Texas Instruments | AM572x Thermal Considerations (Rev. A) | Application notes | Texas Instruments AM572x Thermal Considerations (Rev. A) Application notes

Texas Instruments AM572x Thermal Considerations (Rev. A) Application notes
Application Report
SPRAC53A – December 2016 – Revised December 2018
AM572x Thermal Considerations
Michael Erdahl
ABSTRACT
This application report discusses thermal performance of the Sitara™ AM572x series processors. Data
presented demonstrates the effects of different thermal management strategies in terms of processor
junction temperature and power consumption across CPU loading and ambient temperature.
1
2
3
4
5
Contents
Overview ......................................................................................................................
References ...................................................................................................................
Important Notes ..............................................................................................................
Test Overview ................................................................................................................
Data and Results ............................................................................................................
2
2
2
2
4
List of Figures
........................................................................ 5
.......................................................................... 6
Junction Temperature vs Ambient Temperature ........................................................................ 7
Power Consumption vs Ambient Temperature .......................................................................... 8
Junction Temperature vs Ambient Temperature ........................................................................ 9
Power Consumption vs Ambient Temperature ......................................................................... 10
Junction Temperature vs Ambient Temperature ....................................................................... 11
Power Consumption vs Ambient Temperature ......................................................................... 12
Junction Temperature vs Ambient Temperature ....................................................................... 13
Power Consumption vs Ambient Temperature ......................................................................... 14
1
Junction Temperature vs Ambient Temperature
2
Power Consumption vs Ambient Temperature
3
4
5
6
7
8
9
10
List of Tables
1
Supported OPP vs Max Frequency ....................................................................................... 3
2
Processor OPP Settings .................................................................................................... 5
3
Processor OPP Settings .................................................................................................... 7
4
Processor OPP Settings .................................................................................................... 9
5
Processor OPP Settings
6
Processor OPP Settings................................................................................................... 13
..................................................................................................
11
Trademarks
Sitara is a trademark of Texas Instruments.
All other trademarks are the property of their respective owners.
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
AM572x Thermal Considerations
1
Overview
1
www.ti.com
Overview
The heterogeneous multicore AM5728 general-purpose (GP) evaluation module (EVM) was used in this
experiment to gather thermal data with different processor loading, operating performance points (OPP)
and ambient temperature. Ambient temperature was controlled with programmable environmental
chamber.
The collected data can be utilized to correlate the thermal performance of the processor and power
consumption at a given processor load and junction temperature, based on ambient temperature and
thermal management.
Tests were repeated with the following thermal management:
• Bare package (no heatsink)
• Low-cost heatsink (provided with GP and Beagleboard-X15 community boards)
• Low-cost heatsink with fan (heatsink was not changed)
2
References
•
•
•
3
To learn more about thermal management, visit http://www.ti.com/thermal
Thermal Design Guide for DSP and ARM Application Processors
Thermal models can be found in the Models section of Tools and Software in the product folder:
http://www.ti.com/product/AM5728/toolssoftware
Important Notes
The environmental chamber used to collect this data circulates air internally to maintain homogeneous
internal temperature, and does not accurately simulate the environment on the bench or end product. This
is important to consider in passive cooling applications where air circulation can significantly impact PCB,
package, and heatsink power dissipation efficiency.
Data presented in this test was gathered with a typical device, representing nominal silicon process and
leakage. Thermal performance and power consumption can vary significantly due to process variation.
Extra margin must be designed in to account for worst case process variation (leakage).
4
Test Overview
The following CPU loading schemes were characterized with the AM572x GP EVM for this report.
4.1
OS Idle
AM57x processor is idling at the Linux Matrix graphical user interface (GUI). Here, the processor cores are
active, consuming minimal power.
4.2
Dhrystone Single Core
Dhrystone is a single-threaded benchmark, capable of utilizing approximately 100% of one ARM CortexA15 core. Dhyrstone is included in the TI Processor SDK. Tests were conducted with the A15 running at
1.0GHz (OPP_NOM) and 1.5GHz (OPP_HIGH).
4.3
Dhyrstone Dual Core
Dhrystone benchmark instances running on each ARM Cortex-A15 core. Tests were conducted with the
A15 running at 1.0GHz (OPP_NOM) and 1.5GHz (OPP_HIGH).
4.4
Temperature Measurement
Reported temperature data is measured by on-die sensors to approximate actual junction temperature.
Temperature for each use-case is measured after soaking for 5 minutes. Under lab conditions, it was
determined a 5 minute period allows the processor to reach stable temperature.
2
AM572x Thermal Considerations
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
Test Overview
www.ti.com
The TI Processor SDK provides Linux drivers for these sensors, and can be queried from the commandline, for example:
# cat /sys/class/thermal/thermal_zone0/temp
71800
4.5
OPP Definitions
Operating performance points (OPP) levels imply fixed voltage and frequency targets on a per-subsystem
basis. The table below lists frequency of each subsystem per OPP for the AM57x processor.
Dynamic Voltage Frequency Scaling (DVFS) refers to a software technique where the various SoC AVS
rails are changed from one OPP level to another in order to either adapt to a changing work-load, or in
order to avoid device operation outside of desired temperature bounds.
The SoC only supports DVFS on the MPU rail. Other OPP levels should be set by the initial bootloader.
Fixed OPP levels can affect the thermal and power estimates of your system. Ensure the selected OPP
level meets the application's needs, and all thermal testing is conducted at the desired OPP level.
Table 1. Supported OPP vs Max Frequency
OPP_NOM
OPP_OD
OPP_HIGH
Voltage Domain
Clock Domain
Max Frequency (MHz)
Max Frequency (MHz)
Max Frequency (MHz)
VD_MPU
MPU_CLK
1000
1176
1500
VD_DSPEVE
DSP_CLK
600
700
750
VD_IVA
IVA_GCLK
388.3
430
532
VD_GPU
GPU_CLK
425.6
500
532
VD_CORE
CORE_IPUx_CLK
212.8
N/A
N/A
L3_CLK
266
DDR3/ DDR3L
532 (DDR3-1066)
RTC_FCLK
0.034
N/A
N/A
VD_RTC
4.6
Default OPP Levels
Tests were conducted with Processor SDK 03.00.00.04, with following default OPP levels:
OPP
MPU
GPU
DSP
IVA
NOM
HIGH
HIGH
HIGH
Typically, the MPU domain OPP level can scale dynamically between OPP_NOM, OPP_OD and
OPP_HIGH, in Linux using the CPUFreq driver, based on CPU load and selected governor. For the tests
in this document, the “userspace” governor was used, and MPU frequency was manually set at the Linux
prompt.
GPU, DSP and IVA cores are have fixed OPP levels, and clock-gated when not in use. Today's OPP
levels for GPU, DSP and IVA cores can be changed by recompiling U-Boot with the desired OPP level.
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
AM572x Thermal Considerations
3
Data and Results
5
www.ti.com
Data and Results
This section contains the raw data and graphs of the test experiments described above. All data was
gathered running the latest Linux Processor SDK.
All tests were conducted with an LCD module installed, which adds extra power.
5.1
OPP Settings and Linux Thermal Framework
Operating performance points for each major compute subsystem is given below. For clock speeds, see
Section 4.5. GPU, DSP and IVA cores are clock-gated due to inactivity.
MPU OPP level was held constant for each test. Additionally, the Linux thermal framework was disabled to
hold the OPP constant as the processor heats, and prevents thermal shutdown. This was done for data
gathering purposes, and is not recommended for a production system.
5.2
Power and Thermal Chamber Measurements (Nom Unit) - SoC Power vs Tambient
The tables below contains power consumption and junction temperature measured running OS Idle,
Dhrystone single core and Dhrystone dual core use cases at different controlled ambient temperatures
and varying thermal management schemes. Silicon process type is nominal. Dhrystone tests were
repeated with MPU at OPP_NOM (1.0 GHz) and OPP_HIGH (1.5 GHz). OPP_OD adds only a modest
power increase over OPP_NOM.
Junction temperature and power reported in the following sections were sampled at the same time, and
are presented in separate tables to aid comprehension.
4
AM572x Thermal Considerations
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
Data and Results
www.ti.com
5.3
OS Idle (MPU @ OPP_NOM)
Table 2. Processor OPP Settings
OPP
MPU
GPU
DSP
IVA
NOM
HIGH
HIGH
HIGH
Ta (°C)
Therm Mgmt
25
40
50
60
70
No Heatsink
(°C)
46
61
71
82
94
80
Heatsink (°C)
43
58
68
78
88
100
Heatsink + Fan
(°C)
38
52
61
70
81
89
90
100
Figure 1. Junction Temperature vs Ambient Temperature
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
AM572x Thermal Considerations
5
Data and Results
www.ti.com
Ta (°C)
Therm Mgmt
25
40
50
60
70
No Heatsink
(mW)
1117
1360
1509
1760
2103
80
Heatsink (mW)
1118
1281
1447
1649
1923
2304
Heatsink + Fan
(mW)
1064
1203
1332
1542
1712
1973
90
2323
Figure 2. Power Consumption vs Ambient Temperature
6
AM572x Thermal Considerations
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
Data and Results
www.ti.com
5.4
Dhrystone Single Core (MPU @ OPP_NOM)
Table 3. Processor OPP Settings
OPP
MPU
GPU
DSP
IVA
NOM
HIGH
HIGH
HIGH
Ta (°C)
Therm Mgmt
25
40
50
60
70
No Heatsink
(°C)
53
68
78
88
100
Heatsink (°C)
48
63
72
84
94
Heatsink + Fan
(°C)
41
54
64
73
82
80
90
94
104
Figure 3. Junction Temperature vs Ambient Temperature
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
AM572x Thermal Considerations
7
Data and Results
www.ti.com
Ta (°C)
Therm Mgmt
25
40
50
60
70
No Heatsink
(mW)
2092
2291
2559
2800
3223
Heatsink (mW)
2031
2240
2430
2671
3012
Heatsink + Fan
(mW)
1961
2158
2268
2452
2678
80
90
2992
3380
Figure 4. Power Consumption vs Ambient Temperature
8
AM572x Thermal Considerations
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
Data and Results
www.ti.com
5.5
Dhrystone Single Core (MPU @ OPP_HIGH)
Table 4. Processor OPP Settings
OPP
MPU
GPU
DSP
IVA
HIGH
HIGH
HIGH
HIGH
Ta (°C)
Therm Mgmt
25
40
50
No Heatsink
(°C)
66
84
99
60
Heatsink (°C)
58
74
86
99
Heatsink + Fan
(°C)
46
62
71
82
70
80
92
104
90
Figure 5. Junction Temperature vs Ambient Temperature
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
AM572x Thermal Considerations
9
Data and Results
www.ti.com
Ta (°C)
Therm Mgmt
25
40
50
No Heatsink
(mW)
4076
4758
5523
60
Heatsink (mW)
3953
4429
4879
5572
Heatsink + Fan
(mW)
3735
4070
4350
4744
70
80
5190
5835
90
Figure 6. Power Consumption vs Ambient Temperature
10
AM572x Thermal Considerations
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
Data and Results
www.ti.com
5.6
Dhrystone Dual Core (MPU @ OPP_NOM)
Table 5. Processor OPP Settings
OPP
MPU
GPU
DSP
IVA
NOM
HIGH
HIGH
HIGH
Ta (°C)
Therm Mgmt
25
40
50
60
No Heatsink
(°C)
58
74
84
94
70
Heatsink (°C)
52
67
78
88
100
Heatsink + Fan
(°C)
43
58
65
76
87
80
90
97
Figure 7. Junction Temperature vs Ambient Temperature
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
AM572x Thermal Considerations
11
Data and Results
www.ti.com
Ta (°C)
Therm Mgmt
25
40
50
60
No Heatsink
(mW)
3048
3317
3574
3927
70
Heatsink (mW)
3001
3210
3393
3654
4059
Heatsink + Fan
(mW)
2874
3056
3176
3382
3681
80
90
4001
Figure 8. Power Consumption vs Ambient Temperature
12
AM572x Thermal Considerations
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
Data and Results
www.ti.com
5.7
Dhrystone Dual Core (MPU @ OPP_HIGH)
Table 6. Processor OPP Settings
OPP
MPU
GPU
DSP
IVA
HIGH
HIGH
HIGH
HIGH
Ta (°C)
Therm Mgmt
25
40
50
No Heatsink
(°C)
84
104
Heatsink (°C)
71
89
103
Heatsink + Fan
(°C)
54
69
80
60
70
90
102
80
90
Figure 9. Junction Temperature vs Ambient Temperature
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
AM572x Thermal Considerations
13
Data and Results
www.ti.com
Ta (°C)
Therm Mgmt
25
40
No Heatsink
(mW)
6868
7785
50
Heatsink (mW)
6448
7166
7651
Heatsink + Fan
(mW)
6068
6466
6846
60
70
7220
7859
80
90
Figure 10. Power Consumption vs Ambient Temperature
14
AM572x Thermal Considerations
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
Revision History
www.ti.com
Revision History
NOTE: Page numbers for previous revisions may differ from page numbers in the current version.
Changes from Original (December 2016) to A Revision ................................................................................................ Page
•
The Power vs Ambient temperature graphs has been updated with new one for Dhrystone(1 core, OPP_NOM),
Dhrystone(1 Core, OPP_HIGH), Dhrystone(2 Core, OPP_NOM) and Dhrystone(2 Core, OPP_HIGH)....................... 2
SPRAC53A – December 2016 – Revised December 2018
Submit Documentation Feedback
Copyright © 2016–2018, Texas Instruments Incorporated
Revision History
15
IMPORTANT NOTICE AND DISCLAIMER
TI PROVIDES TECHNICAL AND RELIABILITY DATA (INCLUDING DATASHEETS), DESIGN RESOURCES (INCLUDING REFERENCE
DESIGNS), APPLICATION OR OTHER DESIGN ADVICE, WEB TOOLS, SAFETY INFORMATION, AND OTHER RESOURCES “AS IS”
AND WITH ALL FAULTS, AND DISCLAIMS ALL WARRANTIES, EXPRESS AND IMPLIED, INCLUDING WITHOUT LIMITATION ANY
IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT OF THIRD
PARTY INTELLECTUAL PROPERTY RIGHTS.
These resources are intended for skilled developers designing with TI products. You are solely responsible for (1) selecting the appropriate
TI products for your application, (2) designing, validating and testing your application, and (3) ensuring your application meets applicable
standards, and any other safety, security, or other requirements. These resources are subject to change without notice. TI grants you
permission to use these resources only for development of an application that uses the TI products described in the resource. Other
reproduction and display of these resources is prohibited. No license is granted to any other TI intellectual property right or to any third
party intellectual property right. TI disclaims responsibility for, and you will fully indemnify TI and its representatives against, any claims,
damages, costs, losses, and liabilities arising out of your use of these resources.
TI’s products are provided subject to TI’s Terms of Sale (www.ti.com/legal/termsofsale.html) or other applicable terms available either on
ti.com or provided in conjunction with such TI products. TI’s provision of these resources does not expand or otherwise alter TI’s applicable
warranties or warranty disclaimers for TI products.
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright © 2018, Texas Instruments Incorporated
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals

Download PDF

advertising