Adjusting the thermal behavior of a computing system using indirect information about ambient temperature

• US 8,452,463 B2
• Filed: 06/04/2010
• Issued: 05/28/2013
• Est. Priority Date: 06/04/2010
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: Cox, Keith, Kapoor, Gaurav, Liang, Frank Faqiu
• Primary Examiner(s): Bahta, Kidest
• Application Number:
  US12/794,464
• Publication Number:
  US 20110301777A1
• Time in Patent Office: 1,089 Days
• Field of Search: 700/299, 700/300, 455/574, 455/343.1, 706/12, 706/62, 713/600
• US Class Current: 700/299
• CPC Class Codes: G06F 1/206 comprising thermal management
• CPC Class Codes: G06F 1/3203 Power management, i.e. event-based initiation of a power-saving mode
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A computing system has a thermal manager that changes a power consuming activity limit in the device based on an estimate of temperature of a target location in the system. There are several temperature sensors that are not at the target location. An estimator computes the target location temperature estimate using a thermal model and, as input to the thermal model, data from the sensors. The thermal model produces different estimates of the target location temperature at different ambient temperatures but without computing or measuring the ambient temperatures. Other embodiments are also described and claimed.

Claims

1. A method for adjusting the thermal behavior of a computing system, comprising:

accessing a thermal model algorithm that is stored in memory of a computing system, wherein the thermal model algorithm estimates temperature of a target location in the system based on temperature data from a plurality of temperature sensors that are not at the target location, wherein for the same power consumption state of the computing system, the thermal model algorithm computes a different estimate at a higher ambient temperature than at a lower ambient temperature without estimating or measuring ambient temperature;

receiving temperature data from the temperature sensors;

estimating the temperature of the target location using the thermal model algorithm; and

making a decision on whether or not to change a power consuming activity limit in the computing system, based on the estimated temperature.

2. The method of claim 1 further comprising receiving power consumption data for a component in the system, wherein the decision on the power consuming activity limit is made further based on the power consumption data.

3. The method of claim 1 wherein the temperature estimate at 25 C ambient is higher than at 35 C ambient, while power consumption state of the computing system remains constant.

4. The method of claim 1 wherein by estimating a lower target location temperature at the higher ambient temperature, the decision made on that basis allows the system to run hotter at the higher ambient temperature.

5. The method of claim 1 wherein the target location is an outside surface of the system that is touched by a user of the system.

6. An article of manufacture comprising:

a machine-readable storage medium having stored therein instructions that program a processor in a mobile device to perform thermal control in the device at different ambient temperatures, the programmed processor is to receive temperature data output by a plurality of temperature sensors in the system, compute temperature of a target location in the device that is spaced apart from the plurality of sensors, using a thermal model and the temperature data as input to the thermal model, and make a decision on whether or not to change a power consuming activity limit in the device based on the computed temperature, wherein the power consuming activity limit is changed differently at different ambient temperatures based on the computed temperature but without estimating or measuring ambient temperature.

7. The article of manufacture of claim 6 wherein the power consuming activity limit is selected from the group consisting of:

the speed of a cooling fan including whether the fan is on or off;

a power supply voltage of a data processing unit;

maximum transmit power of an RF antenna;

clock speed of a data processing unit;

whether a data processing unit is enabled or disabled; and

intensity of a light source including whether the light source is on or off.

8. The article of manufacture of claim 6 wherein the thermal model is such that when it is used to predict temperature of a target location while the mobile device is at a lower ambient, and again while the mobile device is at a higher ambient, in both cases while power consumption state of the mobile device remains constant, the predicted temperature is about equal to the actual temperature of the target location at the lower ambient, but is less than the actual temperature at the higher ambient.

9. The article of manufacture of claim 6 wherein the decision on whether or not to change a power consuming activity limit in the device is further based on received power consumption data for a component in the device.

10. The article of manufacture of claim 6 wherein by computing a lower target location temperature estimate at the higher ambient temperature, the decision on the power consuming activity limit causes the device to run hotter at the higher ambient temperature.

11. The article of manufacture of claim 6 wherein the target location is an outside surface of the system that is touched by a user of the system.

12. A mobile device comprising:

a thermal manager to change a power consuming activity limit in the mobile device based on an estimate of temperature of a target location in the device;

a plurality of temperature sensors not at the target location; and

an estimator to compute the estimate using a thermal model and, as input to the thermal model, temperature data from the sensors, wherein the thermal model produces different estimates of the target location temperature at different ambient temperatures but without computing or measuring the ambient temperatures.

13. The mobile device of claim 12 wherein the target location includes a critical point on an outside surface of the device that a user of the system touches while using the system.

14. The mobile device of claim 13 wherein the target location is a point on a back face or side of an external housing of the device that the user touches while using the device as a telephony handset.

15. A method for computing a thermal model, wherein the thermal model estimates temperature of a target location in a computing system during in-the-field use of the system based on temperature data from a plurality of temperature sensors that are not at the target location, the method comprising:

recording temperature data from (a) a plurality of temperature sensors that are embedded in the computing system and (b) a testing temperature sensor that is temporarily placed at the target location in the computing system, wherein the data is collected while the system is operated at different ambient temperatures so that transient and steady state thermal behavior of the system at the different ambient temperatures is recorded;

applying an offset to the recorded temperature data, or to a variable representing the recorded temperature data, that is obtained from the testing sensor at one ambient, relative to the recorded temperature data that is obtained from the testing sensor at another ambient; and

fitting a curve or curved surface, that will be part of the thermal model, to the offset temperature data from the testing sensor, in relation to the recorded temperature data from the embedded sensors.

16. The method of claim 15 further comprising:

determining the offset by

a) beginning with a value of zero for the offset, computing an instance of the thermal model using the zero as the offset, wherein the thermal model includes a plurality of coefficients that have been computed so as to fit a curve to the recorded temperature data from the testing sensor and the embedded sensors, and computing a temperature for the target location using the instance of the thermal model in which the offset is zero,

b) increasing the offset and then using it to compute a new instance of the thermal model, computing the change in the coefficients of the new instance of the thermal model relative to a previous instance of the thermal model, and the change in the inaccuracy of the new instance of the thermal model in predicting the temperature for the target location relative to the previous instance, and

c repeating b) until a combination of the coefficient change and the inaccuracy change has become greater than a set threshold.