Hybrid, Electric, Plug-in, Clean Diesel Cars for the Responsible Consumers.

Design Considerations For Good Battery Pack Design & Integration

By • May 26th, 2009 • Category: Battery Technology for Hybrid and Electric Cars

Battery pack technology is all about managing the temperature of the cells and low resistance of the interconnecting conductors, connectors, and switches. Battery pack technology is not the same as battery chemistry technology. The latter is about building functional individual battery units, while the former is about making those units work together effectively and safely to provide much higher capacities.

There is still a lot of room for creative pack designs in electric and hybrid-electric vehicles with different form, fit and function topologies and connection schemes. The most common requirement is the need for temperature management for high power performance and operational life of a pack. High current charging and discharging through parasitic resistances in the cells, connections, switches, and wire cables generate heat that must be dissipated without damaging the surrounding materials.

Here are some key factors governing good battery pack design.

Internal Heating Factor

Knowing where the heat originates is the first step in designing a cooling system. Just like ultracapacitors, cylindrical cells are rolled up like a jelly roll with an outside cover and dissipate the most heat through the electrode connections in the center at either end of the cylinder. Again just like ultracapacitors, prismatic cells are in the form of flat sheets and dissipate the most heat through the sides of the cell. Cells in the middle of the pack and at the end of the circulating air or coolant stream will be the hottest and deserve the most attention. The most stress on a cell occurs at high temperature and high voltage such as immediately after braking regeneration charging while going down a hill.

Quick, low cost, low resistance connections between cells are a challenge. One bad or corroded high resistance connection can create enough heat to destroy an entire pack very quickly during high current charging or discharging operations. Even if the cell chemistry can accept high current “quick” charging and discharging, resistive heating losses in the connections and wires can easily drop the stored energy efficiency to 50% from the 80%-90% efficiency of more moderate power operations.

MES-DEA Zebra Z5 battery pack
MES-DEA Zebra Z5 battery pack Courtesy MES-DEA.

The challenge is to keep all the cells at a uniform cool temperature to prevent uneven cell aging and premature pack failure. It would be clever if the cooling system would start early in anticipation of high current downhill charging and startup accelerations. Today’s position locations systems make this a real possibility.

Low Ambient Temperature Factor

Keeping a pack warm is generally not a problem except during cold weather startups. Some batteries lose a significant amount of capacity at extremely low temperatures. Using part of the stored energy to power a heating blanket could work nicely prior to startup.

The “Zebra” Nickel Sodium Chloride (NiNaCl) battery requires an internal temperature of 300 °C (572 °F) to keep the NaCl electrolyte melted. Using the stored energy of the battery pack and excellent insulation, the temperature can be maintained over a number of days from the stored energy alone. The downside is that the battery must be preheated before use. Prematurely charging a cold Zebra will destroy an expensive pack.

Lithium Ion Factor

Lithium ion (Li ion) titanate is a unique battery chemistry that is endothermic (cools by absorbing heat from the environment) during moderate to low operational power currents. However, at high currents and with corroding connections over time a Li ion titanate pack has the same heat dissipation problems.

Tesla Roadster battery system - Courtesy
Tesla Roadster battery system – Courtesy

The Tesla electric sports car battery pack uses an older Li ion battery chemistry that is more susceptible to thermal runaway. Even when passively sitting fully charged in a garage the Tesla pack uses the equivalent of two refrigerators power to continuously keep the pack cool.

Electromagnetic Radiation Factor

Often ignored, one final consideration in pack design is the high magnetic field surrounding the high current carrying conductors. The high current DC power of the battery leads to an inverter controller for control of the AC induction motor used in most electric vehicles. The high current AC coming out of the inverter creates enough electromagnetic radiation to drown out any nearby AM radio, so good shielding practices have to be followed in the wire and cable installation. Alternatively, this problem can be ignored if functioning AM radio reception is not a priority.

The Bottom Line

Very few battery pack manufacturers have successfully integrated all the pack thermal and electrical requirements into a mechanical structure that has to withstand the shock and vibration environment of transportation applications. With at least five new plug-in battery cars coming on the market within the next two years, resulting in thousands of electric cars driving on our roads, let’s hope the manufacturers are successful with their battery pack integration.

Be Sociable, Share!
Tagged as: , , , ,

is an industry veteran with 30 years of experience in general business, marketing, project and product management, and engineering research and development. Mr. Bartley provided executive management support including technical and business oversight to heavy-duty hybrid-electric prototype projects as they evolved into production. He developed cost models for energy storage and fuel savings, and power models for ultracapacitor packs. Mr. Bartley is well known throughout the industry of heavy-duty hybrid-electric buses and trucks, having delivered many papers and presentations since 2003. Mr. Bartley maintains a blog at Follow
Email this author | All posts by

10 Responses »

  1. To the best of my knowledge, no one has yet produced a battery pack that provides all of the qualities Mr Bartley suggests are desirable (even necessary). However, this commenter has an invention that comes close. A patent has been applied for and is presently pending. A limited amount of prototypeing and testing has been completed. Further work is required. The inventor is seeking a partnership with an entity interested in advancing a proprietary design that could be highly profitable and quickly brought to a welcoming market. If any party has such an interest, I can be contacted via email (

  2. I am more interested in advancing the technology through the synergy and collaboration of open applied research rather than unknown proprietary intellectual property.
    Tom Bartley

  3. E.A.Baker
    A great amount of research has been and is being conducted on cell and battery pack design. Most of the work that I am aware of is being carried on by “for profit enterprises”. I assume that many if not all of them hope to profit from their investment. My purpose in applying for a patent is simply the best way I know of to protect the methodology described in the application from being advanced as a proprietary product of someone elses work. The primary product of my research pertains generally to the ampacity considerations or lack there of that can and do cause excessive heating and the wasting of the stored energy. You cite the “Tesla Battery” as an example. Therefore my origional comment.

  4. Dear E.A. Baker,
    If you really want “to protect the methodology described in the application from being advanced as a proprietary product of someone elses work,” just publish it for peer review. I don’r know what you mean by “the ampacity considerations or lack there of that can and do cause excessive heating and the wasting of the stored energy.” If you are referring to high currents for charging and discharging, Ohm’s Law, I**2 R, still holds where I is the current and R is the equivalent series resistance. The result is heat and loss of energy. if you haven’t looked already, I refer you to the Lithium titanate battery chemistry that, in some points of operation actually absorbs heat. The real trick in pack designs is to consider all sources of heat, including new and aged or corroded interconnections, and reject that heat before it can raise the temperature enough to damage the pack. Ofcourse, storage efficiency is enhanced if you don’t generate the heat in the first place.

  5. Dear Mr. Bartley,
    When I commented on your origional post, it was not my intention to open a debate. I concur that heat is a primary problem in large scale/high capacity battery packs. I also concur that inadequate conductors/connectors are generally the cause of these problems. A conductor’s current carrying capacity is governed by two different qualities, the resistivity of the material and the size of the material (cross sectional area) generally collectively referred to as “ampacity”. I can only presume that Tesla’s battery pack design has the 69_18650 cells (brick) connected in parallel using light metal strips welded to the terminals of each individual cell to produce a supercell. 99 of these supercells are then connected in series to produce a high voltage battery pack.
    According to Kirchhof’s law of current, the current is the same everywhere in a series circuit, thus the current levels in the Tesla battery would calculate to be about 150 amps. Far to high for the metal strips typically used in resistance welded connections.

  6. Dear E. A. Baker,
    While not wanting to be confrontational, I believe public debate is good to increase interest and spread information.
    Ah yes, the Tesla battery – the devil is in the details. If I remember correctly the pack voltage is about 340 volts. Doubling the voltage would half the required current for the same power and even if the equivalent series resistance (ESR) doubled the net heat loss would be half because of I**2 and the battery pack efficiency would double. Why didn’t Tesla do this and why can’t two of their packs be put in series to double the voltage? i believe the answer is in how the grounds are handled to reject the waste heat that is generated boith during charging and discharging.
    Last time I looked there were over a dozen patents in this area and even more if you look at the required balancing and equalization.

  7. Dear Mr. Bartley,
    Again, I agree with you, debate is good. As a matter of fact, our dialog has led me to reexamine my investigations of the smalll cell/large battery concept. I have recently learned more about the “state” of battery pack and cell design, causing
    me to question further the heat “problem” apparently prevelant in all EV battteries. All battery manufacturers and/or EV
    manufacturers have the same goals (low battery pack weight, high stored energy capacity. Tesla is a good case to examine. They have done more to further the small cell/large battery concept than any other single entity. They have more real experience than any other entity. There is more literature available on the Tesla battery/vehicle than all other projects combined. All making their design a good candidate for investigation.
    There are 2 possible sources of excessive heat, the conductors or connectors and the individual cells themselves. The
    conductors were discussed previously. Is it reasonable to presume that the high current levels in the Tesla battery (or any other battery made up of many small cells) simply cannot be tolerated by the 18650(or other small) cells employed in the design? The question seems to merit consideration. I don’t know the answer.
    By the way my name is Ed.

  8. Hi Ed,
    Thanks for the response. Individual cell balancing and equalization is a major challenge for any transportation pack design, especially one that incorporates the frequent rapid charging from braking regeneration. The massive parallelism in the Tesla pack is one way to do this while adding some level of failure tolerance. Individual cell manufacturing and aging differences, even failures, can be averaged out in the parallel connections of the layers. Also, the many parallel connections creates many parallel current paths, thus reducing the current in each path and minmizing the effects of resistance in the connections. This approach is typically rejected because of the assembly cost of making all those connections.

    The other design considerations are structural integrity and cooling. The structure has to support each cell in an automotive shock and vibration environment with out affecting the electrical connections or cooling system. The cooling strategy has to consider hot spots like the middle of the pack and the downwind edge of the cooling air flow. It is also useful to know that cylindrical cells dissipate most of the heat through the ends and prismatic (flat) cells dissipate most of the heat through the sides. Another curious effect is that individual cells expand at certain states of charge enough to destroy the cell unless the packaging structure keeps it compressed.

    So, it “ain’t” quite as simple as it first appears.

    Getting back to the Tesla, rumor has some owners complaining about the cost of haviing to run the battery pack cooling system after the car is fully charged sitting quietly in the garage. It’s like adding two refrigerators in the garage. More rumors express investors’ concerns about the thermal runaway potential of that early Lithuim ion chemistry. The Tesla reps I’ve talked with plead ignorance.

    Addressing battery pack current levels, most packs are designed for 200 Amps or less with occassional spikes up to 300 Amps. Looking at the available cables and connectors, this seems reasonable.

    Other interesting topics are the Magna Steyr/A123 pack for the Daimler hybrid; and the cell parameters changes with temperature and charge/discharge “C” rate.

    Your turn,

  9. Tom,
    Back to ampacity. My question regards the ability of a single small cell to conduct large currents. If we think of the cell simply as a conducting element (with a resistance) in a complex circuit, we must then concern ourselves with the current that must flow thru that conductor (cell). My understanding of the”law of current” is that the current in a series (complex) circuit is the same everywhere in the circuit. Thus, it seems to me, that a battery pack along with it’s load (a
    circuit) requires that all of the current in the circuit must flow thru all of the conductors in that circuit. While parallel configurations divide resistances, they do not divide current.
    If my interpretation of theory is incorrect, then my conclusions are probably not correct. I’d appreciate your opinion.

  10. Hi Ed,
    Parallel paths divide the current, I, according to the resistance, R, in each path according to I=V/R. The voltage, V, is a constant for each path.