How to achieve consistency, quality and lifetime: the lesser known challenge facing makers of electric vehicles’ batteries?
By Dr Stephen Lambert. Director of Battery Systems, Vayon Group
Much of the public debate around the future for electric vehicles (EVs) concerns range anxiety, and the infrastructure required to enable drivers to charge their vehicle conveniently when away from
home. Indeed this, combined with the high purchase price of EVs, appears to be the main reason why consumer adoption of EVs has been limited to date.
Car manufacturers themselves, however, have other concerns about EVs’ lithium battery power packs. Commercially, the effort to develop viable battery-powered EVs will have been wasted if they
never reach the mainstream. To make money out of battery-powered EVs, car makers need to increase production volumes by more than an order of magnitude compared to today’s levels. And
this means that they face a big challenge in resolving the tension between cost and quality in the EV’s battery.
Degradation over time – ageing – is an inherent property of the lithium battery cells in an EV’s battery pack. There are also marked variances in performance between one new cell and another.
But if the EV is to become a mainstream product at an affordable price, it needs to offer mainstream levels of lifetime reliability and performance. And this means that manufacturers of EV power packs have to find a way to manage the effects of ageing and variances in lithium cells while keeping the number of cells in a power pack as low as possible.
ICE: proven reliability
The internal combustion engine (ICE) is an old (it was invented in the 19 century) and very well understood technology. Moreover, it serves the purposes of car manufacturers very well. Consumers expect a car to work reliably for at least ten years, and to still have some resale value after its first ten years of operation.
The ICE helps today’s cars to meet this requirement very well: with appropriate servicing, the engine will operate almost exactly the same after ten years, in terms of power output, torque and other mechanical parameters, as it did on the day it was made. What is more, both the machined and the electronic components inside an ICE are uniform, so one engine will perform the same as any other engine produced to the same design specifications.
From the car owner’s point of view, this means that the experience of driving an ICE vehicle is more or less unchanged after ten years of ownership: by and large, the car will accelerate as fast, and pull as strongly, as it did when it was new.
Natural variation and degradation in lithium cells
Unlike an ICE, the traction unit of an EV is based on technology which ages, and which suffers from inconsistency between manufactured units. But the consumer’s expectation is for a traction unit like an ICE: predictable, and consistent over a very long lifetime.
And so the development of battery-powered EVs is forcing car makers to learn how to extract the unchanging, uniform lifetime performance of an ICE from a battery technology that is non-uniform
and that deteriorates over time.
This inconsistency is an inherent characteristic of the lithium cell manufacturing process. Cell production is subject to a trade-off between quality and consistency on the one hand, and unit cost
on the other. General industrial and consumer market demand for standard 18650 rechargeable lithium cells calls for a reasonable level of quality at an ultra-low cost. (The 18650 designation
applies to cylindrical cells in a standard form factor which is 18mm in diameter and 650mm long.)
The problem for developers of large battery packs – and an EV battery can contain thousands of standard cells – is that across a large population of new cells, a ‘reasonable’ level of quality means that the variation in parameters such as output voltage and impedance will be wide.
The ageing process extends the variances over time. With every charge and discharge cycle, a cell’s impedance changes. And the rate of change in impedance is slightly different from cell to cell. And so the optimal charging profile and discharging limits are also different from cell to cell, and constantly change over time.
Without electronics systems to compensate for the variances, this would lead to inconsistency in performance: two identical EVs from the same production line would behave differently when new, and the owner would notice the reduced range and peak power output of a ten year old car compared to the same vehicle when new.
Developing more sophisticated battery management systems
Part of the function of the battery management system (BMS) inside an EV’s power pack is to reduce the effects of inconsistency and ageing. This calls for precise and accurate monitoring of the state of health of the battery. Today, the limits to the BMS’s performance are largely defined by:
- the accuracy, precision and sampling frequency of the current, voltage and temperature measurements of the battery pack as a whole, and of the cell modules which the pack contains
- the software controls, and the models of battery operation underlying them, which
continually modify the charging profiles and discharging behaviour of the battery over its
Both these functions of the BMS, the measurement and the control, can be improved. And they must be.
Today’s EVs, such as the Nissan Leaf and BMW i3, are superbly engineered cars, but as niche, premium vehicles they have been able to take a highly conservative approach to the battery pack
design, using a high number of cells which are driven relatively lightly. This gives the BMS an extended capability to compensate for ageing by driving the cells harder towards the end of their life than when new, without any risk of over-stressing the battery, which might result in a premature battery failure.
To bring EVs into the mainstream for high-volume production, cost-optimisation will be essential: a mainstream EV’s power pack will need the absolute minimum number of cells required to deliver the specified performance on the road. A conservative, light-touch BMS policy will not be permissible.
This means the vehicle will be required to squeeze every mAh of capacity and Watt of power out of every cell – and this calls for an extraordinarily sophisticated BMS which can characterise the
battery’s state of health and state of charge with extreme accuracy, and which has the intelligence to use this data to optimise the management of every single cell.
The combined efforts of the car manufacturers, of semiconductor suppliers and of automotive electric power system manufacturers such as Vayon Group will reach this goal – but how soon? All
we know today is that it is going to take the application of intensive engineering effort before battery-powered EVs are ready for the mainstream.