The Eighth Grade Lab Report
If we were grading lab reports for an eighth grade science class, and Practical Sailor1submitted their battery testing article published August 1, 1999, we'd have given them a failing grade.
Besides misinformation that could be deadly, and numerous other errors, that report is lacking any serious data that convinces us actual capacity tests were made. The conclusions drawn fall pretty much in line with marketing claims made by battery manufacturers. We understand that magazines who accept ads must practice the fine art of appearing to evaluate products for their reader's benefit, all the while not offending advertisers.
The Dearth of Data
There is some data presented in the article. The biggest table appears to be a compilation of information taken from battery data sheets ...that is marketing information produced by the firms selling batteries.
The second table is one that shows voltage versus state-of-charge for flooded cells with a fully charged specific gravity of 1.265. This table, according to the article, was adapted from BCI Tech Manual. We presume that BCI stands for Battery Council International, but that isn't explained in the article, and we doubt that most readers would know the meaning of BCI. In any case, we're not convinced that the table is specific to the deep-cycle flooded batteries being tested, and are certain that it isn't for the gel and AGM batteries.
The Practical Sailor article claimed to test each of seven batteries for reserve capacity and 50% capacity. The official term is reserve minutes, rather than reserve capacity, but we understand their confusion, and that of others who ask us about reserve capacity.
Data for reserve minutes is presented in the big table, (they call it the Value Guide), at the bottom of each battery column.
No data showing 50% capacity is presented, and the method they claim was used to measure it seems to us fraught with great difficulty where numerous trial and errors would have to be made. Since there is no data provided for the tests, we think it reasonable to ask, were the tests really made?
Analyzing Reserve Minutes
The data in the table below is taken from the Practical Sailor article, but we have calculated the error between the claimed reserve minutes and measured reserve minutes.
|West Sea Volt||175||151||15.89|
|West Sea Gel||160||171||6.88**|
Note that the West Marine Sea Gel actually had more reserve minutes than claimed, 171 measured versus only 160 claimed. We put a couple astericks by that error since it's in the consumer's favor. Two numbers really glare at us, the percentage of error for the Rolls and the West Marine Sea Volt, both flooded cells. We're in the habit of discounting marketing claims on batteries from 10 to 25%, so we're not shocked by the numbers. We'd interpret the error percentages with a simple question ...who lies the most?
How does Practical Sailor observe the same information? First, they're pretty sure their measurements are reasonable, but with no other data presented, we as a reader, are not so sure that something isn't amiss. But here's what Practical Sailor has to say about the Rolls reading. ``While it fell a little short of its rating in the reserve test, note that it also had the highest listed reserve capacity(sic) of those tested''.
We'd say the the West Marine Sea Volt was a little worse than the Rolls, by 0.04% to be precise, but even if Rolls' overstatement of their rating by 15.85% can be construed as a little short, we are completely baffled by how that can be justified by overstating the rating in the first place.
Impressing the Lab Instructor
No eighth grader is going to write a lab report without embellishing it with names dropped in here and there. In the Practical Sailor article, we learn that an EMROL BATTEST 12-20-3 was used to discharge each battery with a constant current. From that discharge, minute by minute results were downloaded to a Toshiba computer. Did the article present any of that data? No. Why not?
But, a bigger question can be asked? Does a constant current discharge give the same kind of results as a battery being discharged in a typical marine electrical system? No, again. Depending on the load current relative to battery capacity, the difference may be slight, or significant. If you want tests to represent actual usage, why choose a discharge method different than actual?
We presume that capacity testing was also done with a constant current discharge. Interestingly, Practical Sailor presents no data about the results of capacity testing. There's an explanation of how it was done, but as we stated earlier, the method chosen would involve more trial and error cycles than would be convenient, and, if you have to wait 5-24 hours after the discharge to measure open circuit voltage, testing is going to involve many days, if not months.
It's a tenet of big league science to present enough method and data so that your work can be reviewed and duplicated by others. The Practical Sailor article is without data ...which saves potential embarassment should the data turn out to be wrong. Missing data is most suspect, however.
Life Cycle Testing
Practical Sailor readily admits they didn't test life cycles. That's a plus. In the absence of actual tests, the next best thing to do is quote numbers gleaned from battery data sheets and other industry claims. In doing just this, the article makes a claim that the best flooded cells can deliver over 900 cycles. Since none of the batteries tested claimed that many cycles, we're in the dark as to what the best flooded cells might be.
And being a skeptic, we'd ask how the number 900 came about? Was the battery test done on a machine simulating a rough sea where the electrolyte was continually washing the active material on the plates? Was recharge done over many hours, or was the battery pumped to the max for an hour or two? Was the test done within a saltwater environment? Was water replaced religiously, or in the more hit or miss style of life at sea?
Before we'd believe 900 cycles on a flooded cell, there would have to be a lot of convincing tests done ...which to our knowledge have never been done.
Charge Acceptance Rate
Supposedly recharge times were also monitored, and they concluded that the gel battery was a little faster than the flooded cells but the AGM was at least 20% faster. We're not sure what the 20% refers to, but then, we don't have the charge times because they simply aren't reported. It seems to us a little unethical to measure something, draw a conclusion, but omit the actual measurements. Maybe one just needs to develop blind faith in Practical Sailor.
Suppose that Practical Sailor had presented all the test methods and the data so that we could either duplicate their results or logically dispute them? Would anyone really care?
Practical Sailor tested Group 27 batteries which have roughly 100 Amp-hours. Several of these would be required to make a reasonable bank for a cruising boat, but no Group 27 battery will have the cycle life of a 4D or 8D battery, or the 6 Volt golf cart style units. Building a house bank with Group 27 batteries would be a last resort from our perspective.
Sulphuric Acid and Saltwater
Chlorine gas, deadly in small doses, is the result of mixing sulphuric acid with saltwater. We've been pooped at sea, and while we probably only lost a few drops of electrolyte, it's a situation that we never want to repeat. Even if we were convinced that a flooded cell could offer twice the cycle life of a gel or AGM battery, we'd not venture out of protected waters with them.
Of course, we're not convinced of any such claims for cycle life. Perhaps we've seen too many of those expensive flooded cells piled up on pallets ready for the final journey to the recyclers. Maybe our own cycle tests have been unfair to flooded cells because we've tried to duplicate actual life in a vessel, or, perhaps we realize that cycle life isn't the holy grail.
Suppose you could purchase batteries that did indeed offer twice the cycle life, but because of inefficent energy deliver, and slow charge characteristics your cost of ownership was actually greater, maybe as much as 50% greater, would those extra cycles seem like such a bargain?
Watt-hours versus Amp-hours
In the April 1992 issue of Power News, we suggested that rating batteries in Watt-hours as opposed to Amp-hours would be a more accurate way to rate batteries. This theme was elaborated on further in our book Wiring 12 Volts for Ample Power. Watt-hours gives the actual work that a battery can perform, while Amp-hours does not.
It's an indisputable fact that the thicker the battery plate, the less it's ability to conduct current. It takes time to diffuse electrolyte through a thick plate, and without ready movement of ions, current just can't happen. The Watt-hour density of thick plate flooded cells is significantly less than gel or AGM batteries, especially at high rates of current.
In the August 1992 issue of Power News, a comparison of Watt-hours between a gel battery and a moderately thick plated flooded battery showed that the gel battery had about 13% more Watt-hours. In other words, the gel battery can do 13% more work for the same Amp-hour rating.
Hazardous to Your Health, and Other Errors
While there are several outright errors in the article, there is one statement which could turn out to be deadly. They report that an advantage of gel batteries is the ``absence of explosive or corrosive gasses being vented during charge''. Tell that to the owners of vessels who have had on-board explosions caused by hydrogen and oxygen vented from gel batteries! During overcharge conditions, gel and AGM batteries do emit explosive gases. You must provide sufficient ventilation to avoid concentration of hydrogen and oxygen, just with flooded cells. To do otherwise is risking your life.
Another error in the article calls into question what Practical Sailor truly knows about batteries. Their explanation of thermal runaway is inverted from reality. The resistivity of electrolyte is a major determinant of overall battery resistance. It's a documented fact that the resistivity of electrolyte decreases with temperature. According to Practical Sailor, ``heat build-up causes more resistance''.
Here's another boner that could lead to dangerous overcharges. According to Practical Sailor, ``batteries used daily should be equalized monthly''. The fact is, batteries used daily are much less likely to need equalization than batteries which are floated for long periods. With good multi-step, temperature compensated charging, equalization is rarely needed, and its need can always be determined with a hydrometer. If you need to equalize monthly, you have a very poor charging system! Equalization carries a risk of thermal runaway, and should only be done if you completely understand the process.
According to Practical Sailor, ``lead-acid batteries have been with us for decades''. Would you believe a century and a half? The first viable lead-acid battery was developed in 1859 by a French scientist named Planté.
And, ``years ago'' Peukert figured out the relationship of discharge rate to capacity. We think that 1897 is more than just ``years ago''.
According to Practical Sailor, ``even slight overcharging can cause permanent loss of capacity'' in gel batteries. Is slight here in the same range of a little short used to describe Rolls failure to meet their reserve minutes claim? If so, then overcharging a gel battery by 15.85%, or about 16.33 Volts would cause permanent loss of capacity, if done for several hours a day for months. Some people think that applying more than 14.1 Volts to a gel battery turns them into instant toast, and that's the impression we get from Practical Sailor. It's true that you can't replenish electrolyte in a gel or AGM battery, but it's also true that slight overcharges are not damaging.
In the end, it's a simple matter of chemistry. Electrolysis, that is, the breakdown of electrolyte is a direct function of Amps times time. How many Amps of overcharge for how long does it take to vent a damaging amount of hydrogen and oxygen from the battery? Electrolysis consumes water at 0.336 cubic centimeters per Ampere-hour of overcharge. Until the internal pressure of the gasses exceeds the 1-2 PSI where the vent seals release, hydrogen and oxygen are recombined in the gel and AGM cells. Overcharges which force the vented caps open are damaging if maintained long enough.
Practical Sailor report that plates in flooded cells ``are often strengthened with antimony, helping to reduce shedding of active material''. Antimony is added to lead to strenghten the plate, not only for its benefit in automation of plate manufacture, but for its benefit in portable applications where plates are stressed from vibration. Because lead-antimony plates pour in the mold better, they have sharper edges, which does tend to hold active material in place better. Lead-antimony plates have a lesser expansion coefficient than does pure lead. Antimony goes into solution during overcharge, and also leads to self-discharge. It lowers the voltage where gassing starts, so more water loss occurs during recharge. Antimony is also the source of deadly stibine gas produced during charge.
Practical Sailor goes on to say that ``antimony also promotes gassing, which contributes to shedding when overcharging occurs''. Antimony does cause gassing to occur at a lower voltage, which causes corrosion mainly on the positive plate. But shedding is also significant at a low state-of-charge when the active material is soluable and where motion sets the liquid electrolyte into an erosion mode.
According to Practical Sailor, plate shedding, followed by sulfation are the major failure modes of batteries. Neither one of these is correct. The major cause of battery failure is overcharge which causes corrosion of the positive plate ...about 40% of failures. Shedding active material may be the culprit in 7-10% of the failures, whereas sulfation may be the cause of 10-20% of failures.
Had Practical Sailor been interested in your health, they might have pointed out that lead-antimony plates produce stibine and arsine which are known health hazards in closed environments.
Practical Sailor would have you believe that a battery must be ``in top shape'' before Peukert's equation is of any use. In fact, Peukert's equation can be calculated at anytime to match the state of the batteries, so its use is applicable to new and old batteries.
In their sidebar titled ``How to Maximize Battery Life'', they omitted a critical parameter, that of temperature compensation. Battery life cannot be maximized without correction of the charging voltage relative to battery temperature.
As is usual, Practical Sailor quotes the 50% rule without reference to its source, and in a way that mis-leads people to think that 50% discharges are a way to maximize total life cycle Amp-hours. If they have a study that demonstrates that interpretation of the 50% rule, we'd very much like to see it.
Practical Sailor failed to provide any new information in their article of August 1, 1999. They failed to provide any data to substantiate their conclusions. The batteries tested are not representative of a typical energy system. The test method of discharge is not related to the way batteries are used in energy systems. Comparing batteries based on Amp-hours is obsolete.
While some errors in the article are of little consequence to the average user, other errors could be dangerous. In either case, the errors indicate that complete understanding of batteries eludes Practical Sailor and calls into question the value of their article ...we think that the errors noted, and the fact that real data is not reported give the article a negative value.
Practical Sailor totally ignored the cost of ownership of the various battery types. This cost of ownership includes the invoice cost of the batteries as well as the cost of recharging them. If you're tied up to the dock, charging at $0.05 per Kwh, the cost of charging batteries is insignificant. If you're running an engine to recharge, however, the difference between $3 per day versus $9 in fuel and engine wear adds up in a hurry. Unless you're relying mostly on solar or wind power, thick plate flooded cells cost over and over again by the way of slow charging, not to mention the aggravation of having to listen to an engine run incessantly.
If you're going offshore, we'd suggest going with gel or AGM batteries. When you're anchored offshore, run your engine for an hour a day instead of three, and take the extra time and money to find the nearby fishing or watering hole!
Ample Power products are manufactured by Ample Technology,
2442 NW Market St., #43, Seattle, WA 98107 - USA