TMCnet News

Performance Gains in Lithium Battery Charging
[October 07, 2004]

Performance Gains in Lithium Battery Charging

2500 Eisenhower Avenue, PO Box 890, Valley Forge, PA 19482-0890, 610-676-0188, FAX 610-676-0189, www.galaxypower.com, E-mail: [email protected]
Galaxy Power, Inc.
2500 Eisenhower Avenue. PO Box 890. Valley Forge, PA 19482-0890. (610)676-0188. FAX (610)676-0189
Performance Gains in Lithium Battery Charging
Technology using the Galaxy Power
QuickLithium™ Charging Algorithm
Lithium Battery Technology
Lithium-based (Lithium-Ion and Lithium-Polymer) batteries have become the de facto standard for battery powered portable
electronics equipment. Currently, devices such as Laptop PCs, Cell Phones, and PDAs utilize this battery chemistry almost
exclusively. Lithium-based battery technology possesses many technological advantages when used in these applications.
Lithium-based batteries offer the highest volumetric efficiency and mass efficiency of any of the currently available battery
chemistries. For a given weight and volume, they are capable of storing more electrical energy than any of the other readily
available cell chemistries. In addition, Lithium-based batteries have a low rate of self-discharge; a Lithium-based battery
retains its charge when stored far longer than other battery technologies. Although Lithium-based batteries are more costly than
Nickel-Cadmium or Nickel-Metal-Hydride batteries, they are not prohibitively expensive for use in high-end products. As
Lithium-based technology advances, expect to see this battery technology migrate into many other applications.
Charging Lithium-based batteries requires different technologies from those used for other cell chemistries. Attempting to
charge Lithium-based batteries beyond 4.2 Volts per cell permanently damages the battery and may result in the battery
exploding. Over discharging Lithium-based batteries also results in irreversible changes to the batteries. For this reason,
manufacturers incorporate semiconductor-based protection circuitry in the Lithium battery pack to prevent this from
happening. In addition, Lithium-based batteries must be limited in the amount of current they supply; significant over-current
discharge can cause this type of battery to explode. Manufacturers incorporate a PTC fuse device within the battery pack to
limit current. A PTC fuse is actually a thermistor type device that exhibits a relatively low resistance at its rated current, but
increases resistance exponentially if this current is exceeded.
Charging Methods
Contemporary charging methods utilized for charging Lithium-based batteries charge the batteries at a constant current,
typically at a 1C to 1 ½ C charging rate, therefore, a 1 Ampere/hour battery would be charged at 1 to 1 ½ Amp rate. When a
cell voltage of 4.2 Volts is reached, the current is decreased to maintain this voltage constant and current is monitored until it
reaches 1/10th of its initial value. At this point, the battery is considered to be fully charged. If a Lithium-based battery is
discharged to its low voltage limit, then fully charged in this manner, it takes 2 ½ to 3 ½ hours to be fully charged.
Compared to Nickel-Cadmium (Ni-Cd) or Nickel-Metal-Hydride (Ni-MH) batteries, this is a long charging time. Ni-MH
batteries can be charged in approximately 1 hour. The limitation on charging time for Ni-MH batteries is a result of their
exothermic charging characteristics. Attempts to charge them at a faster rate results in excessive heating of the battery pack
which is detrimental to battery life. Ni-Cd batteries can be charged at a much faster rate as they possess an endothermic
charging characteristic. Assuming that the charge is terminated before they enter an overcharge region, they can be easily
charged in 15 minutes without adverse effects.
This relatively long charging cycle compared to that of other battery chemistries is one of the less desirable characteristics of
Lithium-based batteries. Another undesirable characteristic is “capacity fade”. Each time a Lithium-based battery is charged it
loses a little of its storage capacity. All battery chemistries exhibit some capacity fade, however, Lithium-based batteries fade
more rapidly than Ni-Cd orNi-MH batteries. The battery loses a little capacity each time it is charged. Cell resistance increases
with each charge cycle until the battery is no longer usable. It appears as if the act of charging the battery damages it, almost as
if the size of the battery plates decreases with each charging cycle.
2500 Eisenhower Avenue, PO Box 890, Valley Forge, PA 19482-0890, 610-676-0188, FAX 610-676-0189, www.galaxypower.com, E-mail: [email protected]
Galaxy Power, Inc.
2500 Eisenhower Avenue. PO Box 890. Valley Forge, PA 19482-0890. (610)676-0188. FAX (610)676-0189
If we look at an electrical model of a Lithium-based battery pack we can see some of the reasons for these problems.
Terminal and
Lead Resistance
Terminal and
Lead Resistance
Overcurrent
Protection
Cell Protection
Device Internal Cell
Resistance
Ideal
Lithium- Ion
Cell
PTC Fuse
Lithium-Ion Battery Model
As can be seen from this schematic, there are a number of resistances in series with the battery. The Lithium-based battery, as
we already have stated is a voltage sensitive device. All of these external resistances along with the battery’s own internal cell
resistance make it difficult to actually measure battery voltage and to control the charging cycle.
Many of these resistances are a result of the protection devices required when using a Lithium-based battery. However,
terminal and lead resistance, and cell resistance are present with any type of battery. In the case of Ni-Cd or Ni-MH batteries,
these resistances are unimportant as we typically charge with a constant current source until the charge cycle is terminated.
With Lithium-based batteries we cannot do this due to the limitation on maximum cell voltage.
Resistance Source Magnitude Voltage Drop @ 2.5Amps
Terminal and lead resistance .0536? .134 Volts
PTC Fuse resistance .0372? .093 Volts
Protection Device resistance .042? .105 Volts
Cell resistance Varies ˜ .036? .09 Volts
TOTAL .1688? .422 Volts
As can be seen from this data, if this battery was charged at a constant current until the voltage at the charger measured 4.2
Volts, then decrease current to hold 4.2 Volts AT THE CHARGER, the constant voltage phase of charging would start nearly
½ Volt too early. This would contribute to a long charge cycle. This would be worse if cell resistance was higher.
The QuickLithium™ Charging Algorithm
The Galaxy Power, Inc. QuickLithium™ charging algorithm utilizes several techniques to minimize charging time.
QuickLithium™ measures battery voltage at a time when no current is flowing into or out of the battery. This quiet time
sensing was pioneered in our patented QuickSaver® charging method. It permits us to measure the actual cell voltage without
the inaccuracies introduced by all of the series resistances. We also measure the current flowing into the battery and the voltage
at the output of the charger during charge. With this information we can calculate the actual voltage drop across these series
resistances and compensate for it. As a result, QuickLithium™ delays the onset of the constant voltage phase of charging until
the actual battery voltage reaches 4.2 Volts/cell. As cell resistance is dynamically changing during the charging cycle, we
recompute this once per second.
Note:
All resistances measured on a BYD Type
LP103463S Lithium-Ion Battery (1700mAh
capacity). Cell resistance was measured when the
battery reached 4.2 Volts, and will vary considerably
depending on the state of charge of the battery, and
the number of times it has been charged.
2500 Eisenhower Avenue, PO Box 890, Valley Forge, PA 19482-0890, 610-676-0188, FAX 610-676-0189, www.galaxypower.com, E-mail: [email protected]
Galaxy Power, Inc.
2500 Eisenhower Avenue. PO Box 890. Valley Forge, PA 19482-0890. (610)676-0188. FAX (610)676-0189
QuickLithium™ Charging Algorithm Flow Chart
Start
Load Preset
Period & Duty
Cycle.
Soft Start
Measure Battery
Voltage Under
No Charge
Condition (Vbo)
Is
Voltage =
4.2 Volts
Cell?
Yes
No
Measure
Voltage (Vbc)
& Current
Under Charge
(Ichg)
Subtract Vbo
From (Vbc)
=To Vr
Is
Vbc – Vr
= 4.2 Volts
Cell?
Yes
No
Is
Ichg > Preset
Value?
Yes
No
Increase
Charging
Current
Decrease
Charging
Current
Decrease
Charging
Current
Measure Vbc
& Vbo &
I Under Charge
Is
Vbc – Vr
>4.2 Volts
Cell ?
Yes
No
Increase
Charging
Current
Is
Ichg= Preset
End Value
?
No
Yes
End Fast
Charge
Go To Topping
2500 Eisenhower Avenue, PO Box 890, Valley Forge, PA 19482-0890, 610-676-0188, FAX 610-676-0189, www.galaxypower.com, E-mail: [email protected]
Galaxy Power, Inc.
2500 Eisenhower Avenue. PO Box 890. Valley Forge, PA 19482-0890. (610)676-0188. FAX (610)676-0189
The TI MSP430 microcontroller based development board used to develop the QuickLithium™ Charging Algorithm
SmartPulse™
To minimize the effects of cell resistance, we borrow another technique from our QuickSaver® technology. Once per second,
we introduce a short (5 Millisecond) but deep (2 ½ C) discharge pulse to the battery. This technique which Galaxy Power, Inc.
has trademarked as SmartPulse™ has the beneficial effect of redistributing the charge more uniformly over the surface of the
battery plates, preventing the formation of bubbles, and minimizing cell resistance. The plates of the battery are not a flat sheet
of metal, rather, they are highly irregular to maximize their surface area and keep cell resistance low. However, as charge tends
to reside on the peaks of this surface, the peaks enter an overcharge state long before the battery is fully charged. As this is an
electrolytic process, these overcharged peaks tend to evolve oxygen bubbles that shield the rest of the plates from being
charged.
The discharge pulse discharges this excessive charge at the peaks and results in the battery being more uniformly charged. As
no bubbles are present, cell resistance remains low and the battery will accept charge at a higher rate. Since Lithium-based
batteries are highly susceptible to damage from overcharge, this discharge pulse minimizes this effect and results in a much
lower rate of capacity fade.
Results
Testing was conducted on the same BYD Lithium-Ion battery that was used to measure resistances. The battery was discharged
to the low-voltage trip point of its internal protection circuitry. The battery was then charged using contemporary constantcurrent
constant-voltage charging methods at a 2 ½ Amp charging rate until it measured 4.2 Volts.
Charging was continued until the current dropped to 170 mA. The battery capacity was measured using an Alexander Batteries
Optimizer 2003 battery analyzer. Charging time using this method was 2 Hours, 20 Minutes. The battery was discharged again
to its low voltage trip point, and then recharged using the Galaxy Power, Inc. QuickLithium™ method. Charging time was
measured at 55 Minutes. The battery was again discharged using the Optimizer 2003 analyzer.
The battery when charged with the QuickLithium™ method exhibited a 5% increase in charge as compared to the constantcurrent
constant-voltage charging method.
2500 Eisenhower Avenue, PO Box 890, Valley Forge, PA 19482-0890, 610-676-0188, FAX 610-676-0189, www.galaxypower.com, E-mail: [email protected]
Galaxy Power, Inc.
2500 Eisenhower Avenue. PO Box 890. Valley Forge, PA 19482-0890. (610)676-0188. FAX (610)676-0189
Conclusion
QuickLithium™ technology appears to overcome most of the objectionable characteristics of Lithium-based batteries in realworld
applications. Charging times with the QuickLithium™ charging technique are comparable with Ni-MH, and capacity
fade is greatly minimized. QuickLithium™ should extend the utilization of Lithium-based batteries to many new applications.
Bruce Rogers
Sr. Staff Engineer
Galaxy Power, Inc.
Phone: (610)676-0188
FAX: (610)676-0189
E-Mail: [email protected]

[ Back To TMCnet.com's Homepage ]