6 is configured to dissipate an additional amount of power proportional to 10 mA of current for 5*RT (e.g., 6*RT−RT). In an example, each of the Devices 1-6 includes a configurable register that specifies an amount of power or current to be dissipated.

Similarly, because Device 5 consumes 10 mA of current for 2*RT driving its communication signal, Device 5 is configured to dissipate an additional amount of power proportional to 10 mA of current for 4*RT (e.g., 6*RT−2*RT). Continuing this approach, Device 4 is configured to dissipate an additional amount of power proportional to 10 mA of current for 3*RT; Device 3 is configured to dissipate an additional amount of power proportional to 10 mA of current for 2*RT; and Device 2 is configured to dissipate an additional amount of power proportional to 10 mA of current for RT. Because Device 1 is the device responsible for the longest response time, Device 1 does not dissipate any additional power/current. Accordingly, the devices (e.g., battery monitors 134) each consume an approximately equal amount of power during stack response communications, which improves the balance of power consumed from the battery modules 108 in the battery pack 106.

FIG. 5 is a timing diagram 500 of response times of devices in a stack, such as the battery monitors 134 described above, in accordance with an example of this description. The responses in the timing diagram 500 are responsive to a single device read, in which a specified one of the battery monitors 134 provides a value of one or more of its registers in the response. The Devices 1-6 in the example of FIG. 5 correspond to the Devices 1-6 in the example of FIG. 3.

In the example of FIG. 5, the battery management controller 130 provides the single device read command to the stack of battery monitors 134 through the communication bridge device 132. The single device read command specifies, or is directed to, Device 3. Because of the daisy-chain communications between the battery monitors 134, the single device read command is passed from Device 1 to Device 6, irrespective of the device specified by the stack read command. However, because the command in this example is a single device read command directed to Device 3, Devices 4-6 do not respond to the command, and thus Device 3 begins to respond after providing the command to Device 4. Accordingly, before time 502, Device 3 has provided the command to Device 4, and Devices 4-6 can ignore the single device read command.

Accordingly, at time 502, because the single device read command is directed to Device 3 and Device 3 has already passed the command up the stack to Device 4, Device 3 begins to provide its response to the next-lowest downstream device, Device 2. As above, the duration of the response for Device 3 is shown as “RT” for response time.

A time period between the time 502 and time 504 is the Device 2 re-clock time (t_RECLK). Accordingly, at time 504, Device 2 begins to forward the Device 3 response to the next-lowest downstream device, Device 1. In this example, Device 2 does not append any additional response because the single device read is only directed to Device 3. The Device 2 response time is thus RT.

Finally, a time period between the time 504 and time 506 is the Device 1 re-clock time (t_RECLK). Accordingly, at time 506, Device 1 begins to forward the Device 3 response to the next-lowest downstream device, which is the communication bridge device 132. In this example, Device 1 does not append any additional response because the single device read is only directed to Device 3. The Device 1 response time is thus RT, and the longest response time of any of the devices (e.g., battery monitors 134) is also RT, which is the response time of each of Devices 1-3. The communication bridge device 132 provides the response of the battery monitors 134 to the single device read command to the battery management controller 130.

FIG. 6 is a power consumption diagram 600 corresponding to the timing diagram 500 of FIG. 5 in accordance with an example of this description. The timing of FIG. 6 is similar to that of FIG. 5 and is not described again for brevity. As described above, the power consumed by a device (e.g., battery monitor 134) to drive a communication signal is proportional to the duration of the communication signal. In these examples, a device consumes 10 mA of current for one response time, RT. Thus, in the single device read command example of FIG. 5, Devices 1-3 each consume 10 mA for 1*RT, whereas Devices 4-6 each consume 0 mA. This is another example of a command/communication that causes the power consumption imbalance across the battery modules 108, described above.

The power consumption diagram 600 shows that Devices 4-6 are configured to dissipate an additional amount of power, which is proportional to the difference between that device's response time (e.g., 0) and the longest response time (e.g., any of the Devices 1-3 response time, which is RT). For example, because Device 6 consumes 0 mA of current driving its communication signal, and Device 1 consumes 10 mA of current for 1*RT driving its communication signal, Device 6 is configured to dissipate an additional amount of power proportional to 10 mA of current for 1*RT.

Continuing this approach, Device 4 and Device 5 are also configured to dissipate an additional amount of power proportional to 10 mA of current for 1*RT. Because Devices 1-3 are the devices responsible for the longest response time, Devices 1-3 do not dissipate any additional current. Accordingly, each of the devices (e.g., battery monitors 134) consumes an approximately equal amount of power during stack response communications, which improves the balance of power consumed from the battery modules 108 in the battery pack 106.

The above examples which generally address power balancing during response communications. Also, the top-of-stack device (e.g., battery monitor 134) does not forward a command received from a downstream device, whereas each of the other devices in the stack forwards such a command to an upstream device. In this case, the top-of-stack device consumes less power during command communications as well. Thus, the top-of-stack device is configured to dissipate an additional amount of power that is proportional to a forwarding time for the command. Accordingly, the devices (e.g., battery monitors 134), including the top-of-stack device, each consumes an approximately equal amount of power during command communications as well.

In an example, each of the stack devices, such as Devices 1-6 in the examples of FIGS. 3-6, includes a configurable register that specifies an amount of additional power or current to be dissipated by that device. For example, referring to the stack read command of