ELBC 2010 – Automakers Discuss Their Battery Requirements For Stop-Start Systems


John Petersen

Last week I spent three days at the 11th European Lead Battery Conference in Istanbul where I learned that I’ve been far too conservative in earlier articles that discuss the likely impact of stop-start idle elimination systems on the battery sector. To put things in perspective, the 10th ELBC in 2008 had 500 participants and two papers on stop-start systems. The 11th ELBC in Istanbul had 700 participants and 15 papers on stop-start, including three from major automakers. The stop-start papers took a full day of the 2-1/2 day conference program.

The high-level overview is that almost every major automaker is aggressively implementing stop-start idle elimination systems across their main product lines. Most forecasts expect penetration rates of 20 million cars per year within five years and emerging consensus is that stop-start will be standard equipment on all internal combustion engines by 2020, if not sooner.

The logic is inescapable; turning the engine off when a vehicle is stopped reduces fuel consumption and toxic emissions without impacting performance. The estimated fuel savings range from 5% in government mandated tests and 10% under real world city-highway driving to almost 20% in congested city traffic. No matter which figure you choose, it’s a very worthwhile target if implementation is widespread enough and cheap enough.

In his opening remarks, Ray Kubis, the president of Enersys’ European unit, explained that current stop-start systems typically use two batteries instead of one, and use higher quality batteries. This increases the battery content of new vehicles to two or three times historic norms. While a couple hundred dollars of additional battery value has a minor impact on the price of a car, it’s a huge opportunity for publicly traded lead-acid battery companies like Exide Technologies (XIDE) Johnson Controls (JCI) and Enersys (ENS) that can expect their OEM sales and margins to skyrocket over the next decade followed by sustained increases in replacement battery sales as the stop-start fleet ages. It’s an even bigger opportunity for developers of other advanced energy storage technologies that are better suited to the harsh demands of stop-start vehicles.

In the first automaker’s presentation, Andreas Stoermer of the BMW Group (BAMXY.PK) described a joint research effort between BMW, Ford Powertrain Research (F) and Moll Batterien that evaluated the technical requirements of stop-start systems and developed a universal testing protocol to accurately assess the impact of battery aging under real world stop-start operating conditions.

While theory of stop-start is both simple and rational, significant complexities arise from the need for a battery that can support accessory loads during engine-off periods, reliably restart the engine on demand and recharge as rapidly as possible to prepare for the next engine-off opportunity. To date, the European experience with stop-start systems has been less than stellar because the systems work great with new batteries, but rapidly lose functionality as the batteries age. In practice the frequency of engine-off events plunges during the first few months of driving. Based on this experience, automakers are rapidly coming to the realization that conventional lead-acid batteries are not robust enough to handle the demands of stop-start. They need something better.

The primary advantage of the BMW-Ford test protocol is that it’s technology agnostic and can be used with any battery chemistry and any combination of energy storage devices. The protocol is designed to focus on dynamic charge acceptance, or the amount of time required for a battery system to recover from the last engine-off event. The specific steps in the test protocol include:

  • A 60 second discharge at 50 Amps to simulate accessory loads during engine-off periods;
  • A one second discharge at 300 Amps to simulate the engine restart load;
  • A seven second rest period to avoid recharging the battery while the vehicle is accelerating; and
  • Measurement of the time needed to bring the system back to an 80% state of charge in preparation for the next engine-off opportunity.

The most fascinating part of the protocol is that the engine restart load is only 9% of the total energy associated with an engine-off event and the yeoman’s work is carrying the accessory load without interruption.

While BMW-Ford test protocol seems simple, it’s brutally punishing for battery systems because it focuses on maximizing the number of engine-off events in order to maximize fuel savings. There’s no escaping the fact that turning the engine off eight to ten times during a commute saves more fuel than turning it off two or three times.

BMW is apparently working with appropriate agencies to have the BMW-Ford test protocol adopted as the EU’s official standard for measuring the CO2 emissions reductions of stop-start vehicles. It is clearly more accurate than current EU standards that require a 20 minute test of a new vehicle with a fully charged battery. It also offers a more accurate long-term prediction of the fuel economy end-users will experience their daily driving.

In the second automaker’s presentation, Dr. Ed Buiel of Axion Power International (AXPW.OB) summarized the results of a recently completed joint testing effort by Axion and BMW that used the BMW-Ford protocol to evaluate the long-term cycling performance of four types of lead-acid batteries including:

  • A high quality valve regulated absorbed glass mat lead-acid battery;
  • A high quality AGM battery with high surface area carbon additives;
  • A high quality AGM battery with conductive carbon additives; and
  • Axion’s lead-carbon PbC battery.

The performance graphs for the first three types of batteries were nothing short of tragic because their dynamic charge acceptance plummeted within weeks after the batteries were put in service. The only battery to survive a five-year simulation with no appreciable performance degradation was Axion’s PbC.

A couple weeks ago I published a set of battery performance graphs that Axion presented at the 2009 Asian Battery Conference in Macau. At ELBC I learned that those graphs were interim results from the Axion-BMW testing program that was using the BMW-Ford test protocol. I haven’t received my electronic copy of the ELBC proceedings yet, but think two key graphs from Axion’s 2009 presentation in Macau bear repeating.

The following graph shows the rapid deterioration of a high quality AGM battery as the number of engine-off events increases. The blue line shows the maximum charging current the battery was able to accept as it aged. The black line shows the amount of time the battery needed to recover in preparation for the next engine-off event. The only visual differences between this graph and the full testing cycle graph presented at the ELBC are an inc
rease in the number of cycles completed and a gradual flattening of the dynamic charge acceptance curves.

8.26.10 VRLA.png

The next graph shows that Axion’s PbC battery does not suffer any negative effects from sustained rapid cycling, is able to accept much higher charging currents and has a predictably short recovery time. In an application like stop-start where maximizing the number of engine-off events will maximize system efficiency, the differences are critically important. Once again, the only visual difference between this graph and the full testing cycle graph presented at the ELBC is an increase in the number of cycles completed.

8.26.10 PbC.png

I was happy when I found Axion’s 2009 presentation on the Internet. I was even more pleased to learn that the 2009 graphs were interim results from a long-term testing relationship with BMW that was using a test protocol developed by BMW and Ford. I was delighted to see final graphs at the ELBC that tracked the performance of the PbC through a full five-year cycle life instead of the shorter test period presented in Macau. When I get my electronic copy of the ELBC proceedings, I’ll prepare an update to this article with graphs for the four battery types.

In the third automaker’s presentation, representatives from Renault explained the validation and test procedures that battery manufacturers would have to complete before their products could be considered for use in Renault vehicles. The first stage for all batteries is six months of validation testing followed by at least a year of rigorous performance testing. In their discussion of stop-start systems, Renault made it very clear that flooded lead-acid batteries would not work. While the Renault presentation held out some hope for AGM batteries, the results of the Axion-BMW tests make it pretty clear that AGM will not be an attractive long-term solution.

Based on everything I learned at ELBC, it’s clear that widespread commercialization of stop-start systems will require advanced energy storage products that are far more robust than conventional AGM batteries and can stand up to rapid shallow cycling. There is little question that Axion’s PbC is the current lead-dog in the race because it’s already completed over a year of testing with BMW and is the only device that has demonstrated the ability to survive the BMW-Ford test protocol. It still faces a variety of industrial engineering and scale up challenges, but at least for now the PbC appears to be the best emerging technology option.

Other possible contenders for a big share of the stop-start market include the Ultrabattery developed by CSIRO, NiMH batteries, lithium-ion batteries and multiple-device systems like the battery-supercapacitor product that’s being developed by Maxwell Technologies (MXWL) and Continental AG. The big challenge for NiMH and lithium-ion battery systems will be establishing comparable shallow cycling capacity with no performance deterioration at a competitive system cost. The big challenge for multiple-device systems will be overcoming rapid deterioration of the AGM battery that will carry the engine-off accessory load and ultimately be a gating limitation on system recovery time. I wish them all the best of luck because a vibrant market requires several credible competitors and it would be difficult for a small company like Axion to scale up production rapidly enough to satisfy the expected demand from automakers worldwide.

The next year to eighteen months will be very interesting times in the stop-start market. I expect 2011 to be an active year of testing and large-scale fleet demonstrations for competing energy storage products. By this time next year I expect automakers to announce their final design specifications for commercial rollout of stop-start systems beginning with their 2013 model year vehicles. Ultimately I think stop-start vehicles will be introduced with a variety of storage systems that will then have to prove their merit over time. After seven years of hard work, it’s wonderful to know that Axion will be running in the derby and can claim to be a pre-race favorite because of the work it’s already completed with BMW.

Disclosure: Author is a former director of Axion Power International (AXPW.OB) and holds a substantial long position in its common stock.


  1. ‘There’s no escaping the fact that turning the engine off several times during a commute saves more fuel than turning it off a couple of times.”

  2. If a commute has an average of 10 stops two minutes apart, a battery system that recovers quickly can take advantage of every engine-off opportunity while one that recovers slowly might only take advantage of three engine-off opportunities. This may seem like a simplistic example, but it’s exactly the kind of performance people have been getting in Europe. Ten stops per commute during the first week down to one or two per commute after a couple months.

  3. Oh … the words several and couple are close to synonyms to me … I think you might want to tweak that sentence to be more explicit.

  4. The e-mail went out with the original text, but the revised article now reads:
    “There’s no escaping the fact that turning the engine off eight to ten times during a commute saves more fuel than turning it off two or three times.”

  5. Thank you for a well written article. Taking your example of an average of 10 stops 2 minutes apart,how many engine off opportunities would an Axion PbC battery be able to take advantage of? In my capital city (KL, Malaysia), 10 stops 2 minutes apart would not be an exaggeration. To the uninitiated, it is not obvious what TOC, TDCV, PDRV and EODV mean. It would be useful to add a legend to the graphs.

  6. With a recovery time of 30 seconds, it would be able to handle all of them. In fairness, VRLA might be able to do the same. When you go to 10 stops a minute apart, the PbC would be about twice as effective. If your traffic in KL is anything like the traffic I saw in Istanbul, the one minute intervals are probably more common than the two minute intervals.

  7. I didn’t define TOC, TDCV, PRDV and EODV because I didn’t want to make a rookie’s error in terminology. I’ll make a point of addressing those issues after I get my copy of the ELBC proceedings.


Please enter your comment!
Please enter your name here