entire unit. Very small (5AH or less) may get away with less time here.
G - Begin Phase 3 charging, as described.
H - Release for 24 hours. The battery should be at 95% of it's usable capacity at this point. Let it sit and stabilize before proceeding.
I - Begin Phase 3
again, however this time:
Use a dump controller to pull 1C pulses of short duration to keep the regulated voltage just below the 'floating' voltage of the charger (set to the lower band of the absorption range, ie 14.4V), so that the overall tendency is toward charge. If it's set properly, you'll get very short pulses at about 20Hz (varies with conditions, but is close enough), and the battery will stick right around 14.2V. Let this run for as long as you feel warranted, but note that it will take quite some time for this process to work. During my tests, to keep things equal, I stuck with 24 hours. Real world usage of this information may end up being several cycles of this, or one very long run. Depends on you, and what you judge to be happening.
J - Release for 24 hours. This process will cause some heat buildup within the battery, which again needs to be liberated.
K - Begin phase 4, allow to run for 48 hours.
L - Disconnect completely for 24 hours.
If after the first 24 hours, the battery's OTV is about 0.2V higher than it was before you ever tried any of this, you have made some headway. In my tests, battery capacity doesn't improve much after the 0.2V increase in OTV is observed. If you haven't seen the full increase, try again, beginning with step "i" once more.
If you do not see any results out of this, you likely have reached the limit of what can be done with that particular battery. The sulfation may be too hard, and/or the problem is actually 'flaked and baked' plate material lying in the bottom of the cells. Neither of which you can do anything about.
Hopefully someone finds this information useful... It's a bit of time in the making.
I'm sure you guys will find a way to beat me down, but hey, that's why we're all here, now aint it?
Enjoy,
Steve