ePower's concept for building a Class 8 Tractor with a hybrid series electric drive started with Jay Bowman, age 57, who describes himself as a "Jack of all trades" and has spent over 30 years making machines work together in manufacturing environments. Jay and Andy met while Jay was working as a consultant to Andy's former employer, the Ventures Group of Nova Chemicals, which had bought Jay's modular building products company. While Andy was skeptical when Jay first talked about his plan to build a hybrid Class 8 Tractor, he quickly saw the light when Jay built a prototype with a 97 horsepower diesel engine mated to a 60 kW generator and a 60 horsepower electric drive motor in a cab-over Mac chassis.

(click to enlarge)

Jay's Gen-1 prototype was underpowered, but it got an impressive14 mpg at 35 mph with a 45,000-pound load and served as indisputable proof of concept. Based on those results, Andy joined the team and ePower started working on a Gen-2 prototype with a Peterbuilt chassis that offered more "real estate" for heavier components that could run at highway speeds with an 80,000-pound load.

Throughout the development process ePower's primary goal has been to explore the outer limits of Class 8 Tractor fuel efficiency by using the smallest possible components to power an 80,000-pound vehicle at highway speeds. The current configuration uses:

• A 197 horsepower four cylinder John Deere turbo-diesel;
• A Marathon generator that produces 128 kW at 1,800 RPM;
• A 150 horsepower Marathon electric drive motor; and
• An off-the-shelf five-speed automatic transmission.

The engine is governed to run at a steady 1,800 RPM from the moment it's started until the moment it's turned off. That engine speed is apparently the fuel efficiency and emissions sweet spot and it provides a constant 128 kW of power for the drive and battery systems.

I made a point of asking about the transmission since the Gen1 prototype didn't have one. Jay explained that electric motors have great instantaneous torque, but torque falls off as motor speed increases. The drive motor they chose for the Gen2 prototype is most efficient at 3,600 RPM. So instead of forcing the motor to operate faster than its optimal speed, the transmission lets them keep everything in balance.

I took this picture of a spare drive motor and transmission they had sitting in the shop. The plastic chair in the background gives you a rough idea of size. The combination is probably 2 feet in diameter and seven or eight feet long.

(click to enlarge)

When Brad and I drilled down into the question of engine size, Jay explained that it would be relatively simple to use a six-cylinder engine with a bigger generator and drive motor, but each substitution would cut fuel economy. Since ePower's primary focus at this point is maximizing fuel economy, they're using small components for the prototypes even though they assume that some operators will be willing to trade a little fuel economy in return for more power in challenging terrain.

In their description of how the ePower drive works, Andy and Jay explained that the diesel generator running at 1,800 RPM has plenty of power to accelerate an 80,000-pound GVW tractor and trailer from zero to 35 mph, and to power an 80,000-pound truck at 65 mph on level ground. They also explained that without supplemental boost from the battery banks, the truck is a bit of a dog when it comes to accelerating from 35 to 65 mph or climbing hills.

As I noted earlier, the diesel engine is set to run at a steady state speed of 1,800 RPM, which is the optimal fuel efficiency and emissions point for the engine. That engine speed, in turn, keeps the generator output at a steady 128 kW. The drive motor has first priority on the available power, but any power that isn't used for the drive motor is shunted directly to the battery banks to keep them at an optimal state of charge.

When the truck is running steady state at highway speeds, the generator has 6 to 8 kW of excess capacity that provides a slow trickle charge to the batteries. When the driver takes his foot off the accelerator, however, the generator output available for battery charging can quickly ramp to 128 kW. When you add regenerative braking current from the drive motor, it's not unusual for the combined system to dump a transitory 650-volt charging current at up to 300 amps (~195 kW) into the battery banks. In other words, the transitory charging currents during braking events are enormous and those currents frequently gave rise to some fairly catastrophic failures with both flooded and AGM batteries.

While the original Gen2 design used 56 AGM batteries to provide 650 volts of battery boost, Jay decided to increase the number of PbC batteries to 60 so they could keep the batteries at 11-volts and avoid the top end of the PbC's charging curve where the internal resistance increases rapidly. To date, they've limited the cycling range from 11-volts down to about 10-volts, which hasn't put much strain on the batteries. Over the next two weeks they will slowly increase GVW from the current weight of 50,000-pounds to the target weight of 80,000-pounds, and increase the depth of discharge as they reduce the cut-off point into the 5-volt range. They'll also be adding a control chip to the diesel generator that will boost engine speed beyond 1,800 RPM for hill climbing.

When Jay began to describe the differences he's observed between the PbC batteries and the flooded and AGM batteries he used in earlier prototypes, he explained that the other batteries were always in a state of conflict with his machines. As a result, the machines weren't happy and neither were the batteries. When he started talking about the PbC he got a big smile as he explained that his machines are "happy" with the PbC and the PbC seems to be "happy" with his machines. The conflicts and tensions he's observed with other types of batteries simply don't exist with the PbC batteries. Jay's reserving judgment until he's had a chance to really punish the PbC batteries with a deep discharge and heavy recharge, but he's clearly a man who thinks he's found the batteries his system needs.

While the PbC batteries were delivered in mid-December, ePower experienced a number of component failures that delayed the installation and testing. The first problem was an over-voltage spike from the generator that fried the control electronics. Then they had some minor wiring issues followed by a capacitor failure in the drive motor. While each of these problems introduced a delay, none of the problems was directly or indirectly attributable to the PbC batteries. By the time Brad and I arrived, the problems had been resolved and the tractor was fully operational, as we experienced first hand during our brief test rides.

My test ride in the ePower tractor was quite an experience.

The first thing that surprised me was there was no change in engine pitch when the truck was sitting at a stoplight, accelerating or decelerating. The driver explained that the engine is governed to run at 1,800 RPM and it never changes. They apparently plan to add both idle and idle elimination functionality that will kick in if the truck isn't moving and the batteries are fully charged, but for now they view it as a minor issue. They also plan to add electronic cruise control to improve efficiency even further, but they don't want to increase electrical complexity until they've finished characterization work on the batteries.

The second thing that surprised me was the quickness of the tractor's acceleration from a dead stop. It moved smoothly through the gears and didn't feel all that different from riding in an SUV. While part of the quickness was undoubtedly due to the absence of a trailer, the driver explained that he had to be careful in heavy traffic because other drivers frequently underestimate the speed and responsiveness of the tractor. He also explained that even with an 80,000 GVW, the tractor was extraordinarily quick because of the electric motor's low-end torque.

The last subject Brad and I drilled down into was the question of ePower's retrofit cost and the incremental cost of PbC batteries.

Andy explained that they landed on a $70,000 price point a couple years ago because they expected the ePower retrofit to compete with a conventional five year major overhaul that cost about $35,000. While the cost of a conventional major overhaul has risen sharply in the last couple years and can run up to $60,000 for a rebuilt Tier Three engine, ePower has stuck with the $70,000 price point because its the number they've discussed with potential users and they don't want to change their pricing without a lot more proof of performance than they have today. When you account for the incremental cost of the ePower retrofit and the incremental salvage value of the old engine that won't be replaced, ePower reckons that the cost of its retrofit is basically a wash when compared with the cost of a rebuild.

A conventional rebuild, of course, won't offer the 50% fuel savings that ePower expects from their hybrid drive.

The only complaint ePower has about the PbC battery is pricing. The PbC batteries cost about twice as much s the AGM batteries they used on the last version and three times as much as the flooded batteries they used in the Gen1 prototype. That being said, ePower originally expected a two-year replacement cycle on their AGM batteries and they think a five-year replacement cycle is possible for the PbC. Since their prior experience with AGM batteries showed that problems started to arise within six months and got pretty severe within a year, they think the PbC will be worth the price premium if it has the durability and service life Axion expects.

I'm sure I've overlooked any number of issues readers will want to explore in greater detail and I encourage you to ask questions. I'll provide the best answers I can and hopefully Brad will chime in with more color when he gets a chance. I was very impressed with what I saw during our afternoon at ePower and think this could be a tremendous market for the PbC.

Disclosure: I am long [[AXPW.OB]].

Additional disclosure: I'm a former director of Axion and I was paid a discounted day rate for visiting ePower with Mr. Warneke. In the future, I hope to be retained as legal counsel for ePower.

Thanks to some welcome help from HTL, I've created a spreadsheet that starts with the daily data beginning in January of 2010 and then aggregates the values on weekly and monthly basis. As the last step in my analysis, I've calculated a monthly average percentage and the standard deviation among the monthly percentages.

For the 34 month period beginning with January 2010 and ending with October 2012, monthly short sales averaged 32.06% of total reported trading volume and the standard deviation was 9.85%.

In November the short sale percentage dropped to 17.2% of total volume and in December short sales were only 12.49% of total reported volume. It was basically 2 Sigma event.

Now that we're almost done with January and short sales for the month to date are 12.61% of total reported trading volume, I think we're firmly in the grip of a "new normal" where the big uglies are gone and the flow of old shares into the market is basically over.

Today I decided to put the monthly "short sales as a percentage of total volume" numbers into a graph for those who are more visually inclined. Here it is:

(click to enlarge)

Until very recently, monthly short percentages below 25% were very rare. Now it looks like 12.5% may be the intermediate "new normal." It will probably take a couple more months for the short sale numbers to bottom, but I won't be surprised to see something significantly lower than 12.5%.

JANUARY 26th ADDITION: This morning a comment from Iindelco got me scrambling to see if I could find a way to show the short percentage as an overlay on a bar chart that also shows total monthly trading volume. Here it is:

(click to enlarge)

I think the new graph layout was a great suggestion.

Disclosure: I am long [[AXPW.OB]].

(click to enlarge)

The events that increased market capitalization during 2009 were:

• The April 2009 announcement of a strategic alliance with Exide;
• The August 2009 announcement of a DOE grant award to Exide with Axion Power; and
• The December 2009 completion of a $26 million private placement.

Since the effective date of the resale registration statement for the shares sold in December 2009, the selling pressure has been heavy enough to completely eclipse all of these events:

• The June 2010 disclosure of a development relationship with Norfolk Southern;
• The September 2010 disclosure of a development relationship with BMW;
• The March 2011 disclosure that a major US automaker had joined Axion as a subcontractor in a major DOE grant application;
• The November 2011 commissioning of the PowerCube as the first behind the meter frequency regulation resource in the country;
• The December 2011 discovery that the unnamed US automaker was General Motors;
• The April 2012 announcement that Norfolk Southern had completed its laboratory testing and ordered batteries for the NS 999;
• The August 2012 disclosure that BMW had completed its laboratory testing and commissioned an independent peer review;
• The August 2012 disclosure that a Top-5 Asian automaker had decided to go directly to advanced testing on the strenght of the BMW test results;
• The November 2012 disclosure that a testing program for the use of the PbC in auxiliary power units for Class 8 trucks was expected by year-end;
• The November 2012 disclosure that the PbC had been selected as a replacement for the AGM batteries used in a new series hybrid electric retrofit for Class 8 tractors;
• The tacit admission by Exide that its best enhanced and AGM batteries won't stand up to the demand of micro-hybrids in November 2012; and
• The frank admission by JCI that its best enhanced and AGM batteries won't stand up to the demand of micro-hybrids in December 2012.

While each of these events would have been big news in a typical micro-cap company, they didn't register on Axion's price chart because of the unusual market dynamics that prevailed when the announcements were made.

While Axion's stock has been "broken" for the last three years, I believe the market dynamic that caused the problem has been resolved and the only thing that's holding the stock at present levels is fear that higher prices will only give rise to another round of heavy selling. After three years of unrelenting selling pressure despite an increasing body of proof that the PbC is an extraordinary new battery technology, I understand the fear. I also know that Axion has arrived at a transition point and is poised to shed the R&D company market dynamic that prevailed for the last nine years as the PbC earns a place in several billion-dollar niche markets where competitive battery technologies simply can't do the work.

Disclosure: I am long [[AXPW.OB]].

(click to enlarge)

After a series of phone calls and e-mails, Andy graciously sent me a technical presentation on ePower's series hybrid drive and gave me permission to share the presentation with readers and discuss ePower and its technology in greater detail. Click here to download a copy of ePower's presentation.

ePower Engine Systems LLC is a closely-held advanced transportation technology developer that's using inexpensive off-the-shelf components to bring series electric drive, the mainstay of the nation's rail transportation system, to highway transportation. Their goal is to narrow the fuel efficiency gap between 480 ton miles per gallon for railroads and 110 ton miles per gallon for heavy trucks.

In a truck with series electric drive, there is no mechanical connection between the engine and the wheels. Instead, the engine powers a generator and electricity from the generator powers an electric drive motor. This configuration maximizes fuel efficiency by running the engine at its optimal RPM and eliminates the need for complex heavy truck transmissions while delivering the instantaneous peak torque of an electric motor.

In furtherance of their goal to maximize fuel efficiency, ePower takes series electric drive a step further by sizing the generator for steady vehicle state operations at highway speed and using an array of 52 PbC batteries to provide additional power for acceleration and hill climbing, and increased energy savings from regenerative braking. The ePower drivetrain is a true series hybrid electric drive and a first for the trucking industry. The design is suboptimal for mountainous routes with substantial elevation changes, but it's extremely efficient in flatter terrain.

(click to enlarge)

While a typical Class 8 tractor operating in the US with an 80,000 pound gross vehicle weight achieves fuel economy in the 5.2 mpg range, the same truck with an ePower system will deliver fuel economy of 10 to 14 mpg, values that crush the DOE's 2018 SuperTruck target of 6.8 mpg for conventional heavy trucks. It works out to an annual fuel savings of roughly 11,500 gallons per vehicle.

During the startup phase, ePower has focused on the retrofit market because around 37% of the 2.7 million trucks in the US-fleet are more than five but less than twelve years old. These trucks have outlived their original drivetrain warranties and are often less efficient than newer trucks, but they have substantial remaining useful life in their chassis, bodies and other components.

The cost of converting a tractor with a conventional diesel drivetrain to a series hybrid electric drivetrain is approximately $70,000 (batteries included) and ePower believes its retrofits will pay for themselves through fuel savings alone in 18 to 24 months.

Currently, ePower is doing all required retrofit work in its own facility. Once its system is fully developed and proven, ePower intends to provide the necessary conversion components in kit form for sale to certified installers including fleet operators and other service entities. It also hopes to license its technologies and systems upstream into the OEM market.

ePower's original design used absorbed glass mat, or AGM, batteries to provide acceleration and hill climbing boost. Unfortunately, the AGM batteries were poorly suited to long-string use and ePower was not satisfied with the frequency of battery failures. The AGM batteries also tended to degrade rapidly, which impaired acceleration and hill climbing boost while diminishing the efficiency of regenerative braking systems. ePower believes the long-string behavior and high dynamic charge acceptance of Axion's PbC battery will overcome both of these challenges.

The PbC batteries were delivered to ePower in mid-November and installed in swappable battery boxes that will give ePower the ability to switch back and forth between the old AGM batteries and the new PbC batteries in a couple of hours. During the first week in December, ePower plans to conduct a series of benchmarking tests to compare the on-road performance of the two battery systems in the same vehicle. It will then devote the rest of December to a road show for potential customers. In early January, the first PbC powered truck will be delivered to ePower's customer and a second AGM powered truck will be brought back into the shop for a PbC upgrade.

I believe the ePower system is intriguing for several reasons. Firstly, it's a frontal assault on fuel costs, the biggest expense burden in the trucking industry. Secondly, ePower's initial marketing efforts are directed at medium to large fleet operators who are more inclined to assume the risk of testing an idea in real world conditions instead of devoting years to laboratory work. Thirdly, the ePower system is an extremely efficient use of batteries. Finally, it doesn't take much market penetration in a million-unit fleet to represent a substantial revenue base for ePower and Axion.

If results from ePower's prototype demonstrations are favorable, there is a significant likelihood that several large freight operators will purchase multiple retrofits for similar testing programs to determine where series hybrid electric drivetrain would fit into their operations. ePower's series hybrid electric drive system is not a silver bullet solution for all truckers and all routes, but the economics can be very compelling for firms with established routes and schedules where a series hybrid electric drivetrain can do the required work at a lower cost.

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

Disclosure: I am long [[AXPW.OB]].

• A conventional AGM battery;
• An AGM battery in parallel with a 200 Farad supercapacitor;
• An AGM battery in parallel with a 2400 Farad supercapacitor;
• A Furukawa Ultrabattery.

By the time they adjusted for voltage differences between the battery and the supercapacitors, the 200 Farad supercapacitor pack resulted in a 33 Farad system and the 2400 Farad supercapacitor pack resulted in a 416 Farad system.

Since the researchers used a BoostCap from Maxwell Technologies (MXWL) for their 2400 Farad parallel string, the following impedence chart compares the Ultrabattery with the technical equivalent of the Maxwell-Continental system.

(click to enlarge)

As I understand the impedence chart, the Ultrabattery performed better than the AGM battery or the AGM battery in parallel with a small supercapacitor module, but worse than the AGM battery in parallel with a large supercapacitor module.

I was surprised because I always assumed that the Ultrabattery would have more capacitance than the Continental-Maxwell system and that doesn't seem to be the case.

While Axion Power International (AXPW.OB) doesn't typically talk about the capacitance of the PbC, the unsuccessful DOE grant application it filed jointly with GM in February 2011 did note, "The large capacitance of the PbC battery (13,000F) means that it can support typical vehicle loads for up to 600s above 12V without the need for charging."

I'm sure there's more in the University of Sheffield presentation than I've been able to glean for myself, but it should give some of our more knowledgeable friends something to chew on.

Disclosure: I am long [[AXPW.OB]].

(click to enlarge)

It's unclear whether the batteries were flooded or AGM, but Banner is a primary battery supplier to both BMW and Audi so my best bet is that he was comparing AGM rather than flooded batteries.

The thing I like most about this graph is that it clearly shows while carbon additives do improve DCA, the gains are unimpressive when you remember that automakers want DCA in the 100+ Amp range for today's heavy micro-hybrids and even higher for their next generation micro-hybrids. It is worth noting that Exide Technologies (XIDE) has hitched its wagon to the carbon additives star.

In his presentation, Enders Dickenson of Axion Power International (AXPW.OB) reminded delegates that the 2010 test results from Axion and BMW showed stable DCA of 100 Amps through 50,000 cycles. He also reported that Axion's current dual battery system demonstration is using a charge current of 150 Amps for a dual battery system where 91% of the charging load is absorbed by the PbC and 9% is absorbed by the flooded starter battery.

If the PbC performance lines were superimposed on the Maleschitz graph, the old values would be near the bottom of my first paragraph and the new values would be near the top of my first paragraph.

Disclosure: I am long [[AXPW.OB]].