There are at least three variations of the lead-acid battery in current automotive use (1). In the most common configuration, the car battery has six cells, each producing about 2.1 volts. Thus the total battery output voltage is about 12.6 volts.
It is the "about" in the previous two sentences that leads to trouble when a charging system is not matched to the type of battery installed.
The three major contributors to battery chemistry are lead, lead dioxide, and sulfuric acid. Unfortunately pure lead is too soft to withstand the physical abuse of mobile applications, so about 6% antimony was added to strengthen it. This led to another problem -- water usage.
Antimony added to the lead grids acted as a catalyst and made outgassing (loss of hydrogen and oxygen during use) worse, and frequent water replenishing was required. So battery manufacturers looked for another material that could strengthen the lead grids.
Calcium was added to both the positive and negative electrodes in the early 1970s. It reduced outgassing enough to allow manufacturers to claim they building "maintenance-free batteries".
However, lead/calcium batteries are not very resistant to "deep-cycling" (deep discharge followed by a full charge). This made them inappropriate for uses such as to power trolling motors in fishing boats. It also required a higher charging voltage. General Motors studied the charging characteristics of lead/calcium batteries and set the voltage regulators of cars equipped with the "Delco Freedom II" battery at 14.8 volts. Lower settings prevented charging to full capacity. This is too high for lead/antimony batteries and will cause them to lose water rapidly.
The third type of battery frequently used in automotive service uses "hybrid" construction. Its positive grid is strengthened with antimony, and the negative grid with calcium. Water usage is greatly reduced, although regular checking is advisable.
The hybrid battery is more resistant to deep cycling than the lead/calcium, but is still not as good in this respect as the original lead/antimony style. Most cars supplied with hybrid batteries have their voltage regulators set to 14.3 volts, although it is reported that the higher 14.8 volt setting of GM cars does no harm if the water level is checked regularly.
A fourth type, the "gel cell", has appeared in automotive use. Early reports were that its manufacturer recommended charging at 13.8 to 14.1 volts (2), suggesting it used hybrid chemistry. However, it also does not allow the user to add water, so it may indeed be lead/calcium.
Older vehicles with voltage regulators set at about 14.0 volts simply will not fully charge lead/calcium or hybrid batteries. Stories abound of cars with older regulators leaving their owners stranded on cold mornings shortly after a "new, improved, maintenance-free" battery was installed. Even two weeks of sitting in the garage, with no load other than the electric clock and burglar alarm, can discharge a battery if the voltage regulator prevented it from being properly charged.
The challenge is particularly severe in the case of foreign manufacturers. Many firms are so certain, and so proud, of their engineering expertise that they have declined to modernize their specifications to meet present battery specifications. Bosch, for example, still teaches its "factory-trained" service technicians that 13.8 volts is enough. As late as 1992 the Product Support Manager of their Automotive Service Division wrote (3) that "13.6-13.8 VDC may in fact be adequate", and that "a fully charged battery cannot be charged with 14 VDC or higher. Doing so will cause the battery to be overheated and damaged".
Only slightly contradicting him is a statement from Bosch's Customer Relations Representative (4) that "the alternator and voltage regulator should charge between 13.8 and 14.2 volts". This representative also recommended a local Bosch Authorized Service Center, who assured me that "13.7 volts is enough".
Mercedes-Benz (5) offered their usual advice: "we suggest the installation of an original Mercedes-Benz battery".
The National Technical Director of the Mercedes-Benz Club of America (6) was considerably more flexible: "the specified voltage for Mercedes-Benz cars of the mid-1980s is 13.0 to 14.5 volts". But clearly 13.0 volts is not enough, and Bosch says 14.5 "will overheat and damage the battery". In an early 1993 letter (6) he says local MB technicians tell him that alternators usually check out around 14.2 to 14.4 volts, and that they replace regulators that call for less than 14.0 volts - thus contradicting the Bosch Product Support Manager. Are we having fun yet?
Of course, battery chemistry is also a function of temperature. Modern regulator designs usually charge to higher voltages in cold weather. Bosch has offered this graph in a technical manual (7) covering alternators. Although they do not mention different types of battery construction in this 1982 manual, the graph would appear to display the lower limit of appropriate voltages for hybrid batteries. Too bad their own "factory-trained service technicians" choose to ignore it! By the way, these are stated to be the permissible tolerance ranges of the regulator at an alternator speed of 6000 RPM and a load of 5 Amps.
Personal experience clearly reinforces the dependence of battery performance on a good match between the regulator and the battery chemistry. A 1973 Mercedes whose regulator delivers 14.1 volts keeps its battery well-charged even though the car is driven infrequently and only for short distances. A 1982 model whose regulator provided 13.6 volts would not keep its battery charged - a trickle charger was necessary in cold weather. A 1988 300TE charging at about 13.9 volts has frequent battery problems, while a 1991 560SEC which charges at about 14.1 volts has caused little trouble.
In a more recent document (8), Bosch recognizes the effect of calcium on battery chemistry. Elsewhere in this same 1995 handbook they recommend that when external chargers are used, lead/calcium and hybrid batteries be charged at no more than 14.4 volts and the charger have a certain characteristic known as "Type IU". The "IU" characteristic is well-defined in this reference. However, under the more controlled environment of an automotive charging system (the regulator is temperature-compensated), voltages greater than fifteen are sometimes useful. At a modest air-intake temperature of 25º C (77º F), this chart suggests settings between 14.3 and 14.7 volts.
At one time Interstate offered a line of "Extreme Performance" batteries (9). Recently they reappeared on their Web site under the name "Optima Spiralcell", so they may have a cooperative agreement with Optima. Optima has added a little silver to the plates and some sodium sulfate in the electrolyte, and fabricated the lead plates in a spiral. Their main claim to fame is their ability to be charged very quickly, allegedly due to the greater surface area of their plates and the ensuing lower internal resistance (on the order of 3 milliohms). Since there is no way to add water, they may be lead/calcium (+silver), and the higher charging voltages would apply.
Their charging advice supports this theory. They suggest that alternator settings anywhere from 13.8 to 15.0 volts are appropriate, but that they can be rapidly charged at as much as 15.6 volts. Elsewhere in the same document it is suggested that for constant-voltage charging 14.7 to 15.0 VDC is correct (supporting the "lead/calcium" hypothesis).
Owners on various mailing lists have reported their satisfaction with adjustable regulators. One source mentioned is Transpo Electronics, and another is Doug Sogolow.
If you have a choice, try to get your system to charge within the range suggested by Bosch's newer graph. 14.5 volts at moderate temperature is not a bad target. The improvement in life and performance of lead/calcium and hybrid batteries will be worth it!
Additional information can be found at various places on the Internet(10). Those who don't mind a little math and physics can find a more complete explanation from ThermoAnalytics, in the context of hybrid and electric vehicles(11).
(1) "Batteries: Current Designs", Mike Dale, Motor magazine, November 1985.
(2) Email from Ronald Dwelle, 13 February 1999.
(3) Private correspondence from Dieter Richter (Bosch), 14 September 1992.
(4) Private correspondence from John Haralamos (Bosch), 28 August 1992.
(5) Private correspondence from Thomas Trivento (Mercedes-Benz Owner Service), 1 October 1992.
(6) Private correspondence from Frank King (MBCA), 20 January 1993.
(7) "Technical Instruction Manual - Alternators", Robert Bosch GmbH #VDT-U 1/8 En (12.82), June 1982.
(8) "Automotive Electric/Electronic Systems", Second Edition, Robert Bosch 1995.
(9) "Extreme Performance", Interstate Batteries Web site © 1998, URL = http://www.interstatebatteries.com/products/product_line/extreme_specs_charging.htm, (page no longer found)
(10) "Car Battery Frequently Asked Questions", Bill Darden, URL = http://www.repairfaq.org/ELE/F_Car_Battery.html
(11) "Battery Modeling for HEV Simulation by ThermoAnalytics Inc", URL = http://www.thermoanalytics.com/support/publications/batterymodelsdoc.html