The invention relates to an electrical circuit for a parallel array of at least two charge accumulators, particularly batteries, storage batteries, etc., for the voltage supply of a load and/or for the electrical charging of the charge accumulators, with connection terminals at the charge accumulator end and connection terminals at the load or charging ends and with at least one DC converter. The invention also relates to a voltage supply device for a load, which incorporates this type of circuit. The invention further relates to a charging device that incorporates this sort of circuit.
Technical advances mean that in the coming years it will be possible to develop electric vehicles offering attractive driving performance and to produce them in greater numbers. For the foreseeable future, the most expensive individual component of a vehicle of this type will be the battery. A vehicle manufacturer can optimise costs in this respect by using the same battery for all vehicle categories supplied. Small vehicles will only be fitted with one battery; larger vehicles with two or more. This will enable greater unit numbers to be achieved and the cost of testing and certifying the battery will only be incurred once for all vehicles.
A rated voltage of 300-400 V has proved optimal in cars and light-duty commercial vehicles. It stands to reason, therefore, that the battery should be designed with this rated voltage and in larger vehicles several such batteries should be connected in parallel. In this way, the voltage remains the same and only the capacity is multiplied according to the number of batteries. Tests have revealed that there is not always an even distribution of load for two or more batteries connected in parallel. Particularly where batteries have different thermal conditions, their capacities and currents may diverge. The current voltages or states of charge of the individual batteries alter irregularly as a result. The different loads applied to the batteries may lead to premature ageing and thereby compromise the vehicle's reliability. Similar problems occur when charging the batteries. This may lead to individual batteries being charged unevenly or even being overcharged, which causes the life of the battery to be significantly reduced.
There are state-of-the-art circuits for various battery arrays, which are designed to increase the life of batteries or prevent them from being destroyed. The known solutions will be examined more closely below; most of them relate to a circuit for balancing the different voltages or states of charge of the batteries concerned.
U.S. Pat. No. 6,353,304B1 discloses a system for charging and discharging batteries connected in parallel. The charging involves using an AC generator and an AC-DC converter. In a first phase, both batteries A, B are connected straight to the AC-DC converter and uniformly charged. Even before the two batteries are fully charged, the system switches to phase two.
This involves one of the batteries B being connected to the load, while the other battery A is simultaneously connected to the battery B via the DC-DC converter. Battery A is now fully charged by battery B via a step-up device. Battery B is discharged as a result. It is only when battery A is fully charged that it is connected to the load alongside battery B (phase 3). The same occurs in the following charging cycle, except that rather than battery A, battery B is now fully charged. This system should, in particular, increase the life of the batteries.
However, this reference does not envisage any measures for preventing or balancing a differing degree of load applied to the batteries in phase 3, which means that there is no optimum working load. The DC-DC converter also always processes the entire battery voltage in phase 2 and must be of correspondingly large dimensions.
AT 505169 B1 discloses a redox flow battery with two parallel-connected functional units. Each functional unit may be disconnected by a switch. In addition, a device, e.g. a DC-DC converter, is provided, in order to connect a disconnected functional unit, e.g. a battery, to the still active sub-system (another battery). With the DC-DC converter, the residual energy of the disconnected battery may be used to charge the other, still active battery. In this way, the self-discharge of the disconnected battery along with the negative side-effects can be prevented. This reference contains no teaching in relation to the problem of balanced utilization with simultaneous connection to the load.
U.S. Pat. No. 6,774,606B1 relates to the voltage supply, particularly in the case of satellites, and shows an array of batteries connected in parallel, which each form individual strings. The strings are each connected to a charge balancing unit. This reference actually relates to the parallel connection of individual batteries of different strings. No details are disclosed on the inner life of the charge balancing unit.
The following references are further removed from the invention, as they relate to series-connected batteries.
U.S. Pat. No. 5,479,083A discloses a charging device for multiple batteries. In a basic circuit, an inductor branches off between two series-connected batteries. Two switches alternately operated by an oscillator (e.g. at a frequency of 200 kHz and with a 50% frequency ratio) ensure that the batteries are evenly charged. In this case, the inductor should prevent power losses as a non-dissipative shunt. The flow through the inductor undergoes a complete reversal when the switches are changed, at least when the batteries are almost identically charged, which means on the one hand that throttles of the appropriate size are required and, on the other, relatively high losses occur.
U.S. Pat. No. 5,631,534A discloses a battery balancer in the form of a current pump module, which displays great similarities to the previous reference. In order to transfer charge from the first to the second battery, a first switch is operated by means of a PWM signal, while the second switch is left open. When the first switch is closed, the current in the inductor increases, when the first switch is open, it supplies the second battery via the second diode. The whole process also operates in the same way in reverse.
U.S. Pat. No. 7,049,791 B2 discloses a battery balancer circuit for the uniform charging of multiple series-connected batteries. The switches, e.g. MOSFETs, are operated alternately, so that the state of charge of both batteries can be balanced via the charge bypasses with the inductors and capacitors and overcharging can be prevented.
US2009/0278489 discloses a system and a method that dynamically equalises battery voltages. As with the previous references, the batteries are series-connected in this case and the remaining features also strongly resemble the preceding references.
U.S. Pat. No. 6,150,795A also relates to the problem of battery charge equalisation and pursues in circuitry terms a similar or related basic principle as the previous reference. The batteries are once again series-connected.
In the case of the voltage supply of a load with a parallel battery pack, there are also circuits that counter the problems of uneven utilization described above by having the energy flow to and from each battery passing through its own bidirectional DC-DC converter. In other words, each individual battery is separated from the high-voltage BUS by a DC-DC-converter. In this case, all DC-DC converters are parallel-connected to the high voltage BUS at the output end. The voltage to the load may be freely selected in this case within the framework of the DC-DC operating range, irrespective of the battery voltages. However, the disadvantages are that the converters must be correspondingly powerful and therefore occupy a lot of space, because they each convert the entire battery voltage, that high power losses occur at the converters as a result of this and that the provision of its own DC-DC converter for each individual battery requires a high level of expenditure. All these requirements have a negative impact on the weight, space requirement and cost of this sort of circuit.
One of the goals of the invention is to create a circuit to balance out the state of charge of at least two batteries for a voltage supply device or a charging device, which is unencumbered by these disadvantages and in which the losses due to voltage conversion, in particular, are minimised and which is characterised by a simple design. Moreover, the weight, spatial requirement and cost of this sort of circuit should be significantly lower than the state of the art.
The invention achieves these goals and others with a circuit of the type mentioned above, in that the DC converter connects one of the connection terminals at the load or charge end to the corresponding matching connection terminals at the charge accumulator end, which are each assigned to matching terminals of the charge accumulator or combined charge accumulators, wherein corresponding terminals at the charge accumulator end can be connected by switches as active elements of the DC converter alternately or overlapping in time via at least one common passive element of the DC converter with the corresponding matching connection at the load or charging end.
This means that basically the highest differential voltage applied between matching connection terminals at the charge accumulator end or a voltage derived from this differential voltage is converted at most in the DC converter.
Where there are only two matching connection terminals at the charge accumulator end, the differential voltage applied between matching connection terminals at the charge accumulator end or a voltage derived from this differential voltage is essentially converted in the DC converter.
The invention relates to a method of balancing the states of charge of at least two parallel-connected charge accumulators, particularly batteries, storage batteries, etc, during the voltage supply of a load by the charge accumulator or during the charging of the charge accumulator, wherein at least one DC converter is used, characterised in that with switches as active elements of the DC converter the matching terminals of the charge accumulators are applied alternately with the combined matching connection terminal of the charge accumulator chain formed from the at least two charge accumulators.
The new approach is based on the consideration that the voltage difference of the two batteries is usually very small and amounts to only a few percent of a battery's total voltage. This means that it is advantageous to use a converter in such a way that only this differential voltage occurs as the converter's “operating voltage”. This means that the power processed in the converter is significantly smaller than in the circuits described above, even if the converter has to process the full battery current.
The balancing of the state of charge of the batteries therefore takes place during the voltage supply of a load or during the charging of the charge accumulator. The circuit according to the invention switches alternately from one battery to the other, so that in a first switch state the battery voltage of the first battery at the converter becomes effective and in a second switch state the battery voltage of the second battery at the converter becomes effective. The DC proportions remain unaffected by the converter's mode of operation and are allowed through unchanged and applied to the output (hereafter referred to as “load-side connection terminals” in the case of a voltage supply and “charge-side connection terminals” in the case of a charging device). Only voltage changes caused by the alternately operated switches, which only occur on the scale of the voltage difference between the batteries, are changed in the converter.
A DC converter comprises active elements—switches, also referred to as choppers—which generate a time-periodic voltage curve from a DC voltage (they “chop” the voltage in a particular period) and passive elements, which again generate a DC voltage from the voltage curve changing in time. This is a sort of low-pass filter or smoothing circuit. In other words, only that part of the voltage that changes in time is converted, whereas the DC voltage part that is constant in time passes the converter unimpeded. The latter portion only causes ohmic losses, while the ripple voltage results in power losses, particularly due to the constant current change in the throttle. In the present case, the DC-DC converter does not therefore convert the total voltage of the individual batteries, but only their differential voltage. This means that it can be executed as a significantly smaller, more cost-effective design. Since it is essentially only the differential voltage or a voltage diverted from this that is converted, the power losses in the converter also remain small.
The notion of “voltage derived from the differential voltage” relates to variants that are doubtless conceivable but are less preferred, in which voltage dividers and/or other converter stages are interposed, because this would lead to additional losses in these elements. Where there are three or more batteries, the differential voltage of each battery compared with the load voltage is crucial; this means that there are several differential voltages, e.g. in the case of n batteries connected to the negative terminal, n*(n−1)/2 differential voltages can be measured between the positive terminals.
In the following, the circuit is described in relation to a voltage supply for a load. The aim here is for the batteries that contribute to the voltage supply to be loaded depending on their respective states of charge. The battery with the higher voltage receives a correspondingly greater load than the battery with the lower voltage. It is of course conceivable, however, that the same principle will also apply to a charging device that charges two or more charge accumulators. Conversely, the aim here is for the batteries to be charged evenly and for the charge current—likewise depending on the state of charge of the individual batteries—to be divided proportionately between the individual charge accumulators. In principle, the same circuit may, for example, be integrated in an electric vehicle, used both as a voltage supply and also for charging.
The term “connection terminal” not only refers to a physical connection within the meaning of a reciprocal plug connection or the like, but also covers, apart from the meaning as defined by connection, the meanings of input, output, terminal, line, connection, etc.
The invention is explained in greater detail below by means of the drawings. In these:
On the basis of
The principle disadvantage of the circuits shown in
The heart of the circuit 10 is a DC converter 11, also referred to as a DC-DC converter or chopper. It has two connection terminals 3a, 4a at the charge accumulator end, which are respectively connected to matching polarity terminals (in the present case, negative terminals) on the charge accumulators 1, 2, and also one connection terminal 5a at the high-voltage bus 5 end, hereinafter referred to as the load end.
The opposing terminals, in relation to polarity, on the charge accumulators 1, 2 (the positive terminals in the present case) are connected to the connection terminals 3b and 4b in the circuit 10. The connection terminals 3b and 4b are connected straight to the other terminal 5b for the HV bus 5.
So whereas the connections 3b, 4b with the same polarity are connected straight to the HV-bus 5, connections 3a, 4a with opposing polarity are connected to the HV-bus 5 through the insertion of a DC converter 11.
Each DC converter is essentially made up of active elements, which modulate an input voltage, and passive elements, which generate a DC voltage as the output voltage from the modulated voltage. The aim here is for the output voltage to have a different value to the input voltage. As can be seen from
The switches S1, S2 are operated across a controller (not shown in
It is expressly pointed out that the switches S1 and S2 are an integral or functional element of the DC-DC converter and, as its active elements, bring about modulation of the voltage, which is processed in the following passive elements of the converter and then applied to the outgoing terminal 5a. The bipolar switches can be operated by a controller and designed to switch at frequencies of 20 to 500 kHz, preferably roughly 50 kHz to 150 kHz. The switches S1 and S2 are preferably executed using low-voltage MOSFETS, which give rise to very low losses, even when there is a high current (RDS(on)≈1 mΩ). In terms of the principle, however, any type of switch can be used—MOSFETs, transistors (e.g. IGBTs), thyristors, etc.
As can be seen from
Two operating modes are possible with the circuit according to the invention:
Both switches S1, S2 are switched through. This mode is characterised by the expression “overlapping in time”, which has already been used above. Only ohmic losses occur in this case.
S1, S2 synchronise alternately at a given frequency, e.g. a frequency between 50 kHz . . . 150 kHz. In this case, the pulse width modulation or frequency ratio determines the current division I1/I2 or the ratio of the individual currents I1 and I2 to the total current IA.
It should be noted in this case that the switches must be able to interrupt the current flow in both directions and that the voltage across the switches must not be allowed to exceed the permitted inverse voltage (e.g. rated at 10-20% of the battery's rated voltage). In case of emergency, the switches must be connected through or the battery disconnected by means of the contactors.
In the synopsis with
The voltage UA is the total of the two battery voltages U1 and U2 weighted with the relative voltage application duration in each case (pulse width or pulse-duty factor); where
The current IA is divided proportionately between the two batteries in relation to the voltage application times in each case:
Consequently, the current distribution between the batteries may be selected at random by choosing the voltage application times T1 and T2.
The circuit's optimum frequency ratio is preferably regulated by the fact that the controller receives feedback on the current state of charge of the individual batteries. This may result, for example, from the voltage and/or current from the individual batteries being measured continuously or at given intervals in time and the corresponding measurement passed on to the controller.
The advantages of this circuit are that a smaller throttle can be used and that a continuous, ripple-free current flow can be achieved in each battery at a 50% duty cycle, in other words, in the case of T1=T2=T/2. The connection to the battery's positive terminal is only used in this case to supply the drivers, for example (power requirement low watts) and has no function in the main circuit.
I
12
=T
1
·I
1+(T−T1)·I2
I
23
=T
2
·I
2+(T−T2)·I3
I
31
=T
3
·I
3+(T−T3)·I1
When both switches are in a stationary, switched-through state, this is a special case of operation overlapping in time. The general (synchronised) overlapping operation is not of course preferred in all configurations. Particularly in the case of systems with more than one throttle (e.g.
The embodiment shown in
The definition of “charge accumulator-end connection terminals” on the one hand and “load- or charging-end connection terminals” on the other hand is always arrived at from the point of view of the electrical circuit 10 in each case. In other words, those connection terminals that are closer to the charge accumulators or are assigned to these are referred to as being at the charge accumulator end. Those connection terminals that are closer to the load and/or the charging device or HV BUS are referred to as being at the load or charging end. Hence, it may quite easily be, as in
As already mentioned earlier, the circuit according to the invention may also be an integral part of a charging device. The charge voltage is applied to the circuit's connection terminals at the charge end, e.g. via an AC/DC converter, and the charge current divided proportionately via the DC converter according to the pulse width modulation applied or the frequency ratio of the switches operated alternately. The current state of charge of the individual batteries is taken into account when controlling the switches, so that the frequency ratio can change in favour of a battery during the course of the charging process.
Although the “modular approach”, as depicted in
The three batteries illustrated in
In order to extend the corresponding wording to all embodiments shown, it should now be considered or established that in the circuit according to the invention it is essentially the greatest differential voltage applied between matching connection terminals at the charge accumulator end or a voltage derived from this differential voltage that is converted at best.
Number | Date | Country | Kind |
---|---|---|---|
00361/10 | Mar 2010 | CH | national |
This application is a 35 U.S.C. 371 national-phase entry of PCT International application no. PCT/IB2011/050848 filed on Feb. 28, 2011 and also claims benefit of priority to Swiss national application no. CH-00361/2010 filed on Mar. 15, 2010, and also claims priority as a non-provisional of U.S. provisional application Ser. No. 61/313,866 filed on Mar. 15, 2010; parent application PCT/IB2011/050848, Swiss national application no. CH-00361/2010 and U.S. provisional application Ser. No. 61/313,866 are all incorporated herein by reference in their entireties for all intents and purposes, as if identically set forth in full herein.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/IB11/50848 | 2/28/2011 | WO | 00 | 9/13/2012 |
Number | Date | Country | |
---|---|---|---|
61313866 | Mar 2010 | US |