Embedded system designers historically have limited memory offerings for applications that need to operate under harsh high shock and vibration conditions. This is because advances in memory technology and its associated standard dual inline memory module (DIMM) and small outline DIMM (SODIMM) form factors are largely driven by demand in the PC, telecom and server markets. Memory modules designed for these market applications often do not meet key embedded application specifications that must allow for space-constrained layouts, while also delivering high reliability and performance, and long-term operation in harsh or harsh environments. In the embedded market, memory products must support long product life cycles and be cost-effective.

Some memory module suppliers focus on the needs of the embedded market and continue to develop memory technology advancements. Memory vendors have banded together through various standards groups to make these advances in commodity memory modules, enabling embedded system designers to use rugged devices of various capacities. This standardization also brings the added benefit of consistent availability from multiple suppliers, which helps OEMs accelerate time-to-market while reducing overall system cost and project risk.

Great strides in rugged memory technology

Innovations in memory technology offer embedded system OEMs a variety of rugged options, including small form factor module designs, error correcting code (ECC), thermal dissipation, extended temperature operation, and the addition of thermal sensors to monitor module temperature.

Embedded system OEMs view double data rate type three (DDR3) SODIMM memory modules as the backbone of rugged embedded system designs. Adding to the endurance of DDR3 SODIMMs are new low-power, low-power DDR3L memory modules that address a key embedded system design challenge. JEDEC mandates that systems running memory above +85 °C must double the DDR3 self-refresh rate. DDR3L memory modules address the dual refresh rate requirement by choosing the lowest total current, adopting the thermal release copper method PCB design, reducing die count, and utilizing 1.35 V DDR3 dynamic random access memory (DRAM). Compared to current DDR3 designs, DDR3L memory saves up to +10 °C per module and eliminates the dual refresh rate requirement. Vendor-based testing has shown that, depending on the components used,

Table 1: Virtium internal test data shows that, depending on the components used, OEMs can achieve up to 50% power reduction with 8 GB ECC memory modules.

The Blade VLP is a low profile (17.78 mm) replacement for the JEDEC standard VLP with a height of 18.75 mm. Reduces the height of DDR3 VLP memory modules to 17.78 mm, addressing the space-constrained constraints present in many telecom and networking applications where it is difficult to accommodate the memory required in industry standard DIMM or Mini DIMM sockets plus a standard the VLP. This approach enables designers to reduce the total power consumption of systems using multiple memory modules and systems that must operate above +85 °C, a typical design challenge for various AdvancedTCA-based telecom and Ethernet blade switch networking applications .

Designers of telecommunications and networking blade systems are often faced with tight constraints on system height. Additionally, these systems require space on top of the memory modules to allow airflow for effective thermal management. The use of reduced-height DDR3L VLP memory modules helps improve airflow and provides a low profile, enabling OEMs to deliver higher reliability products, reducing total cost of ownership. Certain DDR3L VLP modules also offer single-shot refresh rates, which are critical for maximizing performance in high-temperature systems.

Multiple ways to help reinforce

To help OEMs meet extreme requirements for vibration, temperature or other harsh environmental conditions, memory suppliers offer manufacturing advancements such as side retention clips to reinforce DDR3 SODIMM modules. These universally applicable clips can be easily implemented in a variety of applications. In the recent past, designers were often limited to using weaker commercial-grade retention clips to secure memory modules. In some cases, these retainers can suddenly open and cause system-level failures. Other alternatives involving mounting holes require significant modifications to the motherboard, often resulting in non-standard COTS-based designs that do not adequately address the issue.

Additionally, OEMs can take advantage of the underfill option to provide higher impact resistance for components populated on standard FR-4 PCBs. Conformal coating is another MIL-I-46058C compliant option that provides enhanced protection against environmental degradation.

In addition to mechanical improvements, OEMs can make many electrical upgrades, including extended temperature screening and burn-in and the addition of thermal sensors to monitor module temperature. Designers can typically choose from three temperature options for memory modules:

Industrial temperature: -40 °C to +95 °C

Extended temperature: -25 °C to +95 °C

Standard temperature: 0 °C to +95 °C

Testing is critical to ensuring modules meet temperature specifications. Therefore, it is important to define a standard set of temperature test parameters and that memory suppliers work with OEMs to tailor test methods to specific equipment and test time requirements. Embedded systems often perform mission-critical operations, so after a test definition has been established and validation is complete, it is recommended that memory modules be 100% tested according to the defined schedule.

The best test method to ensure extended temperature operation is through production testing using a customer motherboard or on an approved motherboard with the same chipset and settings. These tests can also be performed using a specially developed oven that matches most embedded motherboard form factors, enabling temperature testing across system performance.

System testing is essential to catch defects such as ECC errors that cannot be found using standard test systems. Note that system level testing may also be necessary depending on the application or system specification +85 °C ambient temperature.

An analysis of DRAM failure modes in memory modules has determined that DRAM components with suboptimal reliability tend to fail within the first three months of use. As newer DRAMs move toward smaller process geometries, chips that contain weak bits (microscopic defects in individual cells) may be at greater risk. This is not enough to cause DRAM failure directly, but single-bit errors can occur within weeks of initial field operations.

Using During Burn-in Test (TDBI) helps eliminate any potential early failures and improves the overall reliability of the memory product. While most DRAM chips undergo static burn-in at the chip level, TDBI provides a more comprehensive test methodology, enabling 24-hour burn-in testing at the module level, while dynamically running and checking test mode conditions while the module is performing under stress. Studies conducted by several memory manufacturers have shown that early failures can be reduced by up to 90% using a TDBI chamber.

New Standards and Form Factors

Several industry groups such as JEDEC and the Small Form Factor Special Interest Group (SFF-SIG) are actively involved in standardizing storage devices for today’s embedded systems. Standardization brings the added benefit of consistent availability from multiple suppliers, which helps OEMs accelerate time-to-market while reducing overall system cost and project risk.

ECC has become the mainstay of embedded systems. However, JEDEC members were initially unaware of the need to accommodate ECC when developing the DDR2 specification for the SODIMM form factor, since most laptop chipsets at the time did not support ECC. Seeing the need to implement ECC on faster DDR2 memory modules in embedded systems, Virtium sponsored the ECC SODIMM specification in JEDEC, which has now been extended to DDR3 and DDR4 modules.

The SFF-SIG’s XR-DIMM specification is another example of a memory device defined for embedded systems to operate reliably under excessive shock and vibration conditions. Designers of these systems need a small, extremely rugged DDR3 module. The standard frees designers from the constraints of previous commercial-grade products that required soldering, strapping, glue or tie-downs to secure modules.

A SFF-SIG collaboration between Virtium, Swissbit and LiPPERT embedded computers resulted in a module with pin definitions very similar to DDR3 standard DIMMs. The pinout utilizes a high-performance 240-pin SMT connector system that uses standoffs with screw attachments to securely hold XR-DIMM memory modules to the motherboard. In addition, this pinout includes a SATA interface to support the development of dual-function modules containing DDR3 and NAND flash memory for combined memory and solid-state drive storage from a single socket. Future standards for SATA and DDR3 combo modules are planned.

Meet demanding memory requirements

Despite the increasing demand for ruggedized embedded devices, memory module suppliers continue to make technological advancements and associated manufacturing improvements to meet OEMs’ needs. DDR3 SODIMM, DDR3L, and small form factor DDR3L are all examples of new technologies that help meet demanding memory requirements. These advancements address many design challenges, including low power consumption, enhanced thermals, and extended temperature tolerance, while delivering the performance required by today’s complex embedded systems.

Standards for XR-DIMM and ECC SODIMM also facilitate the off-the-shelf supply of durable memory products. Additionally, designers have access to underfill side clips and conformal coating fabrication options as well as advanced testing methods to help ensure robust designs.

The challenges of maintaining the highest reliability and availability in rugged embedded system designs will continue, but advances in memory modules will keep pace with these requirements, helping OEMs remain competitive and innovate in the future.

Reviewing Editor: Guo Ting

Leave a Reply

Your email address will not be published.